Inbenta tracks the actions that users perform with every product (ChatbotKnowledge and Search). There are two categories of tracking: Atomic and Sessions tracking. This data is logged separately, and it is used to generate the reports that you see in the product app's Dashboards section.

This section describes the kind of Events that are tracked. To see the different categories of logsclick here. To see possible Content Sourcesclick here.


Events are any interaction between the user and the interface. Note that not all data is the result of user actions. Because product configurations and implementations are extremely varied, an event could be almost anything: send an input in a chatbot conversation, clicking on a result, sending a contact form, etc.

Here is a short list of some of the events that Inbenta usually takes into account:

  • Submitting a Search
    When a user submits a search query, and if this search produces results, it means the application found content that matches the query. This is called matching. It is tracked automatically.
    Depending on the source of the results, matchings can be internal or external (See Content Sources).
  • Clicking on matchings, categories, etc.
    When a search return a single search result, the click action will be automatically tracked. When a search return several matchings, you must manually send the user click event to track this action.
  • Rating a content
    The rating feature enables users to give feedback on how helpful a given content was in solving their query. A rating can be either positive or negative. If it is negative, users may leave a comment as well (optional). Ratings are not tracked by default and must be tracked manually.
  • Submitting a ticket
    This action marks when a user submitted a contact form to escalate to a real agent or contact the company directly. Most of these actions are tracked automatically within a session but you must track explicitly when the ticket is submitted.

What is tracked and how? 

In atomic logs, actions are listed according to a log type. Session logs actions use data keys. This table shows all log types and data keys that come predefined in a new instance.

If an action has a log type, it can be tracked in Atomic logs. If it has a Data key, it can be tracked in Session logs. If it has both, it can be tracked in both. Here are a few examples:

Log type
Data key
- START Marks the start of the session.
- USER_BROWSER Stores the browser information of the user.
- MY_CUSTOM_KEY Custom data key to track user actions.
SEARCH SEARCH Indicates the user performed a search.
MATCH CLICK_MATCH Indicates a user clicked on a result.
INSTANT_MATCH INSTANT_CLICK Indicates a user clicked on an instant result.

For full details about each product, see their respective API Tracking Guidelines page in Chatbot, Knowledge and Search.

Data keys

A data key also has a priority level that is given to its action. The session reports use the priority of an action to decide the most important action performed by the user. This most important action becomes the result of the session. Here are a few examples:

Priority level Data Key
1 (highest) CONTACT_TICKET
10 (lowest) START
0 (no priority) USER_INFO
(custom) CUSTOM

For full details about each product, see the Data Key section in their respective API Tracking Guidelines page in Chatbot, Knowledge and Search.