Your e-commerce conversion rate is the share of visits to your online store that end in an order. It is one of the first numbers store owners check, and one of the easiest to misread: a rate that looks low can be healthy for what you sell, and one that looks fine can hide a checkout that quietly fails on phones.
This guide puts published benchmarks in context, then shows how to get a rate you can trust from Google Analytics 4 (GA4) and find the funnel stage where most shoppers give up.
The short answer
- What’s normal: widely cited averages sit between about 1% and 3%, but rates vary several-fold by sector, price, device and traffic source. Treat benchmarks as rough context, not a target.
- The formula: orders ÷ sessions × 100. For example, 240 orders from 12,000 sessions is a 2% conversion rate. You can use users instead of sessions; just don’t switch between them.
- What holds it back: usually one stage of the funnel, not the whole store. Measure each step from product view to purchase, split it by device and traffic source, and fix the biggest drop-off first.
What’s a normal e-commerce conversion rate?
Search for the average e-commerce conversion rate and you will find numbers that disagree:
- Littledata benchmarked 2,800 Shopify stores in 2023 and found an average of 1.4%, with the top 10% above 4.7%. Desktop averaged 1.9%, mobile 1.2%.
- Dynamic Yield tracks a rolling twelve months. At the time of writing (August 2026), its global average was 2.72%, ranging from 0.72% for luxury and jewellery to 5.39% for beauty and personal care, with mobile (2.88%) ahead of desktop (2.37%).
They don’t even agree on whether phones or computers convert better. Neither is wrong: they measure different stores, periods and traffic, in different ways.
Why your rate differs from any benchmark
| What changes the rate | Tends to push it down | Tends to push it up |
|---|---|---|
| Price | High-value items people research first | Low-cost, familiar, repeat purchases |
| Traffic source | Cold social traffic, broad display campaigns | Email to past customers, brand-name searches |
| Content | A busy blog attracting early researchers | Most visits landing on products |
A sofa store can be doing well at a rate that would worry a coffee store. The comparison that tells you most is your store against itself: this month against the same month last year, plus a rolling three-month view, calculated the same way each time. Our guide to website KPIs covers what to track alongside it.
Conversion rate isn’t the whole story
Conversion rate counts orders, not money. Revenue per session (conversion rate × average order value) shows whether the same traffic earns more: a site-wide discount can lift the conversion rate while cutting margin, and a higher free-delivery threshold can lower it while raising order values.
How to calculate e-commerce conversion rate
E-commerce conversion rate = orders ÷ sessions × 100
Sessions or users. A user-based rate runs higher, because many shoppers visit more than once before buying. GA4 reports both (session key event rate and user key event rate); filter them to the purchase event, or sign-ups and other key events will inflate them. Pick one and label it.
Which orders count. Leave out test orders, orders staff key in for phone or wholesale customers, and marketplace orders that never touched your site. Keep orders later refunded, but track refunds separately.
One source for both numbers. Calculate the rate inside GA4, where orders and sessions are measured the same way and can be split by device, channel and landing page. Dividing store orders by GA4 sessions mixes two systems with different blind spots. Your store’s order count stays the truth about sales, so compare it with GA4’s purchases each month.
GA4 e-commerce tracking: getting a number you can trust
GA4 only understands a shop if it receives Google’s recommended e-commerce events, usually from your platform’s official integration or Google Tag Manager. For conversion, the key ones are view_item, add_to_cart, begin_checkout, add_shipping_info, add_payment_info and purchase, which needs a transaction_id, value, currency and the items bought. Google’s e-commerce measurement documentation lists them all, and integrations vary in how many they send.
Check the tracking before you trust the rate
Watch GA4’s DebugView while you place a real order, then cancel or refund it. Switch on debug mode first; Tag Manager’s preview does it for you. Tick off:
Expect GA4 to record fewer purchases than your store: cookie refusals and ad blockers hide some orders from it. A steady gap is fine for spotting trends; a gap that suddenly widens usually means something broke. Tracking conversions in GA4 covers testing and consent in more detail.
E-commerce funnel analysis: find the biggest drop-off
A single conversion rate tells you whether there is a problem; the funnel tells you where. Build one in GA4 as a funnel exploration (under Explore):
| Funnel step | GA4 event | Rate to watch |
|---|---|---|
| Visit to product view | view_item |
Share of visitors who view a product |
| Product view to cart | add_to_cart |
Share of product viewers who add to cart |
| Cart to checkout | begin_checkout |
Share of shoppers with a cart who start checkout |
| Checkout to purchase | purchase |
Share of checkout starters who buy |
The exploration counts users, not sessions, so it won’t match a session-based conversion rate. For finding the step where the share falls furthest, that doesn’t matter.
