Get answers from your Google Analytics data

You can ask Kite about your Google Analytics data in plain English, with no report names or menu paths. You type the question in a conversation with Kite, and it answers there. Questions like these work as written:

• "Which pages brought the most signups last month?"
• "Where did our traffic come from this week, and how does that compare with last week?"
• "Did the pricing page get more visits after we changed it on the 3rd?"
• "How many people are on the site right now, and what are they looking at?"
• "Which blog posts lead to demo requests?"

Kite reads the question for the four things every GA4 report needs: what you want to count, what you want to group it by, the date range, and any filter. When a question leaves one of those open, Kite picks a sensible default and says so. For example, "this week" becomes the last seven complete days, compared with the seven days before.

When an answer is easier to read as a chart or table, Kite includes one in its reply. In the Kite web app, those charts pull current numbers again when you open the conversation later.

Before Kite answers a question about conversions, it checks which key events your property records. This matters because GA4 only knows what someone told it to track. If your signup form was never set up as a key event, there's no signup number to report, and any answer that pretends otherwise is a guess.

When the event exists, Kite uses it and names it in the answer, such as "signups, counted by your sign-up key event." When the event you're asking about is missing, Kite tells you that plainly and suggests the closest thing your property does track, such as visits to the thank-you page that follows the form. It also tells you what would need to be added to measure the real thing.

Kite does the same with your other settings. If your team built a custom channel group that separates partner traffic from other referrals, Kite uses it. If you track something specific to your business as a custom dimension, such as plan type, Kite can group by it.

Kite answers the questions people usually build GA4 reports for, such as which landing pages perform best and which traffic sources bring in conversions. It runs the report through your Google Analytics connection, so you don't need to open an exploration or remember where a standard report lives.

For top landing pages, Kite groups sessions by the first page each visit started on and counts the sessions, engagement, and key events for each one. It sorts the list by whatever you care about most. A page with 4,000 visits and 12 signups and a page with 600 visits and 30 signups tell different stories, so Kite shows both the count and the rate.

For traffic sources, Kite tells you which attribution it used. By default, it credits the source of the session where the conversion happened, because that's what most people mean by "which source drove it." If you want to know which channels first bring people in, Kite runs the same question by first-user source and shows you where the two answers differ. A common pattern is organic search finding people first and email bringing them back to convert. Seeing both versions keeps you from cutting the channel that starts the relationship.

When an answer shows a problem, Kite can trace it to the step where people leave. Suppose the signups answer shows plenty of visits to your pricing page but few completed signups. Kite builds a funnel from your own events, for example pricing page, signup form started, and signup completed, and shows how many people reach each step and how many drop out between them.

Kite can split the funnel by device or traffic source, which often explains the drop. If mobile visitors start the form at the same rate as desktop visitors but finish at half the rate, the form itself is the likely problem, not the traffic.

Google Analytics can tell you which campaigns and pages bring in signups, but it usually can't tell you which of those signups became paying customers. Kite can answer that question by reading Google Analytics and Stripe together.

Kite takes the campaign and landing page behind each signup from Google Analytics and matches them to customer records in Stripe. The answer shows which campaigns and pages produce paying customers and revenue, not just form fills. A campaign that brings in many cheap signups can look like your best channel in Google Analytics and turn out to be your worst once you see who paid.

A question you ask every Monday can become a report that arrives every Monday. Once Kite has answered a question you'll want again, it can put it on a schedule in your timezone. Each week, it runs the same reports, compares the results with the week before, and explains what changed and why it matters.

Kite can also keep a standing watch on your Google Analytics traffic. The watch reruns a saved report on a schedule and stays quiet until something changes enough to matter, such as visits dropping sharply after a site update. Then it tells you what moved.

An answer is most useful when something happens next, and Kite can do the work the answer points to. If a landing page brings in plenty of visits and few signups, Kite can rewrite the page or propose a test. If a traffic source converts well but sends little traffic, Kite can plan more content or campaigns for that channel.

Kite asks before it publishes a page or spends money. After the change goes live, it reads the same Google Analytics numbers again and compares the period before the change with the period after, so you can see whether the work helped.