AI Console Articles

How to set up an SSL-secured support center

Support centers that run on Bold360 AI servers can be deployed as HTTPS.

If your site is secured and you have implemented the support center using iframe, then a secure connection must be set up in Bold360 AI including installing the necessary certificates.

SSL (Secure Sockets Layer) is the standard security technology for establishing an encrypted link between a web server and a browser. This link ensures that all data sent between the web server and browsers remain private and integral.

Please ask your IT specialist to contact your support representative or Genesys DX support and provide the following information:

  • Account name, for example: bold360
  • Designated Certificate URL, for example: https://support.bold360.com
  • The certificate (CRT or PEM) in a text file
    Note: Firewalls may filter other types of files.
  • The private key in a text file
  • The certificate expiration date, for example: 21/03/2022 (optional)
  • Any additional comments (optional)
Important: For security reasons, text files should be password protected. You can create a password-protected ZIP for your text files, or use a protected shared folder when sending your certificate and private key files.

Monitor your outreach campaign

***These capabilities are in reference to accounts where Bold360 Acquire is currently enabled. We are no longer enabling Acquire on accounts.

You can monitor the progress of your sales outreach on the sales dashboard, which provides you with a visual representation of your company?s sales KPIs (with a 2-hour latency). The sales dashboard allows you to have a better understanding of the overall revenue compared to campaign-driven conversions, the most successful conversion personas, and campaign trend progress indicators. As a result, you can optimize sales performance throughout a specific period, which ultimately increases top line revenue.

With the sales dashboard, you can monitor sales outreach performance to track conversion rate and conversions, determine behavioral patterns among customer segments and notify stakeholders about over- or under-performing product categories.

Since you can optimize your outreach strategy by A/B testing different treatments, the sales dashboard allows you to drill-down to the level of actions that target your customers.

Tip:
  • The reporting dashboard is event-based: you must generate traffic for each of your campaigns and your A/B tested actions to see results. For example, if you defined an outreach campaign with two treatments that are A/B tested but only one of the treatments has been displayed to a customer yet, then the dashboard shows the campaign only with that particular treatment.
  • Every widget of the reporting dashboard (except the funnel) is conversion-based: you must set up and track conversion events in your account to see results.

The top four charts show various conversion related information:

  • Conversion Events ? displays the distribution of conversion events captured in the dashboard?s time interval
  • Overall Revenue ? displays the overall income and average order size based on all conversion events
  • Outreach Funnel ? provides a breakdown of the number of customers who were exposed to an outreach activity, started to interact with an outreach activity, and converted at the end
  • Revenue by Outreach ? displays the overall income and average order size generated by Bold360 Acquire

The middle section of the reporting dashboard displays outreach level progress charts:

  • Conversion Rate by Engagement Type ? shows the distribution of the different personas (outreach activity variations) who successfully generated a conversion event during the given engagement
  • Campaign Success Rate ? displays the distribution of outreach activities presented for converted customers on a given day

The Conversion by Outreach A/B variations chart at the bottom of the page displays a more granular view of your outreach activities, grouped by A/B variations. This chart shows you the distribution of outreach activity variations presented for converted customers on a given day.

Create your first proactive outreach campaign

***These capabilities are in reference to accounts where Bold360 Acquire is currently enabled. We are no longer enabling Acquire on accounts.

On the Engagement > Outreach page, you get an overview of your Bold360 outreach strategy: you see which outreach are running and the number of variations they each have.

To set up your first outreach campaign, do the following:

  1. On the Engagement > Outreach page, click Create outreach at the top of the page.
    Note: You can only edit a sales outreach that has not started yet.
    Tip: You can define your outreach strategy on account level and not on knowledge base level. This is why you see the same outreach campaigns regardless of which knowledge base you use. If you want to separate campaigns for each knowledge base, website, or customer then use naming prefixes and always define a URL as condition to make sure campaigns do not overlap.
  2. In the Outreach title field, name your sales outreach.
  3. Set the duration of your outreach.

    You can define specific dates or decide to manually start and end the sales outreach.

  4. Define criteria for selecting your target audience.

    You can define the criteria with AND/OR relations using the parameters that are listed below.

    Note: If you enable more than one criteria for a particular rule, then - depending on your AND/OR relations - all or some criteria must match for the rule to trigger an invitation. If you do not define any criteria then your sales outreach will target all your customers.
    Criteria name Goal Notes, Tips
    Device type Invite when the customer is using either a desktop or a mobile device.  
    Current URL Invite based on the URL of the page being visited. For example, if you specify /products/shoes.html, then Bold360 only sends an invite if the URL of the page that the customer visits contains that exact phrase. For example, a customer visiting the site www.mysite.com/products/shoes.html would trigger an invite, but a visit to the www.mysite.com/products/help/shoes.html site would not trigger anything.
    Note: It is recommended to define the Current URL criterion in each outreach campaigns to avoid confusing outreach behavior. Always try to define the most granular URL condition: if you define a campaign for a subpage then use the subpage?s URL as a condition, not just the main domain URL.
    Entry URL Invite based on the URL of the first page the customer sees on your site. Since the entry URL for a particular visit does not change, each page view checks this criteria. To avoid repeated chat invitations during a visit, use this together with the Pages viewed in visit criteria.
    Pages viewed in visit Invite only after customers have viewed a certain number of pages on your site.  
    Referrer URL Invite based on the URL of the page from which the customer entered your site. Since the referral URL for a particular visit does not change, each page view checks this criteria. To avoid repeated chat invitations during a visit, use this together with the Pages viewed in session criteria.
    Country Invite when the customer is in a specific country. You can select multiple countries.
    Email Invite customers whose email addresses contain specific characters. If you previously collected the email address of a customer (for example, in a pre-chat survey) and then that customer came back to the set and got identified again and the system knows his/her email address
    Extra info Define external signals that you can use with your outreach campaign.
    Note: This criteria may not be visible to you. Contact your Customer Success Manager to enable it.
    For more information on how to use the extra info condition, see What custom information can you define for your outreach activity?
    Customer info Invite according to information gathered about customers, which contains specific characters. Provided that you collect customer info either via the VMS script from a website element or any other way, you can specify the previously collected info as a trigger for returning visitors.
    Name Invite customers whose names contain specific characters. Provided that you collect customer name either via the VMS script from a website element or any other way, you can specify the previously collected info as a trigger for returning visitors.
    Phone Invite customers whose phone numbers contain specific characters. Provided that you collect customer phone number either via the VMS script from a website element or any other way, you can specify the previously collected info as a trigger for returning visitors.
    Customer ref. Invite customers whose names contain specific characters. Provided that you collect customer ref either via the VMS script from a website element or any other way, you can specify the previously collected info as a trigger for returning visitors.
    Customer status Invite new or returning customers. Choose new or returning from the drop-down list.
  5. Define what to show your customers when approaching them. You can name your actions to easily identify them when you run reports on your strategy.

    By default, the control group is 100% of your customers, who are not offered any of the following options.

    Option Description
    Launch AI chat window Displays a bot chat widget automatically in expanded mode. The drop-down list contains all the AI-enabled chat windows set up in the Bold360 Admin Center and the touchpoints set up in Bold360 AI. Touchpoints where you have defined welcome mode are excluded.

    Optionally, define the minimum amount of time (in seconds) that must elapse after the customer viewed a page and before a chat window is shown. To do so, click the ellipses icon (...) to the right and then click Show more options.

    Launch chat window Displays a minimized agent chat window automatically. As soon as the agent accepts the chat, the window is expanded.

    Optionally, define whether a Bold360 department should be available and the minimum amount of time (in seconds) that must elapse after the customer viewed a page and before a chat window is shown. To do so, click the ellipses icon (...) to the right and then click Show more options.

    Note: The chat window is displayed to customers only if the following conditions are met:
    • Website is open (based on website business hours)
    • Department is open and available ( based on department business hours)
    • ACD assigned the chat to an agent
    • Agent accepted the chat and sent the first message, which can be either a welcome message or something typed in

    When you set up a chat window, it is recommended to use a pre-chat form and define a department specifically for your Acquire campaign-related chats.

    Show custom invitation Displays a pre-defined invitation on the page. Both JavaScript and simple image type invitations can be used. Contact your Customer Success manager to set up a custom invitation.

    Optionally, define whether a Bold360 department should be available and the minimum amount of time (in seconds) that must elapse after the customer viewed a page and before a chat window is shown. To do so, click the ellipses icon (...) to the right and then click Show more options.

    Welcome chat window You can select those touchpoints where welcome mode is enabled. When the campaign action loads, there will be a small chat button with a welcome message. For more information, see How to define welcome messages for chat invitations.

    The widget preview on the right provides a visual representation of the actions of your outreach strategy. To experiment with your outreach strategy, you can split your audience to a control group and a treatment group to perform A/B testing on the performance of your widgets and invitations. By naming your default actions, you can also run reports on those actions. You can run reports to monitor the effectiveness of the individual strategy variations, including your default actions.

  6. Save your sales outreach:
    • Click Save Settings to save it as a draft. When you define a future start date, this option starts your outreach on the specified date.
    • Click Save and Start Now to launch the outreach.
      Note: The actual start date depends on what you defined when setting up an outreach.

How to differentiate customer experience

You can do the following to customize customer experience:
  • Leverage extra signals as triggers and context for the bot
  • Customize touchpoints
  • Set up a welcome message
  • Assign bot or agent depending on your outreach campaign
  • Delay displaying the campaign action for the end-user
  • Leverage the Extended Visitor Information engine to push data from third-parties that can be used later either to trigger a campaign or as context for the bot

How to manage an existing outreach

The outreach list provides an overview of your outreach strategy, including their current status and the number of treatment variations you set up to approach your customers. To update your outreach activities, do the following:

  1. On the Engagement > Outreach page, select an outreach to edit.
    • Select an Idle sales outreach to start it early or update its details
    • Select a Running sales outreach to manually end it.
    Note: To modify a running outreach campaign, you must first stop that outreach, copy it, and then start it with the modified settings.

  2. Update and save your sales outreach.
    • Click Save Settings to update the details of an outreach
    • Click End Now to close a running sales outreach
    • Click Save and Start Now to save the updated details of an outreach and start it immediately

Build a successful Acquire campaign

***These capabilities are in reference to accounts where Bold360 Acquire is currently enabled. We are no longer enabling Acquire on accounts.

With so many products and services to choose from, website buyers have a hard time determining what is right for them. Therefore, they often leave without making a purchase.

Acquire proactively guides website visitors through the discovery process and remove barriers to checkout with seamless proactive chatbot and human engagements that deliver personalized assistance at the right time.

By setting up an outreach campaign, you can decide how to get in touch with customers already on the website and how to approach them before a chat with the bot or a live agent starts.

For example, you can also differentiate between customers based on cart size. In this case, you can define two separate outreach activities: one for low-value cart size when the bot engages the customer, and another one for high-value cart size when a live agent helps the customer purchase items from your online store. Depending on cart size, the customer has different shopping experience.

You can use combinations of predefined invitations and widgets and essentially perform A/B test on your outreach campaign. To follow the progress of your outreach activities, see the Outreach Activity and Conversion Monitor.

You can define both proactive Acquire and regular (reactive) bots and chat windows for your website: the proactive Acquire deployment would serve as a sales bot, while the regular chat window would provide standard customer support. If you already have a chat window implemented, you can easily extend that window with Acquire capabilities.

Why would you define an outreach campaign

Setting up and running an outreach activity enables you to increase your topline revenue. Whether your intention is to provide information, help customers through a sales process, or promote purchasing your products to your potential customers, Acquire provides the option to proactively engage your potential customers and offers the following additional benefits:

  • Differentiate between customers based on low/high value leads and conversions
  • Assign the best persona. For example, based on hesitation or cart size, either a chatbot or a live agent engages the customer
  • Decide when to run and who to target proactively
  • Reactive and proactive deployments co-exist in harmony
  • Personalized treatments, such as custom JavaScript implementation beside regular bot or agent engagements
  • Provide insight into how to improve customer experience
  • Outreach campaigns can co-exist with regular AI and agent implementations in the website. In other words, having a static entry point (chat button) in the website's code does not affect Acquire's proactive actions.
  • You can extend an existing code snippet with Acquire related entries, such as a new website ID

What is Smart Advisor?

Smart Advisor suggests answers to the agents in real-time based on what the visitor is asking in the chat.

Important: Smart Advisor is available to agents if Bold360 Agent is integrated with Bold360 AI and it is enabled on both platforms. For more information contact your Customer Success Manager.

Smart Advisor is the intelligent search available in the Agent Workspace that agents can use while chatting with clients. It provides agents with access to the company?s knowledge base and enables collaboration between agents who chat with customers and knowledge base managers who create support and chatbot content. To learn about the difference between Smart Advisor and canned messages that are available in the Bold360 Agent platform, read this article.

Onboarding Cards

Smart Advisor displays onboarding cards by default which enable agents to get up to speed on how to use the Smart Advisor. The content of the onboarding cards reflects the options enabled for Smart Advisor. For example, if comments are enabled, the related onboarding card is displayed.

FAQ

To increase agents? interaction with Smart Advisor, you can display an FAQ in it. This FAQ can include articles that provide answers to hot topics: for example, if there is an outage, knowledge base managers can create an article about it and add it to the FAQ so that that?s the first thing agents see in Smart Advisor.

Context

To control what content is displayed in the Smart Advisor, the touchpoint can be set to take one or more contexts into consideration and only display articles that have the specified context. To learn more about how to set up the touchpoint in Bold360 AI, see this article.

Article Information

Smart Advisor can display information about who updated the article last and when it was last updated. This information can be useful to agents: they can see whether the information is up to date and who they can contact about the article.

Labels

Smart Advisor can also display the labels associated with the article so that agents can decide if the content of the article is relevant to their query.

Articles with Links

Smart Advisor supports articles with links. Knowledge base managers can create articles that cover the initial question and detail the specifics in separate articles that readers can get to using links. This way agents can find the answers they're looking for in an interactive way.

Comments

Agents can leave comments on how to improve an existing article. These comments are visible to all other agents in the Agent Workspace. Knowledge managers can find these comments on the Comments tab in the Article Editor for the relevant article.

Article Suggestion

Agents have two options to suggest content:

  • By choosing the Suggest as article option next to an answer they just entered in the chat
  • By choosing Suggest Article after their search yields no results

Either way, an article editor is displayed and they can provide a title for the article and enter content in the description field and submit their suggestion. Knowledge managers can find these articles in the Bold360 AI Platform under Knowledge > Suggested articles.

Confidential Content

When Smart Advisor is enabled, knowledge managers can mark parts of an article?s content as confidential. Such parts are displayed differently in Smart Advisor and agents can?t copy them thus reducing the risk of sharing confidential information with people outside the company. For more information see How can I mark parts of an article as confidential?

What is Smart Advisor 2.0?

The enhanced Smart Advisor 2.0 suggests answers to the agents in real-time based on what the visitor is asking in the chat.

Important: Both versions of Smart Advisor are available to agents if Bold360 Agent is integrated with Bold360 AI and is enabled on both platforms. For more information, contact your Customer Success Manager.

Smart Advisor is the intelligent search available in the Agent Workspace that agents can use while chatting with clients. It provides agents with access to the company’s knowledge base and enables collaboration between agents who chat with customers and knowledge base managers who create support and chatbot content. Smart Advisor 2.0 provides the following features:

Onboarding Cards

Smart Advisor displays onboarding cards by default, which enable agents to get up to speed on how to use the Smart Advisor panel. The content of the onboarding cards reflects the options enabled for Smart Advisor. For example, if comments are enabled, the related onboarding card is displayed.

FAQ

Smart Advisor can display an FAQ, which may include articles that provide answers to hot topics: for example, if there is an outage, knowledge base managers can add an article to the FAQ so that agents see that article first in Smart Advisor.

