Ecommerce conversion rate benchmarks by device, category and traffic source

Published ecommerce conversion rate benchmarks cluster around 2 to 3 percent, with figures such as 2.26% from IRP Commerce and 2.72% from Dynamic Yield in Blend Commerce's roundup. The typical store converts less: Shogun's study of 745 Shopify stores found a 1.74% median, and Littledata's Shopify median is 1.3%. Compare against a median, on the same session basis, and within your own category.

What is the average ecommerce conversion rate?

Published averages cluster around 2 to 3 percent. Blend Commerce's roundup puts recent benchmarks around 2 to 3 percent. Each figure comes from a different group of stores, so read every number together with its sample.

Source	Sample	Figure	What it measures
IRP Commerce	July 2026 merchant data	2.26%	Transactions divided by sessions
Dynamic Yield	Rolling 12 months, global	2.72%	Purchases relative to visitors
Shogun	745 active Shopify stores, January to June 2026	2.61% mean, 1.74% median	Store-level rates
Littledata	438 Shopify stores	1.3% median	Sessions with at least one purchase

The first three rows come from Blend Commerce. The last comes from Littledata. That page also puts the top 20% of its stores above 2.7% and the top 10% at 3.6% or more.

Why the mean and median split

Shogun's 2.61% mean and 1.74% median differ, even though both describe the same 745 stores. The median is the middle store. Blend Commerce notes that it is less affected by unusually high conversion rates.

In our reading, a small group of high-converting stores pulls the mean upward. The typical store sits closer to the median. If you want to know where an ordinary store stands, compare against the median.

Check the denominator first

The table mixes sessions and visitors. One returning shopper can generate several sessions, so a visitor-based rate and a session-based rate are not the same measure. IRP divides transactions by sessions, while Dynamic Yield uses visitors. Before you put your number next to any row, confirm that your analytics tool divides by the same thing.

What the gap means in orders

The figures below are example inputs, not a benchmark.

• For example, a store with 20,000 sessions at a 0.02 conversion rate (2%) gets 400 orders.
• In this example, the same 20,000 sessions at Littledata's 1.3% median would produce 260 orders.

We think that difference is why picking the wrong benchmark matters. For example, a 2% store looks average against the published averages, but it sits well above the Shopify medians.

This week, take three steps:

1. Pull your last full month of sessions and orders.
2. Compute your rate on a session basis.
3. Compare it with the median row whose sample is closest to your store.

For page-level comparisons, see our landing page conversion rate benchmarks.

Where each benchmark figure comes from, and what it counts

The benchmark figures quoted earlier in this piece come from datasets that count different things. Its metrics are "calculated directly from trading data within the IRP platform." Contentsquare's 2026 benchmark draws on 99 billion web sessions across 6,500+ websites in 9 industries. Its earlier edition analyzed more than 43 billion sessions across 3590 websites.

What IRP counts

IRP defines conversion rate as Transactions ÷ Sessions × 100, a session conversion rate. Direct visits, organic search and unpaid referrals are all grouped under Direct.

That grouping matters when you compare channels. Your organic search rate set against an IRP Direct figure compares one channel with a bucket of three. IRP also says independent merchants are the area it is primarily interested in. In our reading, that makes it a closer match for a UK or Irish store that does not trade at marketplace scale.

What Contentsquare counts

The 2026 data covers desktop and mobile, with year-on-year comparisons from Q4 2024 to Q4 2025. A blended figure across all nine says little about a store. Filter to retail before you compare.

The earlier edition set a stricter rule for trend figures. A site counted in the year-over-year analysis only if it operated throughout the entire analysis period, October 2022 through December 2023. All other analyses in that edition represent Q4 2023. Q4 includes peak holiday trading, so we would not set a quarterly target from a Q4 benchmark without adjusting for season.

Six-line match checklist

Run your store against a benchmark before you quote it in a plan or a board deck.

1. Region: does the benchmark cover the countries you sell into?
2. Model: business-to-consumer, business-to-business, or both?
3. Unit: is the rate orders per session or buyers per user?
4. Period: which quarter or month, and does it include peak season?
5. Channel rules: which attribution model, and which channels are grouped?
6. Scope: retail only, or blended with other industries?

If two or more lines do not match, treat the benchmark as context, not a target. Track your own rate month on month against your own baseline instead.

What is a good conversion rate for a small online store?