The second row is your add-to-cart rate. Some tools divide by all sessions instead; dividing by product viewers isolates how well product pages persuade. Make the funnel open if shoppers can skip steps, such as adding to cart from a category page. GA4’s Monetisation reports also include ready-made Purchase journey and Checkout journey reports.
Segment every step before you conclude anything
A site-wide funnel blends very different shoppers, so break each step down by:
- Device. A checkout that fails on phones shows up as a mobile-only drop between checkout and purchase.
- Traffic source. A new social campaign can drag down the first step with nothing wrong on the site.
- Landing page. Pages with many entrances and few product views are losing shoppers at the door.
- New versus returning. Returning visitors usually convert better, so a shift in the mix moves the blended rate on its own.
Say the rate fell from 2.0% to 1.6% last month. If desktop is unchanged but mobile checkout completion collapsed the week a plugin was updated, you have a bug, not a design problem. If every step held steady within each channel but paid social doubled its share of sessions, the store didn’t get worse; the traffic mix changed.
Watch counts too: a segment with a dozen orders a month swings by chance, so judge it over longer periods.
What usually holds each stage back
Knowing which step leaks most narrows the likely causes. Some, such as speed and trust, cut across every stage, which is why a new look alone rarely fixes conversion: design, content, development and performance have to work together, as our e-commerce website design guide shows.
Visit to product view: the wrong door, or a slow one
Ads that promise one thing and land on another, landing pages that crawl on mobile data, navigation that doesn’t match how shoppers think, and site search that finds nothing for a misspelling. Google rates a Largest Contentful Paint of 2.5 seconds or less as good; how page speed affects conversions explains what slower pages cost.
Check it yourself. Open your top five landing pages on a mid-range phone using mobile data, then run each through PageSpeed Insights.
Product view to cart: unanswered questions
Shoppers leave when they can’t see the total cost, delivery time, returns policy, right size or stock level, or when photos are unclear and variant selectors fiddly on a phone. Our guide to product page design covers what each page needs.
Check it yourself. Sort products by views and compare add-to-cart rates. Start with any product that has plenty of views and an unusually low rate.
Cart to checkout: surprises and second thoughts
Some abandonment is normal: shoppers use the cart to check totals or save items. Baymard Institute’s average of 50 studies puts the share of online shopping carts abandoned before purchase at about 70% (70.22% in its September 2025 update). What isn’t normal is a gap much wider on one device, or one that widened after a change to delivery pricing.
Check it yourself. Can a shopper see an estimated delivery cost and arrival date before reaching the cart?
Checkout to purchase: friction at the last step
In Baymard’s research with US online shoppers, extra costs such as delivery, taxes and fees are the most common reason for abandoning a checkout, ahead of slow delivery, distrust of the site with card details, forced account creation and a checkout that feels too long. Our guide to checkout optimisation covers the fixes.
Check it yourself. Buy something on your own phone with every payment method you offer, and note every pause.
Turn the drop-off into a fix: a repeatable loop
The general process is in our conversion rate optimisation guide. For a store it comes down to five steps, repeated monthly or quarterly:
- Pick the leak worth fixing. The step that loses most shoppers in your biggest segment, not the worst percentage in a tiny one.
- Find out why. Walk the path on a phone, watch a few real shoppers try it, and read customer emails for questions people ask before buying.
- Write a testable hypothesis. “Mobile shoppers don’t see delivery costs until the cart, so an estimate on product pages will raise mobile cart-to-checkout.” Name the change, the metric and the segment.
- Fix or test. Fix bugs and obvious friction straight away. For uncertain changes, run an A/B test if you have enough orders; if not, compare equal runs of full weeks before and after, noting campaigns and seasonality.
- Check whether it worked. Same step, same segment, same definition. Check revenue per session too, and log the result so the next round builds on it.
Frequently asked questions
What is a good e-commerce conversion rate?
One that is rising against your own baseline for a similar traffic mix. Published averages sit between about 1% and 3%, but a store selling high-value items can be healthy below 1%, while one selling everyday products to returning customers may run well above that.
Why does GA4 show fewer purchases than my store?
Mainly visitors who decline analytics cookies, and ad blockers that stop the tag loading. Payment methods that don’t return shoppers to the confirmation page, and orders placed outside the website, add to the gap. Investigate when it changes suddenly, not simply because it exists.
How long should I wait before judging a change?
Cover complete weeks, since weekdays and weekends often behave differently, and wait for a meaningful number of orders in the segment you changed. For smaller stores that usually means several weeks.
What to do next
Start with measurement: check that GA4 records each funnel step and that its purchases track your store’s orders. Then build the funnel, split it by device and traffic source, and pick one leak to fix this month. One well-diagnosed fix, measured properly, tells you more than any benchmark.
Two leaks that affect every stage, slow pages and a poor experience on phones, can be checked from the outside. Our free website audit checks both, so you know whether they belong on your list before you start the funnel work.