Context

To control what content is displayed in the Smart Advisor, the touchpoint can be set to take one or more contexts into consideration and only display articles that have the specified context. To learn more about how to set up the touchpoint in Bold360 AI, see this article in the Bold360 AI platform support center.

Article Information

Smart Advisor can display information about who updated the article last and when it was last updated. This information can be useful to agents: they can see whether the information is up to date and who they can contact about the article.

Labels

Smart Advisor can also display the labels associated with the article so that agents can decide if the content of the article is relevant to their query.

Articles with Links

Smart Advisor supports articles with links. Knowledge base managers can create articles that cover the initial question and detail the specifics in separate articles that readers can get to using quick buttons or links. This way agents can find the answers they’re looking for in an interactive way.

Comments

Agents can leave comments on how to improve an existing article. These comments are visible to all other agents in the Agent Workspace. Knowledge managers can find these comments on the Comments tab in the Article Editor for the relevant article.

Article Suggestion

Agents have two options to suggest content:

  • By choosing the Suggest as article option next to an answer they just entered in the chat
  • By choosing Suggest Article after their search yields no results

Either way, an article editor is displayed and they can provide a title for the article and enter content in the description field and submit their suggestion. Knowledge managers can find these articles in the Bold360 AI platform under Knowledge > Suggested articles.

Confidential Content

When Smart Advisor is enabled, knowledge managers can mark parts of an article’s content as confidential. Such parts are displayed differently in Smart Advisor and agents can’t copy them thus reducing the risk of sharing confidential information with people outside the company. For more information, see How can I mark parts of an article as confidential? in the Bold360 AI platform support center.

Support Center Customization

The support center is fully HTML supported and you can change the templates within its HTML code.

Tip: Check out this video to see support center configuration in action.
This video is currently under construction. Thank you for your patience.
  1. Go to Touchpoints > Support Center and click the Page Templates tab.

    Result: The HTML Master page set-up is displayed, which you can then customize using placeholders (documented here).

  2. Use the special syntax in the tables below to insert the dynamic content (i.e article body) to your pages.
    Table 1. Entity Selector Placeholders

    Attributes

    Syntax

    Description

    Name

    Type

     

    < nR:Entity type="label" id="B025" >

    Selects the specified label by ID

    label

    Entity Selector

     

    < nR:Entity type="article" id="B025" >

    Selects the specified article by ID

    article

    Entity Selector

    Table 2. Repeater Placeholders

    Attributes

    Syntax

    Description

    Name

    Type

    labelId = -1 articles with no parent labellabelId = < %Label.ID% > current label ID, maxItems, sort(=faq), days

    < nR:Repeater type="article" sort='faq' labelId='1234' > ... < /nR:Repeater >

    Repeater's enclosed code for each article in the specified enumeration. For information on sorting articles in your Support Center, see How to change the sorting of support center articles

    article

    Repeater

    labelId, maxItems, sort(=faq), days

    < nR:Repeater type="attachment"> ... < /nR:Repeater >

    Repeater's enclosed code for each attachment for current article

    attachment

    Repeater

     

    < nR:Repeater type="phrasings"> ... < /nR:Repeater >

    Repeater's enclosed code for each phrasing for current article

    phrasing

    Repeater

    type='root'/'all'/'nestedUnder/list' maxItems='10' (OPTIONAL)

    < nR:Repeater type="label"> ... < /nR:Repeater >

    Repeats for each Label

    Label

    Repeater

      < nR:IF condition="Site.IsMultiKB" > < nR:Repeater type="language" > < a href="< %Lang.URL% >" >< %Lang.Code% >< /a > < /nR:Repeater > < /nR:IF > Repeats for each separate language KB Language KB Repeater
    Table 3. Value Placeholders

    Attributes

    Syntax

    Description

    Name

    Type

     

    < %Site.Name% >

    Configured site name

    Site.Name

    Value

     

    < %Site.Domain% >

    Configured site domain

    Site.Domain

    Value

     

    < %Site.URL% >

    Configured site URL (includes http/s)

    Site.URL

    Value

     

    < %Site.Account% >

    Configured site Account

    Site.Account

    Value

     

    < %Site.KBID% >

    Current knowledge-base ID

    Site.KBID

    Value

     

    < %Label.Name% >

    Label name

    Label.Name

    Value

     

    < %Label.ID% >

    Label ID

    Label.ID

    Value

     

    < %Label.Link% >

    Link to dedicated label page

    Label.Link

    Value

     

    < %Label.Count% >

    Amount of elements in label

    Label.Count

    Value

     

    < %Label.PositionInPath% >

    0 based index of position in current path, -1 returned if label is not in path

    Label.PositionInPath

    Value

     

    < %Article.ID% >

     

    Article.ID

    Value

     

    < %Article.Title% >

     

    Article.Title

    Value

     

    < %Article.Body% >

     

    Article.Body

    Value

     

    < %Article.Link% >

     

    Article.Link

    Value

     

    < %Article.AttachmentCount% >

     

    Article.AttachmentCount

    Value

     

    < %Article.Attachment.URL% >

     

    Article.Attachment.URL

    Value

     

    < %Article.Attachment.Name% >

     

    Article.Attachment.Name

    Value

     

    < %Article.MetaDescription% >

    Articles Meta-Description

    Article.MetaDescription

    Value

     

    < %Article.PhrasingCount% >

     

    Article.PhrasingCount

    Value

     

    < %Article.Phrasing% >

     

    Article.Phrasing

    Value

     

    < %Repeater.Ordinal% >

    The ordinal for current repeater loop

    Repeater.Ordinal

    Value

    Use in condition "Page.Type=Home" to check page type:Home/Label/Article

    < %Page.Type% >

    Determines the current page type

    Page.Type

    Value

    checks if current Label is the current page label

    < nR:IF condition= "Article.IsPageArticle" >

     

    Article.IsPageArticle

    Value (condition only)

    check if the current Article is the current page article

    < nR:IF condition= "Label.IsPageLabel" >

     

    Label.IsPageLabel

    Value (condition only)

     

    < %Site.NanorepDomain% >

    Places Digital DX AI servers domains

    Site.NanorepDomain

    Value

     

    < %Page.Error% >

    The error that is displayed

       

What are Bold360 AI contextual answers?

Bold360 AI contextual answers is a revolutionary automated solution that delivers context-based answers to your customers and helps drive sales and conversion rates while simultaneously reducing site abandonment.

Tip: Check out this video to see how context and article variations work.

This video is currently under construction. Thank you for your patience.

With contextual answers, you can control the visitor's experience. Visitors who ask the same question will receive different, tailored answers based on segmenting information available about them. You can create relevant answers for visitors based on the specific product or page that they are viewing, their subscription type (free/premium), their digital click stream, account balance, age, gender, their geographical location and many more. You can also associate different channeling options to different visitors using context.

In the AI Console, context can be associated with visitors and articles. Both can have none to several context values:

  • The visitor?s context is set by the touchpoint. For instance, on web pages of a travel site which are related to hotels, the script calling the touchpoint will include a Product=Hotel parameter.
  • An article may have several answers, each of which is mapped to one or more context values. For example, for the question "How do I change my reservation?" there could be one answer for flights and another for hotels.

How do I set up contextual answers?

Important: Contact your Customer Success Manager to enable contextual answers in your account and to add the required context categories to your knowledge base.

Once contextual answers are enabled for your account and the context categories are available, proceed as follows:

  1. Add values to context category
  2. Add context values to articles
  3. Add context definition to website script
  4. Implement context in Support Center
  5. Implement contextual answers in FAQ
  6. Use context in channeling policies

How do I define an article's context?

Add context to articles or article variations to control when and how they are displayed.

Context must first be set up for your account, as described here.

Tip: Check out this video to see how context and article variations work.
This video is currently under construction. Thank you for your patience.
  1. In the AI Console, go to Knowledge > Articles and create a new article or open an existing one.
  2. Below the title of the answer, click on a context category to see the list of context values.

  3. Select one or more values from the drop-down list.
  4. Optional: Set other context categories.
  5. Publish the article.
Tip: To create different answers for the same question but for different audiences, use article variations as described here. For example, for the question "How do I change my reservation?" one answer could be for flights and another could be for hotels.

Synonyms and conversational flows

Synonyms are used to enhance your Knowledge Base (KB) and improve search results. There are two types of synonyms in Bold360 AI:

  • Local synonyms: The synonym repository in the KB which is created and maintained by your content management team. Local synonyms typically include words that are related to your business and serve as an extension to the list of global synonyms.
  • Global synonyms: Language-specific global synonyms that Bold360 AI uses which are relevant across all verticals.

For your conversational flows, synonyms are especially important as you will have to anticipate the different ways customers are looking for your goods or services.

The following is an example of global and local synonyms as seen in the Article Editor:

When you look at the synonyms of create, you see that "add" and "make" are defined locally. At the same time, "build" and "open" are defined as global synonyms that are also applied when a customer enters an intent.

What is the contextual answer experience for customers?

As the end user searches in the widget, Digital DX AI will display relevant results from answers with the corresponding context. These results are in addition to the generic answers (answers without context) that are displayed. Answers from conflicting contexts are not shown.

For example, in context 'tv', the user will get answers that are relevant to the TV, such as size.

Learn more

How to use context values in article titles

You can set different context values in the title of an article so that related content is maintained in the same article.

Note: You must have the Enable Conversation feature switched on for your account to use this feature.

Contextual answers Value type

Each contextual category can be defined in various ways, starting from a Normal value where the values are set both in the articles and in the widget (hard-coded) or by determining the context using the user IP address, Country code or domain.

In this article, we will cover the set up for each of the above values in the context categories values types.

Choosing the relevant value type

To choose the relevant context value type you need to enter the context category settings, go to Admin Center > KB Settings > Context tab.

In the context category, click Show additional configuration and all the category configurations will be shown, In the 'Value type' drop-down you will be able to choose the relevant value type for the context category: Normal, Device Type, Geo Location: Country, Geo-Location: Custom Mapping, IP Mapping, and Domain: Custom Mapping

How to configure contextual answers in the AI Console

Add context values to a context category to start using context-based segmentation.

Make sure contextual answers are enabled in your account and context categories have been added to the knowledge base. Contact your Customer Success Manager for more information.
To add context values to the categories created for your account, you must have Modify permission for Settings in the Advanced Credentials settings. For more information see How do I add a new user?
  1. In the AI Console, go to Admin Center > KB Setup.
  2. On the Context tab, find the context category that was created for your account.
  3. In the Context Values field, enter values for the category one by one and use the Add button that appears to add the values to the list.
  4. Optional: Select Show additional configuration to review the context category settings.

    To learn more about what these settings mean, see What options are available for context configuration?.

  5. Save your changes.

How to dynamically instruct Bold360 AI to escalate a ticket into custom email addresses using an API?

You can tell Bold360 AI to send the ticket to additional email addresses by adding the targetEmailRecipient form value.

Example:
 
_nRepData["formValues"].push(["targetEmailRecipient", "direct", "target@email.com"]);

Replace target@email.com with your email address.

Example:
 

To send each ticket to the support team manager, add this line:

_nRepData["formValues"].push(["targetEmailRecipient", "direct", "supportTeamManager@myCompany.com"]);

How do I configure the outgoing email settings (SMTP server configuration)?

To set up outgoing emails, you must configure Bold360 AI ticket email replies to go through your SMTP server

  1. In the AI Console, go to Admin Center > Ticketing Settings.
  2. On the Ticketing Basics tab in the Outgoing email section, configure the following:
    • Under Sending email from, define the name of the outgoing e-mail
    • Under Reply-to address, define the outgoing e-mail address
    • Under Ticket status on reply, decide what happens to the ticket once a respond is sent; will it be closed on first reply or always closed
    • Select Send emails through custom SMTP server
    • Enter your SMTP server settings.
  3. Click Apply.

How to add additional email fields (CC, BCC) to the contact form?

Add an email address field as well as CC and BCC to the contact form.

  1. Go to Channeling > Contact Forms and click Personalize under the widget you want to customize.
  2. Choose the appropriate URL or knowledge base.
  3. Click on the pencil button to customize the form.
  4. Click Add new field and choose Email field.
  5. Click the down arrow on the right side of the email field to see the settings and under the Textbox type choose CC or BCC.

How to set up the display name of the e-mail sender in Bold360 AI?

  1. In the AI Console, go to Admin Center > Ticketing Settings > Ticketing Basics tab.
  2. Change the sender's name in the Sending email from field.

How can I proactively send an email to a customer from Bold360 AI?

Emails in Bold360 AI are handled through the ticketing system. Therefore, you must first create a new ticket:

  1. In the AI Console, go to Ticketing > Open tickets.
  2. Click the New tickets icon above the list of your tickets.
  3. In the User Email field, enter the customer's email address where you want to send an email.
  4. Make sure that the Send replies to this email address option is selected.
  5. Click Create and Open to create the ticket.
  6. Reply to the newly created ticket and click Send.

How do I set email forwarding to my Bold360 AI inbox?

You can associate your support e-mail address with the Bold360 AI ticketing system.

You can have all your support e-mails handled through one interface and based on a single knowledge base. To forward your support e-mails to the Bold360 AI incoming e-mail address, do the following:
  1. Go to Admin Center > Ticketing Settings > Ticketing Basics tab.
  2. Copy the email address from the Inbox Address field.
  3. Click more addresses and select the label to be attached to the incoming ticket. If you do not select any label, emails will go to your Inbox.
  4. Save your changes.

How to configure Digital DX AI after setting up a messaging channel

Once you have set up a WhatsApp or Microsoft Teams messaging channel, complete additional configuration in Digital DX AI to customize the bot's behavior.

  1. Define the answer the bot returns when it can't answer the customer's question.

    You can set up multiple messages or create articles that the bot returns when the customer's search yields no results. See How to configure conversational touchpoints to learn more about how to set these up.

    Note: You can't use an article for the final message as that prevents the bot from triggering the escalation.
  2. Define channeling policies to cover different scenarios.

    Depending on whether you have a bot-only, agent-only or bot and agent messaging channel, you may define different channeling policies. See How do I define a channeling policy? to learn more about how to create a channeling policy.

    Agent only

    If you want your customers to talk to agents only on the messaging channel, set up a channeling policy with a condition where the number of words is greater than 0. Add a message as an error answer that introduces the escalation option and explains customers that they will be connected to a live agent if it's business hours and agents are available.

    Note: To prevent the bot from providing an answer, you must have an empty knowledge base.
    Bot only If you want your customers to talk to the bot only on the messaging channel, make sure you have defined error answers for when the bot can't find an answer to the customer's query.
    Agent and bot experience

    If you want customers to talk to the bot first and outside business hours but have the option to talk to an agent, you might want to:

    • Define a Day and time condition for the channeling policy that is used to escalate chats to agents
    • Use labels on articles for which you want to provide agent support and define an Article label condition for the channeling policy
What's Next?

See What are the best practices for managing a knowledge base? to learn about how you can optimize the bot's content.

How to monitor traffic on messaging channels

You can monitor the performance of your WhatsApp and Microsoft Teams messaging channels in the AI Console as well as the Agent Workspace.

Monitoring capabilities in the AI Console

You have the following options to run reports and monitor messaging conversations:

  • Engagement Dashboard: provides a view on the total number of engagements, the number of engagements escalated to agents and so on. See What do I see on the Engagements Dashboard? for more information.
  • Dashboard: You can look at voices for the knowledge base you have for messaging, or filter the dashboard using the context you use for messaging to see how your messaging channel is doing. See What do I see on the Dashboard? to learn more.