For a small store, the useful comparison is the median and the percentile bands, not the mean. As the earlier section showed, the median sits at the middle store and is less pulled up by a few high converters. A small store is far more likely to resemble that middle store than the blended average.

Sample makeup moves the numbers too. Elogic Commerce notes that "Enterprise-heavy samples report higher rates than samples including early-stage stores." Its figure for established mid-market retailers clusters closer to 2.5 to 3.0%. That describes a different kind of business. A store doing its first year of sales should not treat that range as a floor.

The lines that matter

Line	Figure	Source	What it means for you (our reading)
Bottom 20% of Shopify stores	below 0.5%	Littledata	Check tracking and traffic quality before touching the site
Shogun upper quartile, January to June 2026	3.39%	Blend Commerce	A stretch line for a strong Shopify store
Established mid-market retailers	2.5 to 3.0%	Elogic Commerce	A later-stage reference, not a starting target

Below 0.5% puts a store in the bottom 20% on its sample, so that is the first band to escape.

Category shifts all of this. In Shogun's study, the median was 3.31% for health and wellness and 1.62% for home and garden. Use the category band where one exists.

A worked example

This is an example, not a benchmark. For example, a store with 20,000 sessions a month converting at 2% gets 400 orders.

In this example, that 2% sits above the median and below the top-20% line on Littledata's bands. In our reading, the next target is that top-20% line, not the mean. Moving one band at a time gives a goal the team can test against within a quarter.

Set your target this week

1. Current conversion rate (same definition as the benchmark): ____
2. Benchmark used and its sample: ____
3. Band you sit in: bottom 20% / bottom 20% to median / median to top 20% / top 20% to top 10% / top 10%
4. Category median, if available: ____
5. Next band line = this quarter's target: ____
6. Funnel stage to work on first (product page, cart, checkout): ____

If you sit in the bottom band, fix measurement first. If you sit between the median and the top-20% line, work on the funnel stage where shoppers stop.

How conversion rates differ by category and order value

Category moves the number, but price point may move it more. The two sources below slice it differently, so read them side by side rather than as one scale.

Littledata: medians for Shopify stores, by category

Littledata publishes a median and a top 20% threshold for each category, with the number of stores behind each figure.

Category	Median	Top 20% above	Stores
Beauty & Skincare	2.6%	3.9%	38
Health & Supplements	2.4%	3.7%	34
Food & Beverage	2.2%	3.6%	32
Fashion & Apparel	1.0%	1.8%	74
Home & Furniture	1.0%	1.8%	62

Treat a one-point gap as a direction, not a verdict.

Elogic: ranges compiled from several benchmark sets

Elogic Commerce gives ranges assembled from Dynamic Yield, ConvertCart, Triple Whale and others, and its figures run higher.

Category	Range
Food & Beverage	4.9 to 6.2%
Beauty & Personal Care	4.3 to 4.9%
Fashion & Apparel	2.5 to 3.1%
Home & Furniture	0.9 to 1.4%
Luxury & Jewelry	0.8 to 1.2%

The level differs, but the order holds in both: food and beauty at the top, home goods near the bottom. In our reading, that shared order is the part to plan around. Pick one source and stay with it.

Order value

Elogic's portfolio study of 21 stores found that stores under $60 AOV converted at a 4.63% median, versus 0.95% for stores above $200. AOV is average order value. Elogic argues AOV predicts conversion better than industry does.

The practical step this week: find your own AOV before you choose a category row. For example, a $250 beauty store should not hold itself to the beauty median.

Check the revenue math

IRP Commerce estimates a merchant's online turnover as sessions times conversion rate times AOV. The same formula tests whether a category target is worth chasing.

Revenue = Sessions x Conversion rate x Average order value

For example, 20,000 sessions at a 0.02 conversion rate gives 400 orders. In this example, 400 orders at a $60 AOV gives $24,000 in revenue.

IRP also publishes market-level pages for categories, such as its Health and Wellbeing market page. Use it if you sell in the UK or Ireland and want a category comparison close to home.

Why mobile converts lower than desktop, and when it does not

The two device findings point in different directions, and both can be true for your store. Elogic Commerce puts desktop at roughly 3.2 to 3.9% against 1.8 to 2.8% on mobile, and notes that mobile still drives most traffic. Littledata's Shopify sample reverses the medians: 1.5% on mobile and 1.3% on desktop.

