Ecommerce conversion rate, visitor conversion and checkout completion use different denominators. A purchase divided by all sessions is not the same measurement as a purchase divided by checkout sessions. Both can be correct, but comparing them as one benchmark produces a misleading performance gap.
Littledata's October 2026 benchmark reports a median session purchase rate of 1.4% across 436 Shopify stores. Its measurement window runs from July 27 to October 24, 2026, and the figures were generated October 28. [1] Keep those dates and the session denominator attached whenever citing the number.
A denominator dictionary prevents the biggest mistake
Shopify describes online store conversion as the share of sessions resulting in a purchase. Its analytics field reference separately defines checkout conversion as completed-checkout sessions relative to sessions reaching checkout. [2] The difference is the population exposed to the final step.
A person can visit more than once, while a session describes a visit under the analytics system's session rules. Google Analytics also distinguishes user and session key-event rates. [3] A purchase event must be the selected outcome if the goal is a purchase comparison; another key event changes the question.
| Metric | Numerator | Denominator | Question |
|---|---|---|---|
| Session purchase conversion | Sessions containing a purchase | All eligible sessions | How often does a visit result in buying? |
| User purchase conversion | Users purchasing | Defined user population | How many people buy during the period? |
| Reached-checkout rate | Sessions reaching checkout | All eligible sessions | How often does traffic progress to checkout? |
| Checkout completion | Sessions completing checkout | Sessions reaching checkout | How often does a checkout session finish? |
Use the exact report label alongside the formula. Visitors, users, customers, orders and purchase sessions are not interchangeable counting units. If a tool cannot supply the denominator needed for the comparison, mark the metric unavailable instead of quietly substituting a nearby column.
The same store can have very different valid rates
Consider a hypothetical store with 10,000 sessions, 8,000 distinct visitors, 600 sessions reaching checkout and 200 sessions completing a purchase. Assume for this example that each purchase comes from a different visitor and that all completed purchases are included in the reached-checkout group.
The session purchase rate is 200 divided by 10,000, or 2%. Visitor purchase conversion is 200 divided by 8,000, or 2.5%. The reached-checkout rate is 600 divided by 10,000, or 6%. Checkout completion is 200 divided by 600, about 33.3%. These are illustrative calculations, not observed ecommerce benchmarks.
A slide that puts 2% and 33.3% in the same comparison column creates a dramatic but meaningless difference. The larger rate starts with shoppers already at checkout. It excludes the earlier sessions that never reached that stage.
The example also shows why a funnel diagnosis should preserve counts. If checkout completion remains stable while session conversion falls, the issue may be earlier in the path or in traffic composition. If more sessions reach checkout but fewer finish, the final-stage experience deserves attention. The ratios tell a more useful story together than either does alone.
The Littledata benchmark describes eligible connected stores
Littledata's source page defines purchase conversion as sessions containing at least one purchase divided by all sessions. It calculates each store's rate and then ranks stores for the median and percentiles. Its October cohort uses Shopify stores connected through Littledata to Google Analytics. [1]
That gives the reference a clear unit and collection window. It does not make the sample a census of Shopify, a random sample of all retailers or a guarantee for a new store. Eligibility rules determine which merchants appear in a provider benchmark, so the population description belongs in any responsible citation.
The point of the median is peer positioning, not a target for every offer. A retailer selling considered, high-priced products and a retailer selling familiar repeat-purchase goods can have different paths to purchase. Segment your own results before deciding that the overall market reference explains an operational problem.
Do not combine the store count with another provider's benchmark to imply a larger common study. The underlying merchants may overlap, and the measurement implementation can differ. More citations do not create a pooled dataset unless someone actually obtains and reconciles the underlying observations.
| Reference | Reported rate | Meaning |
|---|---|---|
| Overall median | 1.4% | Middle store in the eligible overall cohort |
| Overall top 20% threshold | 2.7% | Reported boundary for the highest fifth of stores |
| Overall top 10% threshold | 3.7% | Reported boundary for the highest tenth of stores |
| Mobile median | 1.5% | Device-specific eligible cohort |
| Desktop median | 1.4% | Device-specific eligible cohort |
Source: Littledata's original benchmark. Device rows use their respective eligible subsets, so the mobile and desktop medians should not be averaged to reconstruct the overall median.