Monitoring capabilities in the Agent Workspace

The following options are available to run reports and monitor messaging conversations:

  • You can view active bot, queued, assigned to agent as well as closed messaging conversations in the Monitor View of the Agent Workspace.
  • You can run reports on messaging conversation data with Data Extraction API methods. For more information, see the Developer Center.
  • You can run status and summary reports on messaging conversations. For more information see Messaging Apps Reports.

How to define welcome messages for chat invitations

You can now proactively invite your website visitors to chat by displaying a welcome message that could appear above the chat icon in Slim and Harmony widgets.

To invite visitors to chat, do the following:

  1. In the AI Console, go to Touchpoints > Widgets.
  2. Click Personalize under the widget type that you want to change.
  3. Select a knowledge base or URL to customize.
  4. On the Basics tab, make sure you have an Auto Question selected.
  5. On the Position & Size tab, click Customize next to Welcome Mode and select the Invite option.
  6. Optionally, define a delay to display the welcome message in the Welcome mode welcome message trigger time (in seconds) field.
  7. Save your changes.

Define context-sensitive welcome messages

You can use article variations of a selected welcome message to personalize customer experience based on context. Context must be defined in the widget code.

To select a context-specific welcome message, do the following:

  1. In the AI Console, go to Touchpoints > Widgets.
  2. Click Personalize under the widget type that you want to change.
  3. Select a knowledge base or URL to customize.
  4. On the Basics tab, make sure you have an Auto Question selected.
  5. Select the Context-sensitive Auto Question option.
Note: When you select an article for welcome message that has variations and do not select the Context-sensitive Auto Question option, then you must select a variation to be displayed for customers as welcome message.

What are messaging applications in Digital DX AI?

Messaging applications in Digital DX AI enable you to communicate with your customers over popular messaging apps, such as WhatsApp and Microsoft Teams.

Note: Messaging applications are in beta. Contact your Customer Success Manager to learn more.

When you set up a messaging channel, you have the following options depending on your support needs:

  • Provide bot-only support if you do not have human agents
  • Provide agent-only support if you want your customers to always talk to a human agent
  • Use the combination of bot and agent support, if you want the bot to cover out of hours support and help your agents during business hours
Note: You need access to the AI Console if you want to use a bot only on your messaging application. For an agent-only and a bot and agent experience you need to have access to both the Agent Workspace and the AI Console.

Customers can interact with agents on their favorite messaging application in their own pace. The connection between the customer and the agent is permanent from the customer's perspective. Customers can start and stop the conversation in the messaging application anytime without losing connection to the agent.

To learn more about what the experience is like for customers and agents, see these articles:

How to configure messaging channels

Once you created a messaging channel for WhatsApp or Microsoft Teams, you can change the default status messages, define a context, or set up to auto-end conversations.

Make sure you have created a messaging channel.

  1. In the AI Console, go to Touchpoints > Widgets.
  2. Under Conversation configuration, select Messaging applications.
  3. Choose (Configure) next to the channel that you want to configure.
  4. Adjust the options as necessary.
    Status messages

    You can change the default messages for the following cases:

    • The conversation is escalated or closed
    • There is a network issue and the customer's message cannot be delivered
    • There are multiple answers for a customer question
    Contexts If you use context to have article variations for the WhatsApp or Microsoft Teams channel, you can specify the context here to let the bot know which variations it can display to customers. See How do I define an article's context? to learn more.
    Auto-end configurations

    Bold360 can automatically end conversations after the specified time measured from the last customer-sent message. You can define an interval between 30 minutes and 24 hours.

    You can also define a message for customers when their conversation automatically ends.

    Nanorep Entities

    When users ask questions on Microsoft Teams, now they can receive personalized answers . For example, if Bold360 is integrated with a third-party ticketing system and the agent asks How many open tickets do I have?, Bold360 can deliver a response that is specific to the user.

    To enable personalized answers, select the Enable Nanorep Entities option for the Microsoft Teams channel you configured under Touchpoints > Widgets > Messaging applications. Then create articles with personalized answers using entities.

  5. Save your changes.

What are fallback messages?

Fallback messages enable the bot to provide a response when there is no result or there are multiple options based on the user's input.

When there is one result based on the user's responses, you set up the final answer presented to the user in the body of the article that uses the entity. There could be multiple possible results or no result based on the user's responses. you define what happens in these cases in the Fallback Messages section of the entity.

No results

You can define a fallback message that is presented when the user's answers match the data in the entity source file, but there is no result in the entity that fits the customer's answers. Using the example in this article, if the customer is asking for an iPhone 7 with a 10 megapixel camera, the no result fallback message is presented as iPhone 7 and 10 megapixels are both part of the entity file, in model and camera pixels respectively, but there is no iPhone 7 with a 10 megapixel camera in the database. If however, the customer is asking for a Sony phone, which is not one of the brands in the file, the bot returns the error message defined in the widget configuration.

The fallback message for no result can be either a statement or an article. A statement is plain text that is displayed to the user.
Note: You can't select an article that contains an entity as a no result fallback message.

For custom entities, the message or article you define here is displayed when the customer provided wrong input more times than the property's error message is set to be displayed. For example, you have a property where detect is set to Email, the bot query is What's your email address? and the error message is Please provide an email address. which is set to be displayed 3 times. When the user fails to provide an email address for the fourth time, the fallback message or article is displayed.

Multiple results

When there is more than one result to present, you can set up the bot to generate a carousel of these results. In the Fallback Messages section of the entity page, set what cards in the carousel should look like: specify the property to be presented as title and subtitle, and if you have images in the data source, you can also provide an image property. Using the example in this article, you could have:

Tip: If you don't want to have a carousel of different results, you can use Quick Options instead. See What are the property options for data source entities? for more information.

API Invalid response (for API-based entities only)

Note: This option is displayed only when you provide a link to an API in the Data Source URL field. When your entity reads data through an API, the API may return an invalid response. For example, if your API receives invalid input from the customer or when the API server is unavailable. For these cases, you can define an error statement that is to be presented to the customer.

Smart Advisor: Chatbot-assisted conversations

When chatting with a customer, Smart Advisor offers agents potential answers to customer questions either based on the content of the chat conversation or by manual search.

These instant answers ultimately provide better user experience and increase agent efficiency: they allow agents to spend less time searching for answers, and improve Average Speed to Answer (ASA) in reporting.

Note: This feature is only available with a Bold360 AI platform account.

Important: You must have the Actions > Chats > Suggest response as Smart Advisor article permission enabled to use this feature.

During a conversation with the chatbot, customers have the option to connect to an available agent, if there are any. When the chat is transferred, the agent can view the full history of the conversation between the customer and the chatbot and continue that thread. The customer and the agent always use the same set of knowledge base articles.

When the customer sends a message to an agent, the chatbot analyzes that message and offers the most relevant answers to the customer's messages. These answers are listed on the Smart Advisor panel, which replaces the Canned Messages panel when the customer chats in a Bold360 ai chat window. Smart Advisor considers only those customer messages that have been sent since the agent's last interaction.

To copy the best answer to the message panel, click that response. You can edit your message before sending.

If your articles have contexts as well, you can select the relevant article variation from Smart Advisor panel.

Search Smart Advisor for help

You can manually search Smart Advisor articles when you are looking for results not listed automatically. To do so, type your question in the Search field. Automatic suggestions are provided as you type. Titles of the relevant articles are listed dynamically as you type your question. To see the content of an article, click its title. All previously displayed results are removed from the Smart Advisor panel.

To get the best search results, consider the following:

  • Grammatically accurate words return more relevant answers
  • Avoid very long phrases and generic statements

How to rate a Smart Advisor response

When Smart Advisor offers you responses, agents can always rate those responses.

  1. Click the arrow on the right of the suggested response to expand the article if not fully visible.
  2. Click Yes or No at the bottom of the article to rate whether the article was helpful.

    When you rate an article as not helpful, you can further categorize the problem with the article and suggest improvements. This rating improves the results that Smart Advisor provides.

How to suggest a response as a Smart Advisor article

When you cannot find a relevant answer to a frequent question, you can suggest your response as a Smart Advisor article. After approval, these articles are immediately available to every agent. To make a suggestion, do the following:

  1. Hover the mouse over one of your previous answers.
  2. Click Suggest as article .

Your answer must first be approved by a Bold360 ai knowledge base content manager before it becomes an article. You will not get any notification when your answer is approved.

How to set up the Smart Advisor touchpoint

Smart Advisor provides agents access to the company's knowledge base and offers them answers real-time based on what the visitor is asking in the chat.

Important: Smart Advisor is available to agents if the Agent Workspace is integrated with Digital DX AI and it is enabled on both platforms. For more information contact your Customer Success Manager.
  1. In the AI Console, go to Touchpoints > Widgets and choose Personalize for the Smart Advisor widget.
  2. Select a URL or knowledge base to customize.
  3. On the Basics tab, change the configuration based on your business needs.

    You have the following options:

    Name Description
    Display Article Last Update Select this checkbox to display when an article was last updated and who did the last update in the Smart Advisor. This information can be useful to agents: they can see whether the information is up to date and who they can contact about the article.
    Display Article Labels Select this checkbox to display labels associated with an article in the Smart Advisor. Labels can help agents decide if the content of the article is relevant to their query.
    Context info Add the context values to be considered when articles are displayed in Smart Advisor. This controls what content agents can access.
    Auto-complete count Enter the number of suggestions to be displayed in the auto-complete. If you enter 0, auto-complete is disabled. You can display up to 10 articles in the auto-complete.
    Comments Select this checkbox to enable agents to leave comments on existing articles. These comments are visible to all other agents in the Agent Workspace. Knowledge managers can find these comments on the Comments tab in the Article Editor for the relevant article.
  4. On the FAQ tab, set up an FAQ to display in Smart Advisor.
  5. Save your changes.

How to manage localization in the Digital DX  AI platform?

One of the most efficient ways to manage translations/localization is to use GeoFluent, which works in search and conversational widgets alike. However, if you do not have a GeoFluent account, you can still use Digital DX AI's built-in translation management feature: as a Knowledge Manager, you can control the localization of individual articles, directly from the Article Editor.

Important:
  • This solution is limited to search widgets.
  • You must have at least two knowledge bases set up on your account that are part of the same translation group. Contact your Customer Success Manager to set up additional knowledge bases.

Even though this solution is available in search widgets only, it works in conjunction with auto-translation:

  • If there is an article in the user's language, then that is presented
  • If there is no relevant article, then the English KB is searched and the response is translated on-the-fly with Google Translate

To create a localized version of an article, do the following:

  1. In the AI Console, go to Knowledge > Articles.
  2. Open an article that you want to localize.
  3. On the Localization tab of the Article Editor, click Add localization.
  4. From the Choose KB drop-down list, select the knowledge base where you want to create the localized version of the selected article.
  5. Define the appropriate translation for the selected article:
    • On the Choose an existing answer tab, you can search for an article in the target knowledge base that contains the translation of your article (if you know there is such an article)
    • Otherwise, on the Write Answer tab, you can verify that the translation provided by Google Translate is correct
  6. On the Write Answer tab, add the necessary phrasings to the localized articles and make sure the article is Set Live.
  7. Save your changes.

    The Localization tab, now lists all localized versions of the article. The list includes the name of the user who last edited the localized article, the knowledge base that contains the localized article and the title of the article. Click View next to a localization to open the translated version of the article.

  8. Publish the original article to make the localized versions available.

The number of localized versions of an article is displayed on the Articles page:

Digital DX AI setup

The Digital DX AI platform offers customers chatbot-enabled self-service and introduces Smart Advisor, which provides agents with the best possible answers to customer questions during chat. The chatbot handles customer queries and provides a consistent front-line customer care experience. This way, Digital DX delivers the best of both chatbots and agents to solve customer issues.

Note: This feature is only available with a Digital DXAI platform account.

With your Digital DX AI platform account, you also get access to the Web Admin Console, where you can customize your AI-enabled chat window. Here you can also manage the content of the knowledge base that contains help articles that both the chatbot and the Smart Advisor panel show. The Digital DX AI platform is a service that provides a chat window with AI-enabled customer assistance after it is integrated into Digital DX. The layout of the chat window and the content of the help articles are controlled by the AI Console.

To use Digital DX AI, you must set up a link between the Web Admin Center and the AI Console by creating API keys:

  • To use the chatbot, create an API key, which connects the Digital DX AI and Agent platforms
  • To set up channeling, create a Digital DX Agent API access key. Depending on the number of channeling paths, you may have to create multiple Digital DX API access keys.

If set up correctly, customers receive self-service help from the chatbot and agents can use Smart Advisor to answer customer queries. See Smart Advisor: Chatbot-assisted conversations.

About channeling

When your customers request help from a human agent while interacting with the bot, the chat is escalated to a live agent in your organization. In the Web Admin Console (for the AI platform), chat escalation is called channeling. You may set up multiple channels to cater for your customers' needs. For example, you may use different knowledge bases to serve different audiences, in which case you must set up a channel for each knowledge base. With the Digital DX Agent API access key, channel settings are linked to a chat window and during channeling these settings are applied.

Whether customers receive help from the bot or chat with a live agent, they see the same chat window. Customers do not experience any change when a chat is seamlessly transferred from the chatbot to a live agent. They continue the conversation in the same window.


Figure 1. Customer's view when Digital DX AI is activated

How do I add a link to a Bold360 AI answer?

To help your customers find related information, you can add links to other articles in your knowledge base.

  1. In the Article Editor, click on the (link) icon in the answer toolbar to open the Link dialog.
  2. On the Link Info tab, enter the text that you want to display as a link.
  3. Select Article from the Link Type drop-down.
  4. Optional: Select the Link to specific variation checkbox.

    When you create a link to an article with multiple variations, by default, Bold360 AI chooses the correct variation based on the context. This option enables you to create a link to a specific article variation yourself.

  5. In the Select an answer field, start typing the title of the article that you want to display once the link is clicked.

    Result: A list of matches is displayed.

  6. Select an article.

    If you selected the Link to specific variation checkbox, you can find and select a specific variation of an article.

Example:
 

For an example, we have created this link. Clicking it will direct you to the linked Bold360 AI answer.

How to set the bot to remain active in channeled conversations?

When a conversation is channeled to a live agent, the bot remains active and offers auto-complete suggestions for customer questions. The live agent can also contribute to the conversation. This option is disabled by default.

To enable hybrid conversation, do the following:

  1. In the AI Console, go to Touchpoints > Widgets and click Personalize on the Floating widget pane.
  2. Select the Knowledge Base or URL for your conversational widget.
  3. On the Advanced tab, select Bot suggestions remain active after channeling.
  4. Save your changes.

What are the best practices for ending a conversation with the bot?

Learn how to best set up the end of the conversation with the bot to make the most of the bot experience.

We recommend you provide a way to reengage the bot when your interaction is over, for example, as a response to Goodbye set Happy to help! Say 'hi' again if you need anything.. This way you make it easy for the user to reengage the bot when they come back.

What are the best practices for the bot's introduction message?

Learn how to best set up an introduction message for the bot to make the most of the bot experience.

When the widget is presented, display an introduction that follows these guidelines to set proper expectations:

  • Let customers know they are chatting with a bot or virtual assistant and not a live person. We recommend you include terms like bot or virtual assistant in the introduction message. Be consistent in how you refer to the chatbot.

  • Give a few examples or some explanation on what the bot can and can't answer, for example, I understand short and simple questions best.
  • Tell customers what happens if the bot can't help, for example, they can start a live chat or call customer service, depending on what channeling you have set up.

Making clear that the customer is engaging with a bot and not a human being sets the customer?s expectations and prevents disappointment and frustration. It also encourages the customer to phrase their questions appropriately.

Tip: We recommend adding the introduction message as an FAQ title in the widget configuration, rather than specifying an auto question. An auto question may count as an interaction and therefore engagement and as such could end up in reporting.

How to enable hybrid (bot and agent) conversations?