Source	Desktop	Mobile
Elogic Commerce, blended range	3.2 to 3.9%	1.8 to 2.8%
Littledata, Shopify median	1.3%	1.5%
Littledata, top 10% of stores	4.7%	4.3%

In our reading, the median tells you where a typical small Shopify store sits, and Littledata's top 10% tells you where the ceiling is. Desktop's ceiling is higher in Littledata's data. So a store at parity on the median can still be leaving desktop orders on the table, and a store well below the mobile median has a design problem rather than a device problem.

Why mobile loses orders

Baymard Institute explains the core constraint: the lack of screen real estate causes a dramatic loss of page overview for users.

Baymard names the weak areas: Mobile Homepage & Category Navigation: Main Navigation, plus Mobile On-Site Search: Search Autocomplete, Mobile Forms. That gives you a short audit list for this week.

Mobile audit, one phone, 30 minutes
1. Main navigation: can you reach your top three categories in two taps?
2. Search autocomplete: type a partial product name. Do suggestions appear,
   and do they include product names, not only past queries?
3. Forms: complete checkout as a guest. Count fields, check the keyboard
   type for email, phone and postcode, and note any field that resets on error.
4. Repeat steps 1 to 3 on desktop and record where the two differ.

Size the gap before you fix it

If you pull conversion by device from your analytics, use the last 90 days. For example, 20000 mobile sessions at a 0.02 conversion rate is 400 orders, and 400 orders at a $60 average order value is $24,000. Run the same sum for desktop, then decide which device gets the next sprint.

Compare each device to its own benchmark, not to the other device. If your mobile rate sits near Littledata's median and desktop sits well below its top 10%, we think the desktop checkout and product pages deserve the first look.

Traffic source and growth both move the rate

A store's blended conversion rate is an average of its channels, weighted by how much traffic each one sends. Change the mix and the blended number moves, even when no page on the site changed. Elogic Commerce makes the point directly: "Conversion benchmarks should not be interpreted without considering intent distribution, device mix, and acquisition mix." (Elogic)

Growth is the clearest case. Elogic's Q1 2026 portfolio data showed a 22% year-over-year decline in conversion rate alongside a 50% increase in sessions, and Elogic framed this as a scaling pattern rather than a crisis. In our reading, session growth outran the rate decline, so order volume still grew. New traffic from colder channels usually arrives with lower intent, and it pulls the blended rate down.

Content traffic does the same thing. Blend Commerce notes that if blog visitors make up 80% of your traffic and those visits convert poorly, your overall results suffer (Blend Commerce).

Channel ranges to compare against

Channel	Conversion rate range
Email marketing	4.0% to 5.3%
Direct traffic	3.3% to 3.8%
Referral traffic	2.9% to 3.5%
Organic search	2.7% to 3.0%
Paid search	2.0% to 2.5%
Paid social	0.7% to 1.2%

Treat these as loose references. Promodo's own email data puts promo campaigns at an average of 1% and trigger emails at around 2% (Promodo). We think the gap with the table reflects different definitions and samples, so check what each figure counts. For organic traffic, Promodo suggests you "benchmark each segment separately and use your top-performing pages as the reference point."

Break out your own rate first

For example, 20,000 sessions at a 0.02 conversion rate gives 20,000 × 0.02 = 400 orders. In this example, a $60 average order value gives 400 orders × $60 = $24,000 in revenue. That single blended line tells you nothing about which channel produced those orders. Split it before you compare it with any average.

Channel | Sessions | Orders | Conversion rate (Orders / Sessions) | Reference range | Gap
Email   |          |        |                                     |                 |
Direct  |          |        |                                     |                 |
Organic |          |        |                                     |                 |
Paid search |      |        |                                     |                 |
Paid social |      |        |                                     |                 |
Referral |         |        |                                     |                 |
Blended |          |        |                                     | n/a             |

This week:

1. Fill in sessions and orders per channel, using one definition of conversion throughout.
2. Compare each channel only with its own range, never the blended rate with a sitewide average.
3. Note which channels grew in session share. A falling blended rate with rising paid social share is a mix question before it is a site question.

How to calculate your store's conversion rate

The formula is simple. Shopify gives it as Total conversions / Total visits to your website, expressed as a percentage. For a store, a conversion is usually an order. Track signups and other actions separately.

Sessions or users

The denominator changes the answer. Some marketers divide by visitors rather than visits to account for people who come back several times before buying. Shopify notes that it’s referred to as a user-based conversion rate, and that a visit-based conversion rate is the industry standard for conversion rate benchmarks.

