More in Product analytics
- Autocapture and the AI event inbox
- Build a funnel that explains drop-off
- Identify users across web, app and mini-programs
- What is autocapture, and how does it work?
- How do I review and accept events in the inbox?
- What are unplanned events, and how do I manage them?
- When should I use a funnel instead of a goal or path?
- How do I build a funnel?
- How do I read where a funnel drops off?
- What is identified mode, and should I turn it on?
- How do I call login to identify a user?
- How is one person merged across devices?
- How does identity work on apps and mini-programs?
- How do I track clicks on outbound links?
- How do I track file downloads?
- How do I track form submissions?
- How do I track custom events?
- How do I track button clicks?
- How do I create a heatmap?
- How do I track scroll depth?
- How do I track ecommerce purchases and revenue?
- How do I set up goals and track conversions?
- How do I query my data with SQL?
- How do I create a custom metric?
- How do I analyze accounts with groups?
- How do I connect AI agents with MCP?
- How do I create user segments?
- How do I run an RFM analysis?
- How do I record and watch session replays?
- How do I analyze user paths?
- How do I build a retention report?
- How do I analyze how an event is distributed?
- How do I create a tracking plan?
- How do I view a user’s profile and activity?
- How do I read the lifecycle report (new, returning, resurrecting, dormant)?
- How do I compare attribution models for a conversion?
- All guides in this category
Still have a question?
Could not find it? A person reads every message, usually within a few hours on business days.
Chat with usShort answer
Open All-platform analytics › Operations › AI Assistant and type a question with a metric, time range and optional comparison or breakdown. The answer shows the query it ran; numbers always come from that query. Why did it change? splits a change by channel, country, device and other dimensions.
Short answer
Open All-platform analytics › Operations › AI Assistant, type a question such as "visitor trend for the last 7 days versus the previous period" on the Ask tab and press Enter. The answer shows the query it ran, so you can check it, open it in the analyzer or save it. Numbers always come from a query on your project, not from the model.
1. Ask a question
Name the metric, the time range and, if you want one, the comparison and breakdown:
| Good question (sample) | What it becomes |
|---|---|
| Visitor trend for the last 7 days versus the previous period | Visitors, 7 days, compared with the previous 7 |
| Signups last 30 days by channel | Signup event, 30 days, broken down by channel |
| Pageviews this month, mobile only | Pageviews, this month, filtered to mobile |
Follow-ups such as "now by country" or "mobile only" edit the previous query instead of starting over.
2. Check the answer
Every answer includes the query: metric, time, comparison, breakdown and filters. Read it before you trust the number. Open in analyzer and Save as insight are buttons you click; the assistant does not do them for you.
3. Ask why a number changed
Click Why did it change? to carry the current metric into the Why did it change tab. It splits the change versus the previous period or the same period last year by channel, referrer domain, country, device, entry page, platform and related dimensions, one or two levels deep. Contribution is a value's change divided by the total change.
Sample: visitors fell by 1,000 and organic search fell by 700, so organic search contributed 70% of the drop.
4. How the model is used
- A question is parsed by rules first. When confidence is high and it is not a follow-up, the query runs with no model call.
- The model is used when confidence is low, nothing was parsed, or the thread has history. It must produce the same query shape, which is then run.
- The written reading comes from rules. With a model on and under the daily cap, the model only rewrites that draft and must not add numbers that are not in the result.
- With no model configured, the corner badge reads Rules only (no model calls) and every answer stays on the rule path.
- Model calls have a per-site daily cap shown at the top of the page; a cap of 0 means no cap.
5. What it will not do
It does not search a knowledge base, create monitors or send notifications. For alerts, see set up traffic alerts. To let your own AI agents read the same data, see connect AI agents with MCP.
Common problems
| Symptom | Cause | Fix |
|---|---|---|
| The answer misses the question | Metric, time or breakdown unclear | Name all three, or use a suggested question |
| Plain wording, no polished summary | No model, or daily cap spent | The rule reading still has the right numbers |
| Wrong project | Questions use the current project | Switch project first |
Did this answer your question?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.