When a conversation is channeled to a live agent, the bot may remain active and offer auto-complete suggestions for customer questions. The live agent can also contribute to the conversation. This option is disabled by default.

To enable hybrid conversation, do the following:

  1. In the AI Console, go to Touchpoints > Widgets and click Personalize on the Floating widget pane.
  2. Select the Knowledge Base or URL for your conversational widget.
  3. On the Advanced tab, select Bot suggestions remain active after channeling.
  4. Save your changes.

How to enable a conversational bot in an account

Before you begin: If this is the first time you are adding a conversational widget to your account, please contact your Customer Experience Manager or Support team to make sure that the knowledge base you are working on is enabled for conversations.

To enable a conversational bot in an account, do the following:

  1. In the AI Console, go to Touchpoints > Widgets.
  2. Under Floating widget, click Personalize.
  3. Select the Knowledge Base or URL that you want to change.
  4. On the Basics tab, make sure that Widget Mode is set to Conversational.
  5. Save your changes.

What is a conversational bot and why do I need it?

Bold360 AI platform delivers a conversational bot solution that leverages our Contextual-Answers technology together with our proprietary Natural Language Processing solution.

The conversational platform provides a messaging-based user experience in which the Bold360 AI engine gathers information from the customer similarly to how a human agent would do in a regular conversation. Based on customer input and the engine's search capabilities, the conversational bot provides the customer relevant information.

Like other Bold360 AI widgets, the conversational bot can be deployed across multiple touchpoints. Bold360 AI allows for dynamic selection of parameters that can be based either on integration with your API or on dynamic lists that you provide to retrieve information. Here's what makes the conversational bot unique:

  • Continuous conversation
  • Conversational prompts
  • Dynamic content and personalization
  • Buttons and selection options
  • Image Galleries
Continuous conversations

You can customize the bot's "memory" and define the context in which the bot will remember pieces of the conversation. The bot's ability to remember context enables the conversation. In the following example, the bot remembers that the context is currency exchange:

Conversational prompts

Just as with live agents engaging in real conversations, you can teach the bot to ask follow-up questions to gather the information it needs to answer the customer's question. In essence, the bot identifies the customer's intent and prompts for further information.

The following is an example with multiple prompts:

Dynamic content and personalization

You can define one article (one answer) that would answer a question based on dynamic information that comes from either an API or a dynamic table. The integration with your API can drive personalization in the bot and can answer questions like "What is the status of my claim?" or "How many miles are available for me to redeem?"

You can also use a dynamic table to retrieve information based on context as shown in the following example, where a single article is used to retrieve parameter-based information.

Example

Buttons and selection options

Buttons allow you to add structure to the bot conversation and offer article-specific and follow-up actions. These buttons can either point to other articles/content, open a URL, or offer click-to-call.

Example (Buttons leading to articles/options):

Example (Buttons leading website or click to call)

Image Galleries

Image galleries help bring life and color into the bot conversation. You can use multiple images that the customer can scroll through and take action. You can also use them for product pictures and offers and have the user select between different products.

Example

What is a Bold360 AI chat bot?

Bold360 AI delivers a cutting-edge support chat bot solution utilizing the combination of our patented Contextual-Answers technology and proprietary Natural Language Processing (NLP) algorithm.

We enable customers to speak freely and be understood in any language, as if they were speaking to a live agent.

Deploying Bold360 AI's messenger bot as part of our holistic self-service solution creates a seamless, consistent self-service experience across all touchpoints.

How can I embed fonts in a Bold360 AI support center or widget?

To use paid font types in your Bold360 AI support center or widget, store them on your server and allow Bold360 AI to access them.

  1. Create a CSS file with font face declarations and store it on a server.
  2. Enable cross-origin resource sharing (CORS) on the server as described here.
  3. Allow access from the following resources:
    1. < name of account >.Nanorep.co. For example, if your account name is acme, allow access to acme.Nanorep.co
    2. For support centers, allow access to the support center sub-domain URL. For example, if your sub-domain is support.acme.com, allow access to this sub-domain.

How to prevent access to a support center's inner URL

Every article has an inner and an outer URL.

So, for example, you have:

URL # 1: https://support.mysite.com/Contact-us/864734431/Contact-us.htm

  • An internal iframe URL such as: https://support.mysite.com/Contact-us/864734431/Contact-us.htm
  • An outer URL (where the iframe is inserted): https://www.mysite.com/en/support#/path/Contact-us/864734431/Contact-us.htm

There are two ways to proceed from here:

  1. Change the contact us page address from the internal URL to the outer URL
  2. Ask your Bold360 AI Customer Success Manager to leverage the option that prevents the opening of a support center iframe as a standalone. Instead, if we detect that the page doesn't have a parent, we automatically redirect to the outer URL. In this option you don't have to change the addresses.

Please get back to us on which option you would rather use.

How do I create support center structure with labels?

Use labels to give structure to your support center home page.

Note: You can only use user-created labels for this purpose. There are built-in labels that are used to indicate the availability of an article, like Offline or Hide from Support Center and the status of an article, if content management workflow is enabled, like Draft and Never Published, etc.
  1. In the AI Console, go to Touchpoints > Support Center.
  2. On the Settings tab, go to the Content section.
  3. Under Navigation Labels, select categories to display as navigation options in your support center, for example, along the left side of the page.

    You can change how and where these labels are displayed by adjusting the support page templates.

  4. Under Homepage Labels, select categories to display on your home page, for example, in the main body of your site.

    You can change how and where these labels are displayed by adjusting the support page templates.

    Example:

  5. Save your changes.

How can I add icons to labels in the support center?

You can add three icons to your support center labels: main, hover and active.

By default, if a label has a main icon only, it is used as the hover and active icon as well.

  1. In the AI Console, go to Touchpoints > Support Center.
  2. Select the support center and on the Settings tab, go to the Content section.
  3. Select Manage labels.
  4. On the Content labels dialog, choose the label you want to edit or create a new one.
  5. Choose for Main Icon and upload an image.

    Example:

  6. Optional: If you'd like to show different icons for active or hover, unselect Use main icon and upload an icon.

    Example:

Note: For these fields to affect the support center, PSO work is needed (for new customers this will work automatically).

How do I order labels and sub-labels in the support center?

Set the order of labels and sub-labels shown in your support center.

  1. In the AI Console, go to Touchpoints > Support Center > Settings.
  2. Scroll down to the Content section and select Manage Labels.

    Result: A list of labels is displayed.

  3. Under Manage All, drag and drop to set the order of the labels.
  4. In the Edit section, you can customize how a label is displayed.
  5. You can also set an active and a hover icon for each label.
  6. Save your changes.

How to redirect an iframe URL in support center

If your address for the support center is www.myaccount.com/helpcenter then all the rest of the address is being populated into the iframe's URL.

This means that www.myaccount.com/helpcenter/label would become help.myaccount.com/label.

To setup redirection, do the following:

  1. In the AI Console, go to Touchpoints > Support Center.
  2. Choose the support center that you want to modify.
  3. On the Redirect URLs tab, select Add Value.
  4. In the URL From field, enter the source URL.
    Tip: To redirect an article URL, use the following syntax in the source URL: */articleID/*.
  5. In the URL To field, enter the URL you would like to be displayed.
  6. Choose Save Changes.

The URL in the URL From field will return a 301 (redirect) as HTTP response when the page is visited.

How to set the context in an iframe support center

In order to set the context, add a js function to your hosting page:

<script>
function setContextNanorepSupportPage(){
  return "< context Type >=< context Value >"
}
< /script >
Important: Remove the context from the iframe src if it was set there. For example:
<script>
function setContextNanorepSupportPage(){
  return "Device=Mobile"
}
< /script >

How to set up a support center in the AI Console

Digital DX AI offers a built-in, SEO-enabled, fully HTML-customizable and mobile-optimized support center.

While the out-of-the-box look of the support center template is basic, you have the option to completely tailor the external appearance to reflect your brand. The HTML customization incorporates editable HTML pages, the automatic creation of metadata and SEO-optimized links for articles and related topics.

To set up a support center, proceed as follows:

  1. Create labels and add them to articles.

    Use labels to organize content in the knowledge base and to structure content in the support center. Labels are categories that can be assigned to articles in order to make organization of your support center content easier and more efficient. Labels can be organized into hierarchies which is reflected in the support center and they can also be used to search for specific topics and related answers. This ensures that your agents can easily access answers and thus provide the highest levels of customer service.

    Tip: If an article is relevant for more than one category, you can categorize the article with multiple labels and display it in several categories simultaneously.
  2. Add a support center in the AI Console.
  3. Create a support center structure using labels.
  4. Optional: Change the look and feel of your support center by adjusting the support center page templates.

    You can easily change the fonts, colors or other elements that make up the look and feel of your support center using simple HTML customization. To change the page templates, go to Touchpoints > Support Page > Page Templates.

How to add a new support center in the AI Console

With a single Digital DX account you can offer multiple support centers across different parts of your website that offer answers to different user groups or device types.

To get the score for activities on the support page, allocate a subdomain to the support center from your main website's domain. Specify a CNAME record to point from the support center domain to < < account name > >.nanorep.co so that the subdomain can use the Digital DX server.

Tip: You can create multiple support centers per knowledge base.
  1. In the AI Console, go to Touchpoints > Support Center.
  2. Select Add Support Center.
  3. When prompted, enter the name of the support center and choose Save.

    Result: You are taken to the Support Center: Settings page.

  4. Under General Settings, enter the main URL of your website in the Main Site URL field.

    This field contains your website main URL when you use the Support Center in an iframe. If this is the subdomain you want to use, your Main Site URL can be the same as the Support center domain.

  5. Under Site Settings, fill the Support center domain field.

    This field contains the subdomain used to access this support center. You will also need to create a CNAME record for this subdomain which you can typically do at your domain DNS provider side.

    Note: CNAME changes don't take effect instantly because of the DNS propagation, you will have to wait for some time before it is updated.
  6. Save your changes.

Technical Preparations to implement a support center

If you have initial content that can be imported, it is a matter of two to three weeks until you can get Bold360 AI up and running.

Setting up a subdomain with a CNAME

The support center runs on Bold360 AI servers. In order for you to get the score for activities in the support page, you will need to allocate a subdomain out of your main website's domain. In order for your subdomain to use the Bold360 AI server, please specify a CNAME record to point from the support domain to < < account name > >.Nanorep.co.

Once the CNAME is set up, please contact your Customer Success Manager and inform them about this new subdomain URL. The next step would be to update this new URL in the support center settings in the Support center domain field.

Secured (SSL) Setup: If your site is secured, and you've chosen the iframe solution, then we need to set up a secured connection in Bold360 AI and exchange certificates. For more information see this article.

Setting up a parent page for an iframe solution

In order to set up the support center as a part of your existing page (in iframe) you need to add the following code there:

<!-- embedded support center code start -->
<div id="nanorep-support-center"></div>
<script>
! function(t, e, o, n, c) {
    var s = this.nanorep = this.nanorep || {};
    s = s[e] = s[e] || {}, s.host = c, s.path = n, s.domain = t, s.protocol = "https:" === location.protocol ? "https:" : "http:", s.on = s.on || function() { s._calls = s._calls || [], s._calls.push([].slice.call(arguments)) };
    var a = document.createElement("script");
    a.async = a.defer = !0, a.setAttribute("src", s.protocol + "//" + c + n + o), document.getElementsByTagName("head")[0].appendChild(a)
}("your_support_center_domain", "supportCenter", "support-center.js", "/web/", "your_account_name.nanorep.co");
</script>
<!-- embedded support center code end -->

(Replace the highlighted sections with your actual account settings)

If you need to set some predefined context values, place them after the support center domain as query parameters, for example:

("your_support_center_domain/?Brand=Apple","supportCenter","support-center.js","/web/", "your_account_name.nanorep.co")

Advanced configuration

Embed code will insert embedded support center with default settings, but there's a way to set override the settings inside the "Master Page" template like this:

<script src="//<%Site.nanoRepDomain%>/web/support-center.js"></script>
<script>
nanorep.supportCenter.init({  
            noDirectDomainAccess: false, // use "true" to disable direct access to support center     
            supportCenterDomain: '<%Site.nanoRepDomain%>', // Support center subdomain     
            mainSiteURL: '<%Site.MainSiteUrl%>', // Parent page URL     urlPrefix: '#/path' //prefix path for URL hash   });
</script>

How to implement a child support center

If you have several support centers pulling the data from different knowledge bases but have the same support center domain, it is also possible to load them by specifying a unique URL path after the URL Prefix part in the parent page URL. For example, let's say you have 2 knowledge bases:

1) English KB - main support center ( support.your_company.com )

2) Dutch KB - child support center ( support.your_company.com/ dutch )

In order to load the main support center you can just access the parent page URL located and it will be loaded automatically as a default option.

In order to load Dutch child Support center you should specify its unique URL part after URL prefix: http://your_company.com/#path/dutch

where:

your_company.com - parent page URL

#path - default URL prefix

dutch - unique URL part of child support center

Slim Support Center widget customization

For instructions on how to implement these examples see How to customize the Slim widget? guide.

Customize widget width and height

Since this widget type is basically just a search bar from regular embedded widget its width can be limited by changing container div style directly:

#nanorep-embedded-widget {
width: 500px; 
}

As for the widget height, it is highly dependent on font settings in search bar. If you decide to increase/decrease height of search bar, you need to make sure that all the settings below are also increased/decreased:

.widget-embedded--support-center .query-field__placeholder,
.query-field__input {
    padding: 25px 15px 25px;
    /* padding bottom, padding right, padding top */
    line-height: 20px;
    /* should not be less than font-size value */
    font-size: 25px;
    /* should not be bigger than height value height: 90px; }

Answer title font and colors

.answer-header__title-text-node {
    font-family: 'Josefin Sans', sans-serif;
    /* If you use custom font, make sure to link it at parent page */
    font-size: 20px;
    font-weight: bold;
    /* Bold text */
    text-decoration: underline;
    /* Underlined text */
    font-style: italic;
    /* Italic text */
    color: green;
    /* Font color. You can also use color set in HEX or RGB format */
}

Answer text font and colors

Note that we do not recommend overriding any settings except font-family and text color. All other text style changes are to be done in the Article editor.

.answer-body__html { font-family: 'Josefin Sans', sans-serif; }

Customize links

[nr-normalize] a { text-decoration: none ; cursor: pointer ; color: red ; }

How to change the sorting of support center articles

By default, articles in your Support Center are sorted by popularity. You can also sort articles by alphabetical order or last updated time.

  1. In the AI Console, go to Touchopints > Support Center.
  2. Select Page Templates and choose the support center you want to change.
  3. On the Label Page tab, find the following entry, which drives article sorting:
    <nR:Repeater type='article' sort='faq' labelId='<%Label.ID%>'>
  4. Change the value of the sort attribute as follows:
    • Faq - articles are sorted by popularity (number of hits). This is the default value.
    • TitleAsc - articles are sorted by alphabetical order from A to Z.
    • TitleDesc - articles are sorted by reverse alphabetical order from Z to A.
    • LastUpdated - articles are sorted by last updated time, where the most recently modified one is listed first.
  5. Save your support center template.

What are my options for implementing a support center?

You can implement a support center as a standalone or you can use iframe.

What are the benefits of an iframe implementation?

By using an iframe, you can deliver your support center as an integral part of your website while leaving your site's functionality in place:

  • You get full control of the page wrap including the site's header and footer, cart support, logged in users and any other cross-website functionality.
  • Digital DX AI offers a smart iframe that will
    • Automatically change its size depending on the amount of content inside the iframe
    • Change the hosting parent page URL to reflect the changes in the iframe. For example, if the user navigates to a specific article, the article URL will be reflected in the parent URL. This behavior can be disabled if not desired.
  • The iframe solution fully supports search engine optimization (SEO) including redirect to the parent hosting page if the iframe is opened directly from Google.