Littledata's benchmark counts the share of sessions that included at least one purchase. One shopper with three sessions and one order adds three to the session count but only one to the user count. Stores with many returning browsers will therefore see a session rate below their user rate. Compare your session rate to session benchmarks, and keep any user rate in a separate line.

Template

Period: ____ to ____  (same dates for every line)

Orders (purchases)                 = A
Sessions (visits)                  = B
Users (unique visitors)            = C

Session conversion rate            = A / B x 100
User conversion rate               = A / C x 100
Revenue                            = A x average order value

Check before reporting:
[ ] Bot or test traffic filtered out of B and C
[ ] Orders in analytics match orders in the store admin

Worked example

These are example inputs, not a benchmark. For example, take 20,000 sessions and a session conversion rate of 0.02, or 2%. In this example, 20,000 sessions at 0.02 give 400 orders. In this example, an average order value of 60 means 400 orders bring in 24,000 in revenue.

Now change only the rate. Shopify's point is that if your store can convert at least 2% of website visitors into buyers instead of only 1%, your revenue doubles. In this example, at 1% the same 20,000 sessions give 200 orders, which is half the 400 above. This is why a small change in the rate matters as much as a change in traffic. To model your own numbers, use the growth model template.

Why your measured rate can read low

Two measurement problems push the rate down without any change in buyer behavior.

• Bot sessions inflate the denominator. Littledata says its filtering removes fake sessions so bot traffic doesn't drag the rate down. Without filtering, every fake visit lowers your rate.
• Lost purchase tracking shrinks the numerator. Littledata tracks purchases server-side so that orders lost to ad blockers and iOS privacy are counted. If browser tracking misses orders, the rate reads low.

This week, compare the order count in your analytics with the order count in your store admin for the same month. If analytics shows fewer orders, your reported rate understates the real one by that same gap.

Which step of the funnel to fix first

Our step method is short. Count people at each of five steps, find the step that loses the most of them in absolute terms, and fix that step first. The steps are sessions, product view, add to cart, checkout start and order.

Pull the counts for the same date range from your analytics tool and fill in this template:

Step              Count   Lost at this step        Share kept
Sessions          A       -                        -
Product view      B       A - B                    B / A
Add to cart       C       B - C                    C / B
Checkout start    D       C - D                    D / C
Order             E       D - E                    E / D

Cart abandonment  = 1 - (E / C)
Fix first         = the row with the largest "Lost at this step"

Rank steps by people lost, not by percentage. A step that keeps a low share of a small group can lose fewer buyers than a step that keeps a higher share of a large one.

For an outside reference on the cart steps, Baymard has tracked cart abandonment for 14 years, and the global average "currently sits at 70.19%" (Baymard Institute). Compare your cart abandonment line with it. If yours runs well above that average, we think the add to cart, checkout start and order rows are the place to look first.

The checkout itself is often the problem. Across its testing, Baymard found the design and flow of the checkout frequently to be the sole cause of users abandoning during checkout. Users either left in anger or did not know how to complete one or more fields. That points the first checkout test at form fields and step order, before price or shipping copy.

A worked example

These are example inputs, not a benchmark. For example, a store with 20000 sessions and a 0.02 conversion rate takes 20000 × 0.02 = 400 orders. In this example, a $60 average order value gives 400 orders × $60 = $24,000 in revenue. In this example, the template tells you which of the four drops between those 20000 sessions and 400 orders to work on.

Use device to choose the test

Run the template twice, once for mobile and once for desktop. The mobile weak spots covered earlier (navigation, search and forms) are good first candidates when the mobile rows lose the most people. Expect smaller gains there: Baymard notes that even the stronger mobile patterns it identified are, in the end, still a compromise.

This week

1. Segment the funnel by device and fill in the template for each.
2. Pick the row with the largest absolute loss and write one test for it. Log it in the growth experiment backlog template and score it with an ICE score.
3. Size the test with the A/B test sample size calculator using that step's own count and rate.
4. Launch it, following the steps in how to run a growth experiment.

What this page does not cover

This page does not cover conversion benchmarks for business-to-business stores or for industries outside retail, and it does not explain how to set up tracking in a specific analytics tool. The figures are published benchmarks drawn from different samples, not forecasts for your store, and nothing here is financial or legal advice.