Conversion rate is the percentage of visitors, sessions, or interactions that end in a desired action, such as a purchase, a sign-up, or a demo request. It is one of the simplest ways to judge whether traffic is turning into results.
This page covers the formula, how Google Analytics 4 and Google Ads report it, why industry benchmarks are risky to lean on, how traffic from AI assistants fits in, and a practical process for measuring and improving your own rate.
What is Conversion Rate?
The basic formula is: conversion rate = (number of conversions / total visitors, sessions, or interactions) x 100. If 1,000 sessions produce 25 purchases, your conversion rate is 2.5%.
The tricky part is the denominator. You can divide by users, sessions, or ad clicks, and each produces a different number. You also have to define what counts as a conversion. A "conversion" might be a sale for an online store, a qualified lead for a B2B company, or a booking for a restaurant.
Different tools use different terms. In GA4, important actions are called key events, so the conversion rate metrics are named key event rates. Google Ads defines conversion rate as conversions divided by ad interactions that can be tracked to a conversion.
Why Conversion Rate matters
Traffic alone does not pay bills. Conversion rate tells you how well a page, channel, or campaign turns attention into outcomes, and it helps you compare sources fairly. A channel with less traffic but a higher conversion rate may deserve more investment than one with big numbers and few results.
For SEO, conversion rate helps separate rankings that matter from rankings that only produce visits. For AI visibility, it helps answer a newer question: are people who arrive from ChatGPT, Perplexity, or Gemini more or less likely to act than people from Google search?
The honest limit: conversion rate is a ratio, so it can rise simply because low-intent traffic fell. Always look at it alongside total conversions and revenue.
How Conversion Rate works
How GA4 reports it
GA4 offers two main rates. Session key event rate is the share of sessions in which at least one key event was triggered. User key event rate is the share of users who triggered at least one key event. Google added both to the default Traffic acquisition and User acquisition reports in mid-2024, as PPC Land reported. Google's key events help article also lists session key event rate among the metrics available in explorations.
Both rates count a session or user once, no matter how many key events they trigger. By default they include all key events combined. In the standard acquisition reports, you can filter to a single key event, such as a purchase, to see that event's rate on its own.
Conversion rate in Google Ads
Google Ads calculates conversion rate per ad interaction. Because a single click can lead to multiple counted conversions, Google notes that the rate can exceed 100% if you track several conversion actions or count every conversion.
Why benchmarks mislead
You will find many published "average conversion rate" figures. Treat them with caution. Results swing widely depending on industry, price point, traffic mix, device, what counts as a conversion, and whether the rate is per session or per user. Your own historical trend, segmented by channel and page type, is a far more useful comparison than someone else's average.
AI referral traffic and conversion
Studies on how AI referral traffic converts disagree, which is itself a useful lesson. Ahrefs reported in 2025 that AI search visitors made up 0.5% of its traffic but 12.1% of signups, while noting it was data from a single site. A study of 94 ecommerce brands covered by Search Engine Land found ChatGPT sessions converted somewhat better than non-branded organic search. A peer-reviewed Marketing Science study of 973 ecommerce sites found ChatGPT referrals converted below most traditional channels, though the gap was narrowing and smaller for complex products.
The takeaway is to measure your own. GA4 added an AI Assistant channel to its default channel group in May 2026, per Google's release notes. Keep in mind that some AI clicks may arrive without referrer data and show up as direct traffic, which can understate the channel.
Conversion Rate in practice: an example
A hypothetical online kitchenware store reviews last month's Traffic acquisition report, filtered to the purchase key event:
- Organic search: 40,000 sessions, 800 with a purchase, a 2.0% session key event rate.
- Paid search: 10,000 sessions, 300 with a purchase, 3.0%.
- AI Assistant: 1,500 sessions, 36 with a purchase, 2.4%.
Paid search converts best, but organic produces far more total sales. AI assistant traffic is small and converts slightly better than organic, although with only 36 purchases, one month is not enough to draw firm conclusions. The store decides to keep watching the AI channel for several months before changing strategy.
How to measure and improve Conversion Rate
- Define your conversions. Decide which actions matter, such as purchases or qualified leads, and mark only those as key events in GA4.
- Choose the right denominator. Use session key event rate for page and campaign analysis, and user key event rate when you care about people rather than visits.
- Segment before you judge. Break the rate down by channel, landing page, device, and new versus returning users. Blended averages hide problems.
- Tag and separate AI traffic. Use GA4's AI Assistant channel and UTM parameters where you control links, and watch for unexplained changes in direct traffic.
- Fix friction on high-intent pages. Improve load speed, clarity of offers, forms, and checkout steps on pages that already get qualified traffic.
- Test changes properly. Change one thing at a time and wait for enough conversions to tell a real difference from noise.
- Track volume alongside rate. Report total conversions and revenue next to conversion rate so you do not celebrate a higher rate that came from losing traffic.
Common mistakes and misconceptions
"A 2% to 3% conversion rate is normal for everyone." There is no universal normal. Benchmarks vary enormously by industry and by how conversions are defined, so compare against your own history first.
"Higher conversion rate always means better performance." If you cut low-intent traffic, the rate rises while total sales may fall. The ratio only makes sense next to absolute numbers.
"GA4 and Google Ads conversion rates should match." They use different denominators (sessions or users versus ad interactions) and can use different attribution and counting settings, so differences are expected.
"AI traffic always converts better." Some studies say yes, others say no. The results depend on industry, product complexity, and measurement, so test with your own data.
Conversion Rate vs. click-through rate
Click-through rate measures how many people who saw a link or ad clicked it. Conversion rate measures how many of the people who arrived took the action you wanted. CTR is about earning the visit; conversion rate is about what happens after it. A page can have a strong CTR and a weak conversion rate if it attracts clicks the content cannot satisfy.
How Butter approaches Conversion Rate
We start with measurement before optimization: agree on which key events matter, confirm they are tracked accurately, and segment by channel so search, AI assistants, and ads can be compared fairly.
Butter Analytics connects website performance with 85+ tools in one dashboard, which helps put conversion data next to traffic and revenue. Fundamentals first: a trustworthy conversion rate for your own business is worth more than any industry benchmark.
Frequently asked questions
How do you calculate conversion rate?
Divide the number of conversions by the total number of visitors, sessions, or interactions, then multiply by 100. For example, 50 conversions from 2,000 sessions is a 2.5% conversion rate.
Where is conversion rate in GA4?
GA4 calls it key event rate. You will find session key event rate in the Traffic acquisition report and user key event rate in the User acquisition report, and you can add them to explorations.
What is a good conversion rate?
It depends on your industry, offer, traffic sources, and how you define a conversion. The most reliable benchmark is your own past performance, segmented by channel and page type.
Does traffic from ChatGPT convert better than Google?
Research is mixed. Some single-site and agency studies found AI referrals convert better, while a 973-site academic study found ChatGPT traffic converted below most traditional channels. Measure it on your own site using GA4's AI Assistant channel.
Sources and further reading
- About key events, Google Analytics Help
- Conversion rate: Definition, Google Ads Help
- GA4 adds key event rate metrics to user acquisition and traffic acquisition reports, PPC Land
- Does AI Search Traffic Convert Better Than Traditional Search?, Ahrefs
- ChatGPT ecommerce traffic converts 31% higher than non-branded organic search, Search Engine Land
- ChatGPT Referrals to E-Commerce Websites: How Do LLMs Compare Against Traditional Channels?, Marketing Science