What is the standalone solution like?

You can deliver your support center as a standalone, external page with a subdomain address. A standalone support center can't present your site's header and footer and doesn't include other site-specific functionality from your main page either, like user-secured login, cart management or site search.

The benefits of using a standalone support page:

  • Changes in your site are minimal, you just have to add a link to your new page.
  • The page is easy to implement.
  • There is no risk of conflicting code between your site's JavaScript and the Digital DX AI's JavaScript.

How to set up a protected support center

To increase the level of security, Bold360 AI provides cross-origin resource sharing (CORS) policies on its APIs.

CORS enables you to restrict your widgets and support centers to a single URL domain or a list of domains. If a malicious intruder copies your HTML code snippet and adds it to any other site, the widget or support center on that site will not work.

To set up a secure support center, do the following:

  1. On the Touchpoints > Support Center > Settings page, copy your support center URL from the Main Site URL field.
  2. Go to the Admin Center > Preferences page.
  3. In the Allowed Origins field, paste the support center URL you just copied from the Main Site URL field.
  4. Add the URL of your main customer website to the list.
  5. Save your changes.

How can SEO be improved for support centers?

Bold360 AI offers multiple designs for SEO-friendly support pages which include both free search capability and dynamic navigational FAQ hierarchy.

Google can crawl iframes on a website and is indexing the URLs of the iframe itself, not the parent URL. The same applies to the links inside the iframe. The iframe URL and the URL of the links within the iframe point to a subdomain. You can read more about the effect of using a sub-domain on your Google ranking here.

The challenge when setting up a support center is to avoid Google opening the iframe page instead of the support center page.

Bold360 AI solution

In the Bold360 AI solution for a support center, the iframe URL address is a CNAME provided by the customer, for example, support.acmeCorp.com. This iframe is hosted in a page that is defined by the customer as part of the site and will probably have a URL such as acmeCorp.com/support.

In the iframe, navigation links and articles are presented. These include a link to the support page, home page, label pages and article pages. All these links have the following structure: <CNAME URL>/Label1/Label2/Article-ID/Article-Title. For example, support.acmeCorp.com/General/How-to-Contact/12345/our-branches-opening-hours.html.

Search engine flow

The search engines crawl the hosting page and into the iframe. In the iframe, it crawls deeper and collects the information for all the articles and label pages. All of this content is part of the sub-domain assets. The score and links are, therefore, related to the subdomain.

When you want to find an article using Google, you search for the text that is in one of the articles. Google finds the article page and redirects to a page, which has the subdomain URL.

When the user selects an article, the subdomain URL opens. Bold360 AI redirects all these links to the parent page while maintaining the direct path to the relevant article. Eventually, the parent hosting page has a URL like acmeCorp.com/support?path=/General/How-to-Contact/12345/our-branches-opening-hours.html while the iframe inside maintains the URL received by Google support.acmeCorp.com/General/How-to-Contact/12345/our-branches-opening-hours.html.

How does the subdomain structure affect your SEO ranking compared to a subfolder structure?

In late October, 2012, Matt Cutts of Google's Webspam team clarified the subdomain/subdirectory situation as far as Google was concerned. He said "The historical reasons why you might have wanted to go for a subdomain don't really apply as much, and that leaves you with, okay both are on the same domain, overall, and so it's really a question of which one is easier for you." Therefore, your entire site benefits from the effects of search engine optimization.

How can I track subdomain analytics and include it with my main site analytics?

See the multidomain tracking guide for Google Analytics here.

How do I promote my internal support center?

A few tips on how to promote your internal support center and bot.

In case you recently set up an internal support center or a bot for your agents, here are a few tips on how you could promote their use among your agents:

  • Naming competition: ask agents to name the support center or bot and announce the winner.
  • Most questions competition: highlight agents or teams who ask the most questions.
  • Easter eggs: hide articles about off-topic questions in the knowledge base and ask agents to find them, announce the winner.
  • Ask agents for feedback: make sure your team has the option to provide feedback about the support center or the bot. Also ensure that their imporvements are made based on their feedback.

How to add context to the support center

Context is saved in the browser cookie, so the first time the user navigates to the support center, the page must be provided with context.

Make sure you have set up contexts in your account.
  1. In the AI Console, go to Touchpoints > Support Center.
  2. Choose a support center.
  3. On the Settings tab, add the context to the support center URL.

    Add the context name and value as defined in the Bold360 AI platform. Use an equals (=) sign between the two.

    Example:

    For example, if the support center URL is mydomain.com/mysupportcenter.html, change it as follows:

    • If you set up a context called "country", then the Support Center URL looks as follows: mydomain.com/mysupportcenter.html?country=us
    • If you want to combine two or more contexts in the same URL, use the "&" separator: mydomain.com/mysupportcenter.html?ver=3&country=us
  4. Create article variations that use different context values. For more information see How to define an article's context?.

What are the Support Center Analytics?

The Support Center is also trackable with the Bold360 AI Analytics functionality.

From within your Bold360 AI account, go to Analytics > Custom Charts > Support Center.

The following parameters are available for analytics tracking:

  • Visits
  • Page Loads
  • Organic Visits (visits generated via search engines rather than paid advertisements)
  • Total resolved by Support Center

What is the Support Center referrer?

The SC Referrer is a field in reporting. This field enables you to view which page in SC the user opened before reaching an article (Whether it was from FAQ in homepage, Label page or link from another article).

In the report the Values will be displayed as a URL (of the SC page). If the referrer is not in the support center (for example, an external link), the value will be displayed as empty.

To create this report, go to Analytics > Reports. Create a report according to your needs. Below is an example of how to create a report:

The results will be displayed as follows:


How do I upload content into the knowledge base?

Content can be extracted from your current content management system and uploaded into Bold360 AI using the attached Excel template.

The fields to be captured per article:

Column 1- article title

Column 2 - article body. The article body is preferably to be provided with HTML formatting.

Column 3 - article labels. For general information about labels click here.

Column 4 and on - article contexts and context values. Each column should represent a single context type. Within this column for each article capture the relevant context values. For general information on Context see What are Bold360 AI contextual answers?

In order to upload the content please contact your Success Manager or send the file to support.

What are the best practices for content creation?

Follow these guidelines about how to phrase intents and create article content to make the most of Bold360 AI's Natural Language Processing (NLP) algorithm.

When phrasing intents, keep in mind the following:

  • It?s best to phrase the intent as a question in natural language.
  • Bold360 AI?s NLP algorithm works best with questions that are three to eight words long.
  • Boring is good, so be consistent in how you formulate intents.
  • Use a single statement as intent, for example, What do I do if I can't log in? and avoid phrasings like What do I do if I cannot login, my password is rejected??. Add alternatives ways of asking about the same issue as phrasings to the article.
  • Avoid using acronyms and descriptions in titles like Is PayPal (PP) accepted?. Add synonyms for acronyms. Create phrasings for acronyms that are composed of three or more words.
  • Turn off autocomplete for articles like error messages and negative articles like I want to cancel by unselecting Search and Auto-Complete on the Visibility tab for the article.
    Note: You can also turn off autocomplete for phrasings if necessary.

When composing the answers, remember to:

  • Keep them short.
  • Engage readers using a variety of media: pictures, videos, gifs, tables and so on.
  • Use formatting options like bold, italics, underlining, bullets and so on.
  • Display content on multiple pages if needed; it is a good way to break up look articles to make them more digestible to the reader.
  • Take into consideration that the content may be consumed on a variety of interfaces: mobile, floating widget, support page and so on.
  • Create focused questions and answers and consider what the user really needs, for example, full user guide or steps on how to use the image button.
  • Use channeling instead of including phone numbers directly in the article answer body, especially when deflection from the phone channel is measured?.
  • Periodically revise your FAQs to keep suggested responses relevant.
  • Use conversational flows and article trees when you need more information from customer to provide an answer.
  • Use hyperlinks and ensure that the bot follows, or the link opens in a new tab:
    • Use when answering the question is best done by pointing to a page with a lot of information for user to explore.
    • Use when information is constantly changing and updates are made to that page rather than having to update the article.

How to define a content management workflow

With the content management workflow you can define editors, reviewers, publishers and approvers for knowledge base articles.

To introduce a review process, you can set up a collaboration-enabled workflow for a content management team, including editors, reviewers, publishers, and approvers. In a content management workflow, articles can be assigned to approvers, users can follow the approval process, and articles can be published in a controlled way. To create a review process, select users to manage the workflow:

Important: You must have the Enable approval workflow feature enabled for the knowledge base where you want to introduce a content management workflow. Contact your Customer Success manager to enable it for you.
  1. In the Bold360 AI platform, go to Admin Center > Users.
  2. Select a user who you want to include in the workflow.
  3. Open the Advanced Credentials list and under Knowledge-Base, select the role of the user in the review process:
    • Edit - the user can author the article and send it for approval
    • Approve - the user can either approve the article by clicking Send for Publishing or reject the article
    • Publish - the user can either reject or publish the article
    Important: A user must have modify permission for Knowledge-Base to be included in the workflow.
  4. Click Back to save the changes and set up the remaining workflow members with the proper credentials.
When rejecting an article, you should provide a reason on the Rejections tab of the Article Editor to help the editor revise content.

What workflow statuses are available?

On the Knowledge > Articles page, articles that are in the content management workflow are marked by the following system labels:
  • Waiting for approval - The content editor has sent the article for approval
  • Waiting for publishing - The approver has approved and sent the article for publishing
  • Rejected - Either the approver or the publisher has rejected the article, in which case the editor should revise the article's content.

All articles in the workflow are in draft mode until published.

How to define an approval process

To set up an environment where certain users can only author articles, but cannot publish those, do the following:
  1. Assign the Account Watcher role to the person who writes articles.
  2. Assign the Account Manager role to the reviewer, who approves and publishes articles.
  3. On the Admin Center > KB Setup > Notifications page, do the following:
    1. Enable Activate Email Notifications
    2. Enable Send notifications to assignee
    3. Select the reviewer in the Email recipient for assignee changes field.
Whenever an article is assigned to a reviewer, Bold360 AI automatically sends an email to the relevant user (Account Manager).

How to limit a content manager's access to a subset of a knowledge base

Create permission groups, add users to them and specify them for selected articles to control the article's visibility.

When you have all your content in a single knowledge base and you want each of your content managers to focus on a particular area, you can create permission groups. These permission groups allow your users to view only a selected set of articles that you define manually.

Important: To set this up, the Enable users groups permissions option must be enabled for your account. Contact your Customer Success Manager for more information.
  1. In the AI Console, go to Admin Center > Users and click on the name of a user.
  2. Click the Permission Groups icon at the top of the page:

  3. Click Create group.
  4. Name your group and make sure you select the right Knowledge Base where you want to associate your articles with permission groups.
  5. Click Save.
  6. Click the name of a user to edit it and click Permission Groups at the top of the page.
  7. Make the user the member of a permission group by selecting a group. On the user details page, you see the associated permission groups in the Member of field.
  8. Once you are done, you must also assign articles to permission groups:
    1. Go to Knowledge > Articles and select an article.
    2. On the Visibility tab, select the Permission Groups that you want to assign to the article.
    3. Publish and close the article.

Your Bold360 AI users are now restricted to view only the articles defined by the permission groups.

How to copy support center settings between knowledge bases

When you already have a support center, you can copy the settings of the support center from one knowledge base to another so that you don't have to start from scratch.

  1. In the AI Console, go to Touchpoints > Support Center.
  2. At the top of the page, choose Copy Support Center.

    Result: The wizard is displayed.

  3. In the Select Target Support Center step, select the knowledge base where you want to replicate the existing support center in the Target KB field.
  4. In the Target Support Center, choose the support center in the target knowledge base into which you want to copy the settings.
  5. Select Next.
  6. In the Select Templates to be copied step, choose the support center templates that you want to copy to the new knowledge base.

    These are the templates that you can find on the Page Templates tab of the Touchpoints > Support Center page.

  7. Select Next.
  8. In the Select Support Center Settings to be copied step, choose the settings you want to copy to the new support center.
  9. Select Next.
  10. In the Summary and Confirmation step, choose Apply and when prompted, confirm that you want to overwrite the existing support center configuration in the target knowledge base.
    Important: Once you overwrite a support center configuration, you can't revert it to an earlier version.

Bold360 AI SEO and knowledge-base export API

The SEO and knowledge base export API "pushes" question and answer information from the Bold360 AI knowledge base into the page source.

This way the information becomes visible to search engine bots, resulting in more traffic being driven to the website, and users being able to ask questions via Google, or any other search engine, and being directed straight to your website without using the Bold360 AI widget.

API specifications

The base URL format of the API is the following:

https://my.Nanorep.com/common/api/kbExport.xml?param=value&param=value

The following is a sample script with all relevant data provided:

https://my.Nanorep.com/common/api/kbExport.xml?account=ACCOUNTNAME&apiKey=APIKEY&kb=English&plainText=true

General API parameters

Parameter

Usage

Example

username Your Bold360 AI login name. You must have Knowledge Base access privileges to use the API (required) username=john@company
pw

Your Bold360 AI password (required)

pw=1234567

kb

The Knowledge Base to export from (required)

kb=English

maxItems

The number of items to retrieve

maxItems=50

skip

The place in the list from which to start retrieving. For example, use skip=10 to skip the first 10 list items and retrieve the rest.

skip=10

format

xml/json (default:xml)

format=json

Filter parameters

Parameter

Usage

Example

byPopularity

Set this parameter to true (recommended), to display the most popular articles first in the search results.

Note: Offline articles will not be included if the parameter is set to true.

byPopularity=true

days

Displays the most popular answers given by the Bold360 AI widget during the last x days. This parameter does only have an effect if byPopularity is set to true.

days=30

labelId

Displays answers from a specific label only. You can find this information in Bold360 AI when editing a label.

labelId=ABC123

textFilter

Displays answers that contain a certain word or phrase.Important: The value of this parameter must be base64 encoded. As it is used in a URL, replace "=" with "_", and "+" with "|". For example, "hello" must be defined as "aGVsbG8_"

textFilter=aGVsbG8_

context

Displays answers that have the specified context in category:value format.

You may use more than one category:value pairs separated by commas. For example, Manufacturer:Sony, Device:iPhone **

context=TWFudWZ

hY3R1cmVy Ol Nvbn

ksRGV2aWNlOm

lQaG9uZQ__

modifiedSince

Format: yyyy-MM-ddTHH:mm.

If used, only articles modified since this date are included.

modifiedSince=2013-11-01T17:00

includeHiddenFaqItems

Includes hidden FAQ articles, which are marked by the "hide this answer from the dynamic FAQ widget" option.

includeHiddenFaqItems=tru

API URL Examples

Full export with stripped body
https://my.Nanorep.com/common/api/kbExport.xml?account=ACCOUNTNAME&apiKey=APIKEY&kb=English&byPopularity=true&plainText=true
Articles with the word 'iPhone' in them
https://my.Nanorep.com/common/api/kbExport.xml?account=ACCOUNTNAME&apiKey=APIKEY&kb=English&byPopularity=true&maxItems=20&textFilter=aVBob25l

XML Response example

<?xml version="1.0" encoding="utf-8"?>
<NanorepExport account='Nanorep' kb='633763325358580000'>
        <article id='633915887117827492' isOffline='False'>
                <title>
                        <![CDATA[How much does it cost?]]>
                </title>
                <body>
                        <![CDATA[<p>For detailed packages &amp; pricing information, <a href="http://www.Nanorep.com/index.php?page=prices" target="_blank">click here</a></p>]]></body>
        </article>
        <article id='633915887117827588' isOffline='False'>
                <contextInfo>
                        <context id='Page Type' value='General' />
                </contextInfo>
                <title>
                        <![CDATA[What is Nanorep?]]>
                </title>
                <body>
                        <![CDATA[<p>Bold360 ai is a self-service customer support &amp; service solution. It is designed to improve customer engagement and prevent website or cart abandonment by delivering instant, accurate answers to customers&#39; questions with no waiting queues what so ever.<br />&<br /></p>]]></body>
        </article>
        <article id='141286' isOffline='False'>
                <title>
                        <![CDATA[What happens when I exceed the number of questions in a given month?]]>
                </title>
                <body>
                        <![CDATA[<p>We understand that due to seasonal changes, promotions or campaigns, there can be spikes in the Q&amp;A&rsquo;s traffic.</p><p>In these cases, it won&#39;t be necessary to upgrade to a premium package.</p><p>However, in cases where there is a consistent increase in Bold360 ai&#39;s usage for more than a month, a notification on the outgrown support package&will be sent to the customer and an upgrade plan will be offered.</p>]]></body>
        </article>
                <meta skipped='0' exported='3' totalArticles='3'/>
</NanorepExport>

How can I find answers with links/images in my knowledge-base?

