What is a good website conversion rate? Benchmarks and how to read yours

A good website conversion rate is one measured the same way every week and moving in the right direction. Published benchmarks give you a rough range: Littledata’s November 2026 benchmark of 438 Shopify stores puts the median share of sessions with a purchase at 1.3%, while Unbounce’s 2024 report puts the median landing page at 6.6%. Both numbers are right. They count different actions on different kinds of pages, which is why a single “good” figure doesn’t exist.

What website conversion rate means

Conversion rate is the share of visits (or visitors) that complete an action you care about. Every rate rests on three choices:

1. The action. A purchase, a contact form, a booked call, a newsletter sign-up. A store that counts newsletter sign-ups will show a higher rate than one that counts only orders.
2. The denominator. Sessions or users. One person can visit three times before buying, so the same purchase is 1 in 3 per session and 1 in 1 per user.
3. The pages included. The whole site, or one landing page built for one offer. A blog post that ranks for an informational query is not trying to sell, and it pulls the site-wide average down.

Change any of the three and the number changes, even if nothing about your business did.

How GA4 calculates conversion rate

GA4 has no metric called “conversion rate”. It works with key events (what GA4 used to call conversions): any event you mark as important to your business, as Google’s documentation puts it. Two metrics turn them into rates. Google defines them in its metrics reference:

GA4 metric	Google’s definition	Use it for
Session key event rate	Number of sessions in which a key event happened, divided by total sessions	The standard website conversion rate, per channel or landing page
User key event rate	Number of users who performed a key event, divided by total users	How many people convert at some point, across visits
Key events	The number of times users triggered a key event	The volume behind the rate

The rates count sessions or users, not events. A session with two purchases counts once in session key event rate. Key events, the count, would show two. So key events divided by sessions is not the same number as session key event rate.

By default, they include every key event on the property. If you have marked purchase, a newsletter sign-up and a contact form as key events, a session with any of the three counts as converted. Google’s About key events page explains that the Key events column in reports has an All events menu where you pick one specific key event. For a purchase-only rate, divide purchases by sessions yourself (close enough, though a session with two orders counts twice), or use an exploration, where the same page lists session key event rate as an available metric.

Session key event rate sits in Reports > Acquisition > Traffic acquisition, next to sessions and key events for each channel (Google’s documentation). If you are new to these reports, start with how to see your website traffic in GA4.

Check your key events before you read the rate

When a GA4 conversion rate looks wrong, check the list of key events before the site.

Google’s setup page states that GA4 marks purchase as a key event by default. It also states that when you enable Ads Personalization with a linked Google Ads account, these events are marked as key events too: add_to_cart, add_to_wishlist, begin_checkout, session_start, view_item, view_item_list and view_search_results.

If session_start is a key event, nearly every session triggers one, and your session key event rate sits near 100%, which tells you nothing about sales.

To check, go to Admin > Data display > Events and look at which events carry the key event star. The same page notes that marking an event as a key event applies from that moment on and doesn’t change historic data. So if someone added or removed a key event in the middle of your comparison window, the rate before and after measures different things. Write the date down.

Website conversion rate benchmarks you can actually use

Most “average conversion rate” articles don’t say what was counted. These two sources publish their method.

Source	What it measures	Sample	Headline figures
Littledata ecommerce conversion rate benchmark	Share of sessions with at least one purchase, from GA4 data	438 Shopify stores using Littledata, 90 days from 3 August to 31 October 2026	Median 1.3%, mean 1.9%, top 20% above 2.7%, top 10% above 3.6%, bottom 20% below 0.5%
Unbounce Conversion Benchmark Report (2024)	Conversions on landing pages built with Unbounce	Over 41,000 landing pages, 464 million unique visitors, 57 million conversions	Median 6.6% across all industries; nine industry medians from 3.8% to 12.3%

Littledata is the closer match for a DTC store. Its definition is the same as session key event rate filtered to purchase, measured in GA4. It also breaks the median down by category (as published in early November 2026): Beauty and Skincare 2.6%, Health and Supplements 2.4%, Food and Beverage 2.2%, Fashion and Apparel 1.0%, Home and Furniture 1.0%. Mobile and desktop medians are close, at 1.5% and 1.3%. Keep in mind that Littledata sells tracking for Shopify, the sample is its own customers, and the page is regenerated, so the figures will move.

Unbounce measures landing pages, not websites. A landing page has one offer and, usually, paid or email traffic sent to it on purpose. The report also shows how much the source matters: email traffic converts at 19.3% on average, against 12% for paid social and 10.9% for paid search. A services business comparing its whole site with a 6.6% landing page median will always feel behind, because most of its sessions land on pages that were never built to convert.

Shopify’s own guide to ecommerce conversion rate puts it bluntly: “The idea of a universal benchmark for ecommerce conversion rates is a fallacy.” A benchmark tells you whether you are in a plausible range, not whether this week was good.

If you run a Shopify store, also expect GA4 and Shopify to show different conversion rates for the same week. We explain why in Shopify Analytics vs GA4.

GA4’s peer benchmark for key event rates

GA4 has its own benchmarking on the Home page overview card. Both session key event rate and user key event rate are on the list of metrics it covers (Google’s documentation). Turn it on and the card shows your trend line, the peer group median as a dotted line and a shaded range between the 25th and 75th percentiles. GA4 picks a default peer group from the industry category you chose at setup and signals such as your site URL, and you can change it. It requires the Modeling contributions & business insights setting in Admin, under Account settings.

Treat this one with more care than the engagement rate benchmark we covered in what a good bounce rate is. Engagement rate is mostly comparable across properties, though it too counts a session with a key event as engaged. Key event rate depends entirely on which events each property marked as key events, so your peers may be counting newsletter sign-ups, add to carts or even session starts. A gap with the peer median can be a difference in setup rather than performance.

How to read your own conversion rate

Your own history is the benchmark that tells you something. Five steps:

1. Pick one definition and write it down. For a store, sessions with a purchase divided by sessions. For a services site, sessions with a form submission or booked call, if those are tracked as events and marked as key events. Keep the same one every week.
2. Split by channel. In Traffic acquisition, read session key event rate per channel. Channels convert at very different rates (see the Unbounce figures above), so a change in the mix moves the site-wide rate on its own.
3. Split by landing page. A blog post and a product page do different jobs. Compare each page type with itself.
4. Take 8 to 12 weeks as a baseline. Note the usual range per channel. The week-to-week spread is your noise. A move inside it is not news.
5. Compare like with like. A week with the previous week, a Tuesday with recent Tuesdays. Weekends often behave differently.

Then read the rate next to the counts. Here is an illustrative example (not real data). A store has 10,000 sessions and 150 sessions with a purchase: 1.5%. A press mention adds 2,000 sessions from cold readers who mostly browse, and purchases rise to 160. The rate falls to 1.33% while sales went up. Nothing broke. The traffic mix changed.

The reverse happens too: a rate that rises because a weak channel dried up can hide flat sales. That is why session key event rate sits in our list of marketing KPIs worth checking weekly next to key events, not on its own.

Getting the trend without opening GA4

This only works if you look every week. Dialytics, our product, does that reading for you. Every Monday it posts a digest in Slack that compares the past week with the one before, sessions and key events by channel included, and says what moved and why it probably moved. A separate daily check sends an alert only when traffic or key events are unusual for that day of the week, compared with the same weekday over about 12 weeks, and never more than one a day. If a number looks off, mention the bot in the thread and ask which channel drove it.