- 23 Jan 2026
- 3 Minuten zu lesen
- Drucken
- DunkelLicht
- pdf
Ask Whatfix AI Best Practices
- Aktualisiert am 23 Jan 2026
- 3 Minuten zu lesen
- Drucken
- DunkelLicht
- pdf
Ask Whatfix AI is an AI-powered capability that automatically converts end-user engagement queries into Trend Insight, Funnel Insight, or User Journey charts.
With Ask Whatfix AI, you can,
Ask analytics-related questions, such as, “What is the most engaged User Action?”, or “How many users engaged with the Pop-up called Feature release?” to automatically generate Insights (Trends, Funnels, and User Journeys)
Ask follow-up questions based on the generated Insight in Ask Whatfix AI.
View the AI-generated Insight to make changes and then save the Insight.
Use Deep Think to uncover key patterns in user behavior and get suggestions to improve adoption.
The following are some recommendations to help you get accurate responses from Ask Whatfix AI:
Expand the following accordions for more details:
Best practices to optimize content naming for AI discovery
Follow a standard naming convention
Name User Actions using the Page + Action format. The format provides context for the User Action location and the UI element.
Examples: Leads save, Contact add
Non examples: save1, delete
Publish content and User Actions
Publish Whatfix content and User Actions so that Ask Whatfix AI can show engagement data for content that is live on the application and is engaged with. For more information, see Push all Whatfix content and widgets to Production and Publish User Actions.
Best practices for phrasing AI prompts
Provide descriptive context
Use descriptive sentences to provide Ask Whatfix AI with context.
Include the following details in your query:
What: The content type such as a Flow, Pop-up, Beacon, or Page View, and name of the User Action or content.
When: The time period, such as the last 7 days, this month, or this quarter.
How: The metric, such as Total events, Unique users, or Insight type (Trend, Funnel, or User Journey).
Filters or breakdowns (optional): Add filters or breakdowns to view data by location, segment, or device type.
Examples:
How many users engaged with the Leads save User Action in the last 30 days break down by country?
Where was the drop-off in the Create a contract Flow in the last 7 days?
Non examples:
Show Whatfix engagement
Break down event by country
Avoid Persönlich identifizierbare Informationen (PII) data
Ask Whatfix AI does not display answers for queries containing PII data such as User ID, User type, User Attributes configured using Advanced Customization (AC), Reserved variables, and Cohorts. Ensure that your questions don’t include any PII data.
Examples:
How many users interacted with the application using the Chrome browser in the last 30 days?
How many users from Bangalore city are dropping off from the process with Competitor Tip, Standard Record View, and Ocean FCL actions?
Non examples:
Has johndoe@abc.com interacted with the Pop-up?
Has the Sales team interacted with the new feature?
Who is a power user in this application?
Ask follow up questions
Ask follow up questions and continue the conversation with Ask Whatfix AI to explore your data further. Ask Whatfix AI remembers the previous query context and generates a new Insight based on the follow-up question.
Example:
How many users completed the Flow - Create a Lead?
Follow up question: Where did users drop-off in the Flow, Create a Lead?
Use Deep Think
Use Deep Think to understand data patterns. Deep Think surfaces trends in user behavior and suggests insights to help make informed decisions.
Example:
How many users completed the Create a contract Flow?
How can I increase the completion rate of the Create a contract Flow ?
Understand unsupported functionalities in Ask Whatfix AI
Ask Whatfix AI does not support the following functionalities:
Navigate to the different parts of the Whatfix Analytics dashboard.
Search for existing Insights or User Actions created on the Whatfix Analytics dashboard.
Create a User Action, Dashboard, Event Group, or Cohort based on your analytics queries.
Answer Whatfix feature-related questions, such as, “What is a User Action?” or “How can I export a Trend Insight?”
Ask Whatfix AI does not support functionality currently unavailable in Product Analytics (PA), including comparing engagement data across time periods or creating Funnels using DAP events.
Understand supported and unsupported engagement queries
Use the following table to understand the engagement queries supported by Ask Whatfix AI:
Engagement query | Supported or not supported by Ask Whatfix AI |
|---|---|
Engagement with Whatfix content, including Flows, Beacons, or Pop-ups |
|
Engagement with User Actions |
|
Page View engagement |
|
Engagement with Whatfix content, Page Views, or User Actions based on all user filters except User ID, User type, User Attributes configured using AC, Reserved Variables, and Cohorts |
|
Engagement with Event groups |
|
Engagement with Session metrics |
|
Engagement with Event Attributes |
|