You can customize the filter in the Article List so that the search returns articles with links or images.

To find answers that contain hyperlinks in your knowledge-base, do the following:
  1. Go to Knowledge > All Articles.
  2. Click the Filter button next to the search field.
  3. On the Options tab, select Search titles and answer body.
  4. Click Apply.
  5. Search for an article:
    • To search for links, type "href" in the search field and press Enter.
    • To search for images, type "img " in the search field and press Enter.

The biggest hurdle with artificial intelligence is building a repository of answers for the AI to draw from. But with Bold360, building that knowledge base is faster and easier than ever. Existing FAQs can easily be imported, and our clean and simple interface makes it easy for any business user to create new articles from scratch. With Bold360, building your knowledge base is quick and painless, so AI can start helping right away.

Can you support multiple websites from one knowledge-base?

Yes, Bold360 AI supports multiple websites from a single knowledge-base, but you must set up a separate widget for all your websites.

However, it is recommended to separate the knowledge-bases in case of completely different content.

For more information, see How do I add another knowledge base? and What widgets do you provide and how can I add them to my website?

How to limit a content manager's access to knowledge bases

Create user groups that provide access to selected knowledge bases and assign them to users.

When you have multiple knowledge bases (KBs) and content managers, you may want to restrict users to have access to those KBs that are relevant for their daily work. For example, your organization may support multiple languages or products, where a each language or product has its own KB and is curated by a specific content manager. In this case, you can create a user group and assign a KB to that group. Then you can add users to the same group.

  1. In the AI Console, go to Admin Center > Users.
  2. Click the Groups icon.
  3. Click the Create Group link.
  4. Give the group a name, and on the Advanced tab, select the knowledge bases that you want group members to have access to.
  5. Click Save.
  6. Add users to the group:
    1. On the Users page, click on the name of a user.
    2. Click the Groups icon.
    3. Select the groups that you want the user to be a member of.
    4. Click Apply.

Your users are now limited to access the selected knowledge bases only.

What are the best practices for managing a knowledge base?

Follow these best practices and guidelines to build a knowledge base that knowledge base managers can work with efficiently and where customers can find answers to their questions with ease.

Bold360 AI offers various options you can use to optimize your content:

  • Links to connect related articles to each other for easier findability
  • Labels to organize content in your knowledge base and group articles in the support center
  • Phrasings to make sure readers find relevant content when they use wordings other than the intent
  • Synonyms to cover similar words thus minimize the number of phrasings
  • Context to keep internal and external content separate, offer different escalation channel options based on customer type and so on
  • Article variations to create alternative answers to the same question for different audiences
  • User groups and permissions to provide access to a specific knowledge base or a subset of one
  • Content management workflow to ensure the quality of the content
  • Search Optimizer to help knowledge base managers follow up on customer voices and improve content and phrasings

What are the best practices for labels?

You can increase productivity using labels because you can save time looking for content. Organizing the content helps to keep the content grouped together per owner which in turn also creates easy accountability when filtering the reporting data by label.

When creating and applying labels to articles, keep in mind the following:

  • It?s best to use labels based on how you group content and assign content-related tasks internally.
  • It?s good practice to use a label for conversational articles and articles that include videos to easily find them.
  • To have control over what articles are presented in the FAQ, use a specific label for FAQ articles.
  • To create channeling policies and reports for certain articles, use the same label for them.

What are the best practices for phrasings?

When creating phrasings, remember the following:

  • Search works best with phrasings that are
    • to the point
    • three to eight words long
    • simple sentences
    • formulated as a question
  • Use phrasings that are unique to a single article to ensure the accuracy of search results.
  • Increase variety in the wording of phrasings? as the search algorithm looks at the combinations of the different terms used across all phrases in the article.?
  • Use negative phrasings when you have articles with almost identical titles and substantially different content.
  • Use synonyms instead of phrasings to cover words that are close in meaning.
  • Add up to 10 phrasings for an article.
  • Use the full term in the title and add acronyms as phrasings.
  • Use entities to turn your data into smart chatbot conversations.
  • When leveraging voices to create new phrasings, clean up the voice, for example, remove typos, redundant words and shorten long sentences to improve the appearance of phrasings in autocomplete.?

What are the best practices for synonyms?

Keeping a large list of synonyms helps improve the Bold360 AI Natural Language Processing (NLP) algorithm.?

When creating synonyms, remember the following:

  • Use narrow synonyms. If a synonym is too broad, it negatively impacts search results and it is difficult to identify the root cause.?
  • There?s no need to capture typos as the NLP algorithm automatically corrects them.
  • Create synonym groups for common abbreviations of a word.
  • Add synonyms that are specific to your business.
  • Use synonyms for acronyms and word combinations of two or more words.

How do I specify Knowledge Base dynamically using an API?

Note: Use this reference for Arcade and Modern widgets.
The following API allows you to specify the name or ID of the Knowledge Base domain to be used.
Important: If you have already customized a domain for the current URL, the API cannot be used.
_nRepData["kb"] = "hebrew";
// or 
_nRepData["kb"] = "496477";

To use the API for a specific widget add the following:

_nRepData["embed"] =
{ 
  "kb": "hebrew"
}

Knowledge Base Analytics: Most Escalated, Most Popular, Most Liked/Disliked

To view Knowledge Base analytics, go to Analytics > Knowledge Base.

Most Popular
The most commonly asked questions.
Most Escalated
The questions that most often lead to escalations.
Most Liked
Questions that received the most "like" responses.
Most Disliked
Questions that received the most "dislike" responses.
Tip:
  • Click any question to see a graphic visualization of a question's trend over time.
  • Use the search box (by keyword) to filter by any specific question. The filter applies to all four charts.
  • When applicable, Context is shown in parentheses.

How can I add a new article to my Knowledge Base?

Use the Article Editor to create new content in your knowledge base.

  1. Open the Article Editor.
    • In the left navigation menu, click Knowledge > Add Article..
    • In the left navigation menu, click Knowledge > All articles, then click New Answer.
  2. Type a question in the Add title field and type the answer in the text editor.
  3. Save your article:
    • Click Save if you have not finished editing the article and you want to save an offline draft.
    • Click Publish to publish the new article in the Knowledge Base.
    • Click Publish & close to publish the new article and return to the All Articles page.

How to add notes to your article

You can add notes to an article if you want to add additional or internal information about the article itself. This may be useful when other content editors update your article.
Important: You must have the Enable article notes tab account setting enabled to see the Notes tab. Contact your Customer Success Manager to enable it for you.
  1. Open the article that you want to add notes to.

    You can only add notes to an existing article.

  2. In the Article Editor, go to the Notes tab.
  3. Type your remarks and click Publish when you are done.

    You can only add notes to the article itself and not to the individual article revisions.

Knowledge Base Manager - Role Overview

The knowledge base manager role provides access to the AI Console in a way that the user can perform tasks related to maintaining and optimizing the knowledge base.

The Knowledge Base Manager holds an extremely important role, as they are responsible for ensuring that the Knowledge Base is kept up-to-date and optimized with the most effective and relevant information, which guarantees the smooth running of the Digital DX AI system and effective customer service.

The Knowledge Base Manager performs their tasks the Knowledge and Search Optimizer pages. These tasks comprise searching for unanswered questions (articles) and ensuring that they are given an accurate and appropriate answer, writing new answers, editing existing answers, merging existing answers, adding phrasings to articles and adding synonyms.

Managing the Knowledge Base comprises several tasks which should be performed on a regular basis. You are responsible for general "housekeeping" of Search Optimizer, by maintaining it daily. This involves deleting unnecessary questions, merging questions asked in a different manner to previous questions with existing answers that also fit them, and adding new answers.

The tasks of the Knowledge Base Manager are all intended to optimize the Knowledge Base:

  • Optimize the Knowledge Base
  • Finding questions that require answers (Filtering, Answering, Merging and Phrasings)
  • Adding to and enhancing the Knowledge Base (Writing new answers, publishing existing drafts)

Optimize the Knowledge Base

This is the main task of your daily activities as the Knowledge Base Manager. You must use Search Optimizer to ensure that the Knowledge Base reflects the most accurate and up-to-date information, in order to best serve your customers. You can access Search Optimizer under Voices > Search Optimizer. See What is Search Optimizer? and How to use Search Optimizer to learn more about it.

Write and publish answers

You can write a brand new answer or you can enhance an existing one, and then publish it to the Knowledge Base.

Write New Question/Answer
  1. In the AI Console, go to Knowledge > Add Article.
  2. Write the question, using the most appropriate language for the subject to make the question as accessible as possible to as many customers as possible.
  3. Write the answer.
  4. Click Publish.
Publish draft answers to the Knowledge Base
  1. In the AI Console, go to Knowledge > Articles .
  2. Click the question that you want to publish.
  3. In the Article Editor, click Publish.
Manage phrasings
When you merge answers, Phrasings (or keywords) are added to the Knowledge Base. Phrasings group questions together that are related to similar content, but which are phrased slightly differently. The Knowledge Base can recognize this and provide the correct answer automatically. Phrasings can also be added manually, although Search Optimizer 2.0 adds the phrasings from merged questions automatically.

How do I import my knowledge base from ServiceNow?

Bold360 AI uses ServiceNow's API to import knowledge base articles.

Remember: Your account must have the Enable ServiceNow article import setting enabled for each knowledge base where you want to import articles. Contact your Customer Success Manager for more information.
Note: If you are looking for information about importing articles in a batch using Bold360 AI's API, see this article.

Integrate ServiceNow with Bold360 AI

Before you can import knowledge base articles, you must first integrate ServiceNow with Bold360 AI as follows:

  1. In the AI Console, go to Admin Center > KB Setup and choose the Integration tab.
  2. From the Third-party service drop-down list, select ServiceNow. From your ServiceNow account copy the following details:
    • Copy the instance URL from your browser. Look for https://< instanceURL >.servicenow.com
    • Enter your ServiceNow user credentials.
    • From the Application Registry page in ServiceNow, create a new OAuth API endpoint for external clients that you want to use to set up a connection between ServiceNow and Bold360 AI. Then copy Client ID and Client secret to the Bold360 AI platform.

    The following is an example of the ServiceNow integration parameters:

  3. Choose Connect.

ServiceNow is now integrated with Bold360 AI.

Import Knowledge Base articles

To import knowledge base articles, do the following:

  1. Select the Bold360 AI knowledge base where you want to import articles. Note: You can only import articles to a single Bold360 AI knowledge base. Your account must have the Enable ServiceNow article import setting enabled for this knowledge base.
  2. Go to Admin Center > KB Setup and choose the Integration tab.
  3. Select the ServiceNow knowledge bases that you want to import. These are pre-populated during the integration process.
  4. Optionally, select article labels to filter the articles in the selected knowledge bases.
  5. Choose Fetch Articles and wait for the list of articles to load. Depending on the number of articles in the selected knowledge bases, this process may take a while.

    Articles that you can update in your Bold360 AI knowledge base are listed. Only those articles are listed that have changed in ServiceNow since your last knowledge base import.

  6. Select the articles that you want to import. You can click the arrow next to the title of an article to view its content.

    Articles are color-coded in the list to reflect changes in the ServiceNow knowledge base:

    • Green: new article. A new Bold360 AI article will be created.
    • Yellow: updated article. The article in the ServiceNow knowledge base has been changed. If you have edited the article in Bold360 AI then those changes will be overwritten when importing from ServiceNow.
    • Red: deleted article. The article will be removed from the Bold360 AI knowledge base.
  7. Choose Import selected articles at the top of the list.

The ServiceNow articles are now imported into your Bold360 AI knowledge base.

What happens after articles are imported?

The imported articles receive a ServiceNow label and their expiry dates are also synchronized. When an article expires, it is removed from the live site. To see expired articles in Bold360 AI, go to Knowledge > Expired.

Finally, the revision list is updated to reflect the changes in the number of imported articles.

How do I export my knowledge base to Excel?

You can export the content of your knowledge base to an Excel file that contains details like the ID, the creation date, the intent and the answer of the article and the labels assigned to it.

To export your knowledge base to an Excel spreadsheet, please follow these steps:

  1. Log in to your account and navigate to Knowledge > All Articles.
  2. Choose (More).
  3. Select Export Knowledge Base and then Office 2007+ in the drop-down that appears.

    The export file contains the original IDs of the articles. This way, links to other articles work the same way as in the original knowledge base.

    Note: You can export a maximum of 5000 articles with this option.

For more complex exports, you must use our Export knowledge base API.

What is the Knowledge Base Usage Chart

The Knowledge Base Usage chart shows you how many questions were added or deleted in the knowledge base over a period of time.

To view the chart, in Bold360 AI, go to Analytics > Knowledge Base.

The Knowledge Base Usage Chart provides you with a brief overview of how customer support has been operating over a specified period of time. This data enables you to ascertain how your customer team is functioning.

The legend for the Knowledge Base Usage Chart is as follows:

  • The green line indicates the number of answers that were added to the Knowledge Base.
  • The blue line indicates the number of answers deleted.

See also: Knowledge Base size Chart

What is the Knowledge Base Size Chart?

To view the Knowledge Base Size Chart, in Bold360 AI, go to Analytics > Knowledge Base.

The Knowledge Base Size Chart shows you the volume of questions (including phrasings) and answers in the knowledge base. It also displays the volume of tickets in the inbox and the volume of items in the publishing center under the publish answers tab on a time scale.

All these metrics give you the progress of the system's learning process and knowledge base growth.

Notes:

  • The yellow line shows the number of questions and phrasings
  • The red line shown the number of answers.
  • The blue line shows the number of items in the publishing center under the publish answers tab
  • The green line shows the number of tickets in the inbox

See also: What is the Knowledge Base Usage Chart

How do I start creating my Knowledge Base ?

You can import an existing FAQ from a website, create articles manually or upload articles in a batch to fill your knowledge base with content.

You can populate the knowledge base software in 3 different ways.

Importing FAQ from a Website

You can create an instant Knowledge Base by importing an existing FAQ. The Import FAQ tool enables you to import an entire existing FAQ page into the Bold360 AI knowledge base.

  1. Go to Knowledge > Import FAQ.

    Result: The Import FAQ tab is displayed.

  2. In the Enter your FAQ URL field, enter the address (URL) of the FAQ that you want to import.
  3. In the Context section, specify the page type. This can be an internal knowledge base page for your support staff to use, or a public FAQ page.
  4. In the Product ID field, enter the name of the product for this Knowledge Base.You may have more than one product, each of which requires its own Knowledge Base.

Adding new answers manually

There is no need to put a lot of effort in collecting questions for the initial knowledge base, since Bold360 AI will learn new questions from your customer inquiries and new answers from your representatives.

How can I add a new article to my Knowledge Base?

Upload via an Excel sheet

We can upload articles via an Excel file directly into your knowledge base in Bold360 AI.

The Excel file must be Excel 2003-2007 Format.

How to import multiple articles in a batch with the API

For more information regarding uploading the articles via an Excel file, contact your Customer Success Manager, or contact our support.

How do I add another knowledge base?

Learn what to consider before adding a new knowledge base.

When you use a single Knowledge Base for multiple websites

Bold360 AI can support multiple websites from a single knowledge base.

If you have similar content with some modifications based on different categories or websites, you shouldn't create another knowledge base but rather use just one knowledge base with contextual answers. Web visitors can ask the same question, and get different answers based on any differentiating information you have on a customer such as the product or page they're currently looking at. See What are Bold360 AI contextual answers? to learn more.

When you have multiple Knowledge Bases

It is recommended to have separate knowledge bases for completely different content, for example, for internal and external use, completely different websites, a dedicated knowledge base per language and so on.

If you'd like to have an additional knowledge base, contact your Customer Success Manger.

Knowledge Base Labels

Knowledge base labels are categories you can use to organize content in the knowledge base or in the support center.

Knowledge Base Labels are categories that you assigned to articles to make organization of your Support Center easier and more efficient. You can arrange labels in a hierarchy to define how your Support Center is displayed. Labels also make it easier for you to search for specific topics and related answers.

Tip: Looking for information regarding Ticket Labels? See How do I assign a label to a Ticket?

Creating a Label

  1. Go to Knowledge > All Articles and click the Label icon.
  2. Click Create Label.
  3. The Content Labels window is displayed, where you can add, remove, edit and sort labels.
  4. Under Label Name, give your Knowledge Base Label an appropriate name and then select a color for your label.
  5. Save your changes. The label is displayed in the list of Knowledge Base Labels, and is displayed next to any article to which it is assigned.

Categorizing Knowledge Base Answers Using Labels

Categorizing Knowledge Base answers mean that you can easily organize your support page into groups of related topics for easy browsing of your customers.

  1. Go to Knowledge > All Articles, and select the article(s) to be categorized.
  2. Click the Label icon.
  3. Choose the relevant label to assign it to the article.

Categorizing Multiple Labels at Once

  1. On the Knowledge > Articles page, select the check box of an article (or articles).
  2. Click the Label icon () in the toolbar.
  3. Click Select multiple....
  4. Select labels and click Apply.

How to use the knowledge base launchpad

Use the knowledge base launchpad to create a knowledge base quickly and easily.

Note: The knowledge base launchpad is not available by default, it must be enabled for the knowledge base on account level. For more information contact your Customer Success Manager.
  1. In the AI Console, go to Knowledge > Launchpad.
  2. Select one of the predefined packages depending on your business needs.

    Result: Once you select a package, a list of articles appears.

  3. Review and modify the articles as required.

    You have the following options for each article:

    Table 1. Options for knowledge base launchpad articles
    Option Description
    Live This option controls whether the article is imported to the knowledge base in online or offline state. All articles are set offline by default. If you set an article live, it will be immediately available on touchpoints once you launch the article.
    Article Title Displays the article title and the number of phrasings available for the article. Select the number to see the phrasings.
    Content The body of the article. You can add the content you want to present as a response to the question that is in the Article Title.
    (Delete) This option enables you to delete specific articles from the list.
    (Launch) With this option you can launch a specific article.
  4. Optional: Add new articles as you see fit.
  5. Launch all articles using the button at the bottom or launch articles one by one.
    Note: Once you launch a package in a knowledge base, it can't be launched to the same knowledge base again.
The articles you launched are available in your knowledge base under Knowledge > Articles.

What is the knowledge base launchpad?

The knowledge base launchpad in Bold360 AI enables content managers to jumpstart a knowledge base.

Note: The knowledge base launchpad is not available by default, it must be enabled for the knowledge base on account level. For more information contact your Customer Success Manager.

The knowledge base launchpad offers several packages content managers can choose from depending on the nature of the knowledge base they want to set up. The packages contain a set of intents with phrasings which serve as examples and guidance for content creation and can be changed. To access the knowledge base launchpad, go to Knowledge > Launchpad.

With the knowledge base launchpad, content managers can:

  1. Select from a list of predefined content packages
  2. Review the content of the package
  3. Edit the content of the package such as edit the intent, the phrasings and the content of the article
  4. Delete articles from the list that are not needed and add new ones
  5. Set articles live
  6. Launch the entire package or specific articles of the package to the knowledge base

See How to use the knowledge base launchpad to learn more about how to use the launchpad.

Automatic Language Translation using Google Translate API

Bold360 AI offers automatic language translation using multiple language translation engines. The translation engine supports more than 34 languages.

Features and Benefits

  • Language is detected automatically.
  • Answers and widget interfaces are automatically translated.
  • Once the translation symbol is shown, customers can see the text in the original language by clicking on the symbol.

How to activate the translation

You do not need an existing Google account to use Google Translate, but you must have the Google Translate Auto-Translation for Search feature enabled for your Bold360 AI account. Contact your Customer Success Manager for more information.

Translation Analytics

Bold360 AI offers comprehensive translation analytics at Analytics > Translations.

The analytics show the most popular languages, the most recent translated questions, and the number of questions that were asked, answered, and escalated via translation.

Creating Dedicated Language Knowledge Bases

Bold360 AI allows you to have a separate knowledge base per language. When a customer asks a question, the answer is provided from the knowledge base that matches the language of the question.

Make sure dedicated language knowledge bases have been created for you. Contact your Customer Success Manager for more information.
  1. In the AI Console, go to Touchpoints > Widgets > Get the code and select a widget.
  2. Choose the knowledge base of that specific language from the Choose Knowledge Base drop-down menu.
    Note: For slim widget, select Enforce KB and then select your knowledge base. For legacy code, select Get legacy embedded code and then select your knowledge base.
  3. Copy the code and insert it into the foreign language pages script.

How can I handle various languages on the same topic?

There are two recommended ways to support your customers in various languages:

  • Use one knowledge-base for your local language and provide on-the-fly automatic translations. Read more here.
  • Create several knowledge-bases; one for each language. This option will ensure more accurate translations. Read more here

Which languages does Bold360 AI support?

Depending on which feature of Bold360 AI you use, the following languages are supported:

In Conversational widgets
Dutch English French
German Hebrew Hindi
Norwegian Spanish Turkish
In Search widgets
Arabic Azerbaijani Bulgarian Chinese Croatian Czech
Danish Dutch English Estonian Finnish French
German Greek Hebrew Hindi Hungarian Icelandic
Indonesian Italian Japanese Korean Latvian Lithuanian
Malay Maltese Norwegian Polish Portuguese Romanian
Russian Serbian Slovak Slovenian Spanish Swedish
Thai Turkish Vietnamese      
In the customer chat window
Afrikaans* Arabic* Bengali* Bulgarian* Chinese (Simplified)* Chinese (Traditional)*
Croatian* Czech* Danish* Dutch English (UK)* Finnish*
French French (Canadian)* German Greek* Hebrew* Hindi*
Hungarian* Icelandic* Indonesian* Irish* Italian Japanese
Korean Malay* Norwegian (Bokmal)* Norwegian (Nyorsk)* Polish* Portuguese*
Portuguese (Brazilian)* Punjabi* Romanian* Russian* Slovak* Spanish
Swedish* Thai* Turkish* Ukrainian* Vietnamese*  

* You need to translate the branding of the chat windows.

In the Bold360 Agent Workspace
Dutch English French French (Canadian) German Hungarian
Italian Japanese Korean Portuguese (Brazilian) Spanish  

You can set up the support center as an integral part of your existing website using iframe.

  1. To set up the support center as a part of your existing page in iframe, add the following code to the website:
    Note: Remember to change the highlighted parts to your actual account settings.
    <!-- embedded support center code start -->
    <div id="nanorep-support-center"></div>
    <script>
    ! function(t, e, o, n, c) {
        var s = this.nanorep = this.nanorep || {};
        s = s[e] = s[e] || {}, s.host = c, s.path = n, s.domain = t, s.protocol = "https:" === location.protocol ? "https:" : "http:", s.on = s.on || function() { s._calls = s._calls || [], s._calls.push([].slice.call(arguments)) };
        var a = document.createElement("script");
        a.async = a.defer = !0, a.setAttribute("src", s.protocol + "//" + c + n + o), document.getElementsByTagName("head")[0].appendChild(a)
    }("your_support_center_domain", "supportCenter", "support-center.js", "/web/", "your_account_name.nanorep.co");
    </script>
    <!-- embedded support center code end -->
  2. Optional: To set predefined context values, place them after the support center domain as query parameters.

    Example: For example, the following code sets the Brand context category to be Apple:

    ("your_support_center_domain/?Brand=Apple","supportCenter","support-center.js","/web/", "your_account_name.nanorep.co") 
  3. Optional: The embed code in step 1 inserts the embedded support center with default settings. Override the settings in the Master Page template using the following code:
    <script src="//<%Site.nanoRepDomain%>/web/support-center.js"></script>
    <script>
    nanorep.supportCenter.init({  
                noDirectDomainAccess: false, // use "true" to disable direct access to support center     
                supportCenterDomain: '<%Site.nanoRepDomain%>', // Support center subdomain     
                mainSiteURL: '<%Site.MainSiteUrl%>', // Parent page URL     urlPrefix: '#/path' //prefix path for URL hash   });
    </script>

How do I add an additional phrasing to an existing article?

To add an additional phrasing to an existing question/answer, do the following:

  1. Go to Knowledge > All Articles in the Bold360 ai navigation menu.
  2. Search your question in your online Knowledge Base.
  3. Select your question (article) and click Phrasings.
  4. Enter a new phrasing and click Add.
  5. Click Publish to save the new phrasing.

How to create good phrasings for your article?

  1. Add synonyms. (For example: I want to buy tea: buy|porches|sell|stock|get etc.)
  2. Do not add more than 5-10 phrasings per article. That will make precision less accurate.
  3. Add keywords in your phrasings (from the previous example: tea).

For more information about phrasings, see these articles:

How do Synonyms work?

Best practices for phrasings and synonyms

How to create data source entities

Use data source entities in Bold360 AI to turn your data into smart conversations.

Make sure your widget is set to conversational. Learn more here.
In this example, we have a list of mobile phones and we want to create a bot that recommends users the best mobile devices based on their needs. We are going to set up a data source entity based on this list of mobile phones and then create an article that uses this entity.
  1. Create the list of mobile phones in a CSV file.

    A sample CSV file is attached at the end of this article. Learn more about the different data source options and requirements in this article.

  2. In the AI Console, go to Knowledge > Entities.
  3. Choose (Create Entity).
  4. Select data source entity for entity type.
  5. Provide a name and optionally a description for the entity.

    The entity name can contain letters, numbers and underscores. Spaces are not allowed.

    You can use the description to give the entity a more meaningful name or description that helps you identify it later.

    Example: Name your entity MOBILE_PHONES.

  6. Fill out the Data Source URL field.

    You can select the CSV file from your local computer and upload it to Bold360 AI' secure server or you can upload the file to a file hosting service like OneDrive and enter its URL here.

    Note: If you provide a link to the file's location, make sure it's publicly accessible.
    See What are the data source URL options? for more options.
  7. Set the response type to CSV.
  8. Choose LOAD.

    Result: The properties table is now loaded. Properties are the columns that are available in the data source file.

  9. Fill out the Properties table.

    See What are the property options for data source entities? and How do life cycle options work? to learn more about the options.

    Example:

    You should decide what questions the bot should ask the end users to offer the best possible recommendation about phones. In this example, the bot asks questions about brand, price range, and camera quality.

  10. Optional: Define fallback messages.

    See What are fallback messages? to learn more.

    Example: For this example, we enter Unfortunately, there's no mobile device that matches your preferences. for no results and set up a carousel to be displayed with the image, the brand and model, and the price if there are multiple results based on the customer's responses.

  11. Save your changes.
  12. Now go to Knowledge > Articles.
  13. Create a new article or open an existing one and add the entity in the article's body.

    In the article's body, create the final response of the bot: this is what the bot responds when there is a single mobile device to recommend. The response should use the entity to present the details of the selected mobile device. To insert an entity, you can either choose or enter [[ in the text editor and have a drop-down of all entities displayed. Once you add an entity to the response, the entity is automatically added to the article. You can expand the entity to see the questions that the bot may ask the user and adjust them to fit the article if needed.

    Note: You can drag-and-drop entities in the Article Editor to control the order of the questions in the conversation.

    Example: Intent: I want to buy a mobile phone

    Body: ?

  14. Publish the article.

You can now see the conversation in action in your widget ? just enter I want to buy a mobile phone.

Important: The bot doesn't always ask all the questions we set in the entity. It can automatically identify when there is no need to ask a question, for example, when there is only one possible result left. In addition, the order of the questions might differ as the bot automatically chooses the order based on what it estimates as the fastest way to get to a possible answer.

How to create a post-chat survey

You can create a post-chat survey for bot conversations using a custom entity that captures the survey questions.

You can configure the conversational widget to present a post-chat survey if the user has been inactive for the specified amount of time. Customer responses collected with such a survey are stored as entities and can be extracted through the reporting API in the form of a table that includes all survey responses.
  1. Create a custom entity that includes the survey questions.
    1. In the AI Console, go to Knowledge Base > Entities.
    2. Choose (Create Entity).
    3. On the New Entity page, Select Custom Entity for Entity Type.
    4. Enter a name and a description for the entity.

      Example: For example, use survey as name and post-chat survey as description.

    5. Select Intent in the Life Cycle field.
    6. Add the survey questions to the Properties Table.

      Example:

      Table 1. Example questions for post-chat survey
      Name Detect Bot Query Error Message Life Cycle
      NPS Number On a scale of 1 to 10, how would you rate your experience? Please enter a number. Intent
      Comment Text What can I do to improve? You've entered a wrong format. (Default message) Intent
      Note: If you set up the entity as it is in this example, you can use the API configuration in this article with minimal adjustments.
  2. Create an article that uses the custom entity you just created.
    1. In the AI Console, go to Knowledge Base > Articles.
    2. Choose (Create Article).
    3. Enter an intent for the article.

      Example: For example, Post-chat survey.

    4. On the Content tab, choose Add Entity and select the entity you created.
    5. In the text field of the Content tab, enter the text you want to present to the user once they completed the survey.

      Example: For example, you could add Thank you for your feedback.

    6. Publish the article.
    7. Open the article you just created and copy its ID.
  3. Set your conversational widget to present the chat survey after the visitor has been inactive for a specified amount of time.
    Note: Besides presenting the survey article after the visitor has been inactive for some time, you can also set the survey article as a persistent menu value or add it to specific articles using a quick action button.
    1. In the AI Console, go to Touchpoints > Widgets.
    2. Select Personalize under the type of widget you want to present the post-chat survey.
    3. Select the knowledge base or URL to customize.
      Note: Make sure the widget you select is a conversational one.
    4. On the Conversation tab, enter the ID of the article you created for the chatbot survey in the Customer Inactivity Message field.

      Result: The Customer Inactivity Timeout field appears.

    5. In the Customer Inactivity Timeout field, specify the time in seconds that should pass before the survey is presented.
    6. Save your changes.

What are quick action events?

When using Quick Action buttons, a special event can be created.

An event can be setting a new context or changing an entity value. Events can help you gather more information about your users and their specific preferences so that the bot can present the most relevant content for them.

For example, let?s say we have a context called User Type, and its value are Basic and Advanced. Suppose we have an article providing a general overview of our product. We can add a quick button like Learn more with an event that changes the User Type context to Advanced. Therefore, when the bot displays the Learn more article, it presents the variation of the article that has Advanced context.

Important: For a context value to be successfully set using an event, the context category must have Detect from user search enabled.

How else can you use context-based quick action buttons?

Let's say, you want to offer different shipping information for website visitors who are currently looking at an item and visitors who have actually purchased that item. In this case, you can create a shipping-related article with two variations: one where the context called purchased is set to true and one without any context defined. Then in your main article, you create the quick action button and define an event to open the shipping-related article when the visitor clicks that button. In this case, visitors who are just looking for an item may find general shipping information, while customers who have purchased that item may see the date when they are supposed to receive their products.

Tip: You can use synonyms to cover different user inputs. For example, if the buttons show Yes and No, you can create a synonym group that includes yes, yeah, ya and so on. This way, if the user types yeah instead of choosing the Yes button, they still receive the relevant answer.

Does Digital DX provide an accessibility-enabled experience?

Digital DX AI provides an out-of-the-box accessibility-enabled experience with our Slim Widget, which means we are certified to WCAG 2.0 AA .

For slim widgets, both conversational and search modes support accessibility for floating, side, and embedded widgets.

If you do not make changes to the basic functionality of the widget and UI, it will remain accessible. However, if any changes are made to color, fonts, or functionality of the widget, you may lose part or all of the accessibility experience.

Align with the accessibility requirements, or ask Digital DX AI to do for you, It is your responsibility to ensure that the changes you make do not influence accessibility.

What is the Bold360 AI Mobile SDK?

You can find the Bold360 AI Mobile SDK documentation in the Bold360 Developer Center.

You can also find information about our APIs and SDKs at the following websites:

New repository for mobile and HTML SDK documentation and code samples - https://github.com/genesys/

Legacy references still available:

How to reply to a ticket?

To answer an open ticket, do the following:

  1. Select a question to display the answer editor.
  2. At the bottom of the page, select one of the existing answers suitable for this question by using the Search for quick answers field, which presents suggested answers taken from your knowledge base. You can type in a query to find additional answers from the knowledge base.
  3. Click on a relevant answer to send this answer to the customer. You can edit the answer before sending.
  4. If no answer was found or you want to create a new answer, go to the Write Answer pane. A variety of links, images, videos and text types can be easily added to each answer using the toolbar.
  5. Choose one of the following options to choose what to do with your answer:
    • No, It's personal - to discard an answer after it is sent to the customer
    • Publish answer - Create an article from the answer and publish it into the knowledge base making it available for everybody
    • Decide later - Save the the answer and move it to the 'Suggested content' section under Knowledge. This answer will also be suggested to other agent's when answering a relevant ticket.
  6. At the Ticket Status field on the bottom left of the screen, select one of the following options:
    • 'Open' to define that this question still requires attention. The ticket will stay open and then added to the Open tickets queue.
    • Close to define that this question has been completely answered and no longer requires any more interaction with the customer. This ticket is automatically saved in the All Tickets branch, which contains all tickets' history.
  7. Click Send.

What browsers does the Digital DX AI widget support?

The Digital DX AI widget supports the following browsers to offer customers self-service on the go.

  • Google Chrome on Desktop and Android (Last 20 versions)
  • Mozilla Firefox on Windows (Last 10 versions)
  • Safari on iOS, iPadOS, and macOS (Last 3 versions)
  • Microsoft Edge (Last 5 versions)

Internet Explorer 10 and above is also available to use but it is not a recommended browser as it does not maintain support modern security and web standards.

TLS encryption support

The Digital DX AI platform uses TLS 1.2 encryption. Earlier TLS versions are not supported.

You can enable TLS 1.2 in the Windows Registry.

How to set up auto-replies in Digital DX AI?

You can set up automatic replies to your customers' emails. This is useful when you want to confirm that you have received an email.

  1. On the Admin Center > Ticketing Settings page, make sure you have a ticketing rule. See How do I create a rule in the Ticketing system?
  2. On the Ticketing Basics tab under 3rd party integration, type your organization's email address where you want to forward emails to.
  3. Select Basic for ticketing integration mode.
  4. Select the Pass customer mail address in Reply-To field option.
  5. Create the content of your automatic reply. See How to create an auto-reply when tickets arrive?
  6. Click Save Changes.

Setting Up Your SMTP Server and Your Ticketing System

Setting up the Digital DX AI ticketing system with your SMTP server is part of your initial configuration and setup process. The Digital DX AI ticketing system is the default ticketing system available to you, offering the ease of out-of-the-box technology, and automatically connecting to your knowledge base. The system also enables agents to leverage the Digital DX  AI knowledge base, meaning that they can provide full and accurate answers to customers in a speedy, efficient and timely manner.

There are three steps involved in setting up your ticketing system to function correctly and work in the most efficient manner possible:

  1. Setting up incoming mail
  2. Setting up outgoing mail
  3. Assigning Groups

Step 1: Setting up incoming mail

Once you have your Digital DX AI account, the first step in setting up your ticketing system is to configure your SMTP server to forward incoming customer support emails to the Digital DX AI ticketing system.

  1. Digital DX AI provides you with a default email address that will serve as the inbox address to be specified in your SMTP server. To access this default address, from your Digital DX AI dashboard, navigate to Admin Center > Ticketing Settings > Ticketing Basics.
  2. The default email address is displayed in the Inbox Address field.
    Note: If you have more than one Knowledge Base in your system that contains external information used by and relevant to customers, you have more than one default inbox email address and a separate procedure should be executed for each Knowledge Base.
  3. Configure your SMTP server to forward incoming email messages to the default Digital DX AI email inbox address.

Sending Ticketing to Specific Departments

You can further hone the scope of how customer support tickets are handled, by creating email ticketing specific to an internal department of the company, such as Sales, Marketing, IT Support or Billing.

  1. Create a specific Labels inbox for each department.
  2. Click Admin Center > Ticketing Settings > Ticketing Basics.
  3. Click the More Addresses link, under the default email address in Ticketing Basics. In the drop-down menu available, a unique Digital DX AI inbox address is shown for each label.
  4. Select the label you require. This enables you to configure your SMTP server and create a specific, drill-down email address for a specific internal department in our company, that are sub-set emails of the initial email address displayed in the previous procedure, and that you can configure in your SMTP server.
Important: You must create these labels prior to performing the setup procedure.

Step 2: Outgoing Mail

The next step is to specify how outgoing customer support mail is handled, by configuring your SMTP server for outgoing mail.

  1. From your Digital DX AI dashboard, navigate to Settings = > Ticketing = > Ticketing Basics and go to the Outgoing Mail section.
  2. Specify the following fields:
    • Sending email from: Enter the name you wish to appear as the sender of any customer support email in the first field, and the email address from which these emails should be sent. Note: These settings must also be configured in your SMTP server.
    • Ticket status on reply: Choose either Closed on first reply or Always Closed, according to your company's customer support policy.
  3. Select the Send emails through custom SMTP server checkbox.
  4. Enter your SMTP host, username and password, and select the Server requires authentication and Use secure connection (SSL) checkboxes according to your SMTP setup.
  5. Click Save Changes.

Digital DX AI then performs a verification of the entered SMTP settings to check that the SMTP server can function. The settings will only be saved once the verification has determined that they are correct. The verification may take a few minutes.

If the verification fails, a message is displayed, with an explanation as to why the SMTP settings were not saved.

Step 3: Assigning Groups

You can further subdivide labels into groups to easily manage which agent handles what kind of ticket. User groups help differentiate between agents and their responsibilities, by enabling permissions to a group that determines which tickets can be assigned to members of that group. To create a group, do the following:

  1. In the AI Console, go to Admin Center > Users.
  2. From the menu on the top, click Groups > Create Groups link to display the Group Editor.
  3. Click the settings icon to configure the group settings.
  4. On the Advanced tab, select whether tickets are assigned to all agents or agents only from a specific Knowledge Base.
  5. Optionally, select whether agents can see all tickets to "cherry-pick" those, see all tickets but can choose only certain tickets, or see only those tickets which are assigned to them.

What are the ticketing integration modes?

Digital DX AI supports all e-mail forwarding integrations, which can be used across CRM and ticketing systems such as Salesforce, Kayako, or Zendesk.

Digital DX AI supports two e-mail forwarding modes:

  • Advanced: each e-mail is forwarded to the CRM with a reply address to the Digital DX AI system. When the CRM system 'replies', the answer is saved in Digital DX AI and sent to the customer by Digital DX AI.
  • Basic: each e-mail is forwarded to the CRM as if it was sent from the customer himself (with the customer's e-mail as the actual reply-to address). The CRM 'replies' directly to the customer. The CRM can 'cc' the Digital DX AI inbox so that Digital DX AI to learn the answer.

The email forwarding integration can be set under: Admin Center > Ticketing Settings > Ticketing Basics tab and then select 3rd party integration.

Digital DX AI Ticketing System Overview

You can use the Digital DX AI ticketing system to process customer escalations.

You can access the ticketing system of Digital DX AI by selecting Ticketing from the menu on the left.

Once you open it, you have the option to view:

  • Open tickets
  • Tickets assigned to you
  • Unassigned tickets
  • All tickets

Open tickets is the main interface for the ticketing system. When a customer escalates a question to an agent, it is displayed in Open Tickets. This category contains the following:

  • New tickets, that is, tickets that haven't been answered yet
  • Non-closed tickets, that is, tickets that have been answered but not closed yet

With the ticketing system, you can:

  • Assign open tickets to agents
  • Create labels and assign them to tickets to group them
  • Search your knowledge base to find an answer to the customer's issue
  • Write a new answer or leave a comment on the ticket
It is possible to set up Digital DX AI to work with 3rd party ticketing systems. For more information see What are the ticketing integration modes?.

How do I define conversational flows?

There are multiple ways to enhance the conversational bot with call-to-action activities that enhance the user experience and help guide the flow and suggest options to the user. These can be added to the regular answers using the Conversation tab in Digital DX AI.

  1. In the AI Console, go to Knowledge > Articles and create a new article.
  2. On the Conversational tab, click the Conversation Enabled toggle to enable the conversational editor in the Article Editor.
  3. Select Add Flow.
  4. Add the main answer in the Content field that appears.
    Tip: If you want to display the content in multiple bubbles, enter the <page_separator/> tag in the text where you want to separate the content.
  5. Optional: Add rotating content.

    You can add variations for the content to create a more human-like user experience.

  6. Select +Child to create the options.
    Note: It is recommended to use between 5-7 quick buttons to avoid scrolling on the page.
  7. In the Title field that appears, enter the text you want to display on the button.

    The text on the button can be up to 20 characters long.

    Tip: Use synonyms to cover different user inputs. For example, if the buttons show Yes and No, you can create a synonym group that includes yes, yeah, ya and so on. This way, if the user types yeah instead of choosing the Yes button, they still receive the relevant answer.
  8. Select the type of button you want to use.

    You can either use quick buttons or persistent buttons. Quick buttons appear under the content as separate buttons and once the user selects an option, the rest of the options disappear. Persistent buttons appear as a list under the content and all of them remain visible to the user after they select one. Use the quick button if only one of the options is relevant to the user. Use the persistent buttons if multiple options could be relevant to the user.

  9. Add the content that is to be presented upon clicking the button.

    You have the following options:

    • Content: Enter the content that is to be presented to the user when they click on the button.
    • Variation ID: You can specify the article variation that is to be presented to the user.
    • Article: Use this option to direct the user to an answer in the knowledge base. You can search for articles by title.
    • Statement: Use this option if the answer has more variations associated with different contexts and you want the bot to display the relevant variation based on context. Enter the title of the article with multiple variations.
    • URL: You can create a link to a webpage with this option.
    • Phone Number: You can use this option to add a phone number and call the number defined when clicked.
      Note: Using this option affects the channeling score for phone on the Dashboard. For more information about the channeling score see What do I see on the Dashboard?.
    • External: Use this option to transfer parameters to an external interface. This is useful for handshaking with a customer app that uses certain parameters from the conversational bot.
  10. Optional: Besides plain text, you can also insert HTML elements in the conversation tree to enhance the conversation's look and feel, as well as its functionality. For example, you can enter a link in the conversational statement by adding an HTML tag in the Article Editor's text field.

How can I report that I found personal data?

Digital DX AI has a built-in mechanism to replace personal data with asterisks in user searches, but sometimes you may still find personal data in Search Optimizer, on the Dashboard or in Reporting.

In the AI Console, users with view permission for Analytics have the option to mark user queries that contain personal data so that they can be removed. You have modify permission for Analytics? Follow these steps to obscure and remove the data.

  1. Go to Analytics > Reports.
  2. Run a report that contains the Query field.
  3. Find the query that contains sensitive data and choose (Personal Informaition in Query).
The icon turns red and the value of the Holds PII column changes to True for the query.

How do I review overlapping phrasings?

While editing an article, you notice that it has overlapping phrasing.

Digital DX AI enables knowledge managers to easily optimize phrasings by providing a list of articles that use the same or similar phrasing.

  1. In the Article Editor, select Overlapping phrases above the intent.

    This button is only available for an article that contains overlapping phrasings with other articles.

  2. On the Overlapping phrases dialog, select an article from the list to open it for editing on a new tab.
  3. Edit the phrasing of the article you just opened or go back to the original article and edit the relevant phrasing there.

    Unique phrasings ensure visitors get the most relevant results during search.

  4. Publish the articles.

What about security in the Digital DX AI platform?

We work hard to ensure your security and recognize its importance:

  • Bold360 AI servers are hosted on Amazon's EC2 cloud, it is secured via AWS firewall services to insure that customer knowledge & data are safe.
  • Manage access to our servers is restricted to TLS cryptography.
  • Each of our customers is assigned a unique set of login credentials that only allows access to the customer's own cluster of data. All passwords are enforced for a strict complexity policy. Access to your data is only granted to privileged users in your organization and Bold360 AI privileged users.

Add increased security for Slim and Harmony widgets

To increase the level of security, Digital DX AI introduced Cross-origin resource sharing (CORS) policies to the APIs. This enhancement enables you to restrict your widgets and Support Centers to a single URL domain or a list of domains. If a malicious intruder copies your HTML snippet code and adds it to any other site, the widget or Support Center on that site will not work.

Until now, you could define the URL of a secure Support Center in the Main Site URL field on the Touchpoints > Support Center > Settings page. From now on, to set up a secure Support Center, do the following:

  1. On the Touchpoints > Support Center > Settings page, copy your Support Center URL from the Main Site URL field.
  2. Go to the Admin Center > Preferences page.
  3. In the Allowed Origins field, paste your Support Center URL that you have copied from the Main Site URL field.
  4. Add your main customer website URL to the list.
  5. Save your changes.

For more information and to see more Bold360 certifications, please visit the LogMeIn Trust & Privacy Center.

Announcements

Genesys DX/Bold360 End of Life: January 2024

The Genesys DX (Bold360) platform will end of life on January 31st, 2024. This difficult decision was announced in March, 2023.  

Genesys continues to make a strong commitment to Genesys Cloud, while tightening the portfolio to further accelerate feature growth on the platform. Part of that included bringing over key Genesys DX features to Genesys Cloud CX, such as Knowledge Optimizer that focuses on ease-of-use knowledge management. Digital only licenses for Genesys Cloud were also introduced late last year, which are suitable to those who are not looking for voice capabilities or who need agent seats that only feature support for digital channels. 

Details on the end of life timeline

As of January 31st, 2024, access to Genesys DX product interfaces and customer-deployed components stop to function. Users will no longer be able to log into product interfaces, and all of the boldchat/bold360/nanorep domains will become unavailable for use. If you are curious on what the code on your website related to this might look like and how to remove it, we encourage referencing this post on the DX community

After January 31st, 2024, admins will still be able to get access for an additional 30 days. This period is meant to allow for extracting the necessary data from the platform. Historical data extraction from your account will be available to retrieve by data extraction APIs (Bold360 APIs and Nanorep APIs).