A median store is different from pooled traffic
A median of store rates gives each store a place in the ordered distribution. A pooled conversion rate gives more influence to stores with more sessions. Either statistic can be informative, but it answers a different question.
Take a second hypothetical example. A small store has 100 sessions and 10 purchasing sessions, a 10% rate. A larger store has 10,000 sessions and 100 purchasing sessions, a 1% rate. The simple mean of the store rates is 5.5%. Pooled session conversion is 110 purchasing sessions divided by 10,100 sessions, or approximately 1.09%.
The mean describes the average of those two store-level ratios. The pooled result describes conversion across their combined traffic. A publisher who changes a median or store-level mean into a general shopper conversion claim has changed the statistical object.
Shopify's funnel mode changes the path being counted
Shopify's behavior-report documentation distinguishes open and closed funnels. An open funnel can include sessions entering partway through the path, while a closed funnel requires the selected steps in order. [4] Those choices affect how a funnel should be interpreted.
A customer may arrive through a direct checkout route rather than following a product-page-to-cart sequence. If the funnel requires the earlier steps, that visit can be absent from later funnel counts despite a legitimate purchase. Read the report configuration before using its step counts as a complete order ledger.
For troubleshooting, keep an open-path view and a strict-path view separate. The first can help describe overall behavior; the second can help evaluate a selected sequence. Neither should be presented as evidence that excluded sessions never existed.
When a team reports checkout abandonment, ask whether the complement of completion is being calculated from the same checkout population. An abandoned-cart rate from another definition may not be the opposite of your checkout-completion rate. Arithmetic complements only hold when the underlying populations and outcomes match.
Measurement changes can move conversion without moving orders
Shopify documents a session-measurement rollout dated September 21–23, 2026. It describes changes to session handling and default identified-bot filtering, while stating that orders, sales and customer counts are unaffected by that update. [5]
That is an important reason to annotate a before-and-after conversion chart. A different denominator can move the rate even if purchase demand has not changed. It would be premature to credit a page redesign or blame a campaign without checking the measurement boundary.
Preserve the raw counts, traffic filter and reporting configuration alongside every comparison period. If you maintain separate analytics implementations, reconcile their definitions rather than forcing identical totals. Agreement is useful only when the tools are intended to count the same thing.
A merchant benchmark workflow starts with the decision
Choose the question before choosing a number. For acquisition, a session-level purchase rate may help evaluate traffic quality and landing-page performance. For checkout friction, use a checkout cohort. For repeat buying, a customer cohort is usually more directly relevant than a session metric.
Then write a measurement card: source, collection period, numerator, denominator, event definition, cohort and aggregation. Compare your own metric only after those fields are understandable. If a public benchmark leaves an important field unstated, use it as broad context and say what remains unknown.
Finally, connect the rate to commercial outcomes. Higher conversion can accompany smaller baskets or deeper discounts. More traffic can reduce the rate while increasing total orders and contribution. TrueMargin's practical recommendation is to assess conversion beside net revenue and the costs attached to the resulting orders.
A useful next step is to save a baseline report before changing the store. Record the expected mechanism: a faster page might help shoppers reach products; a clearer shipping charge might help checkout completion. A narrower hypothesis makes the following comparison easier to interpret than a broad promise to improve conversion everywhere.
Methodology and limitations
This is an editorial comparison of public metric definitions and one original provider benchmark, accessed November 2, 2026. TrueMargin did not run a new store study. Littledata owns the stated October cohort. Shopify and Google documentation establish their reporting definitions. Both worked examples are hypothetical and their outputs are calculated from the assumptions shown.
No customer-level records were combined, and no causal lift is claimed. The article deliberately avoids a universal ecommerce average. Public references are most useful when readers can see the population, period and counting unit that make the number meaningful.
Frequently asked questions
What is ecommerce conversion rate?
For a session purchase rate, divide sessions with a purchase by all eligible sessions. Check the tool's definition before comparing it with another source.
Is checkout conversion the same as store conversion?
No. Checkout conversion uses sessions reaching checkout as its denominator. Store conversion generally starts with all eligible store sessions.
Why does visitor conversion differ from session conversion?
One person can generate multiple sessions. Using users instead of visits changes the denominator and the question being answered.
Can conversion improve without more sales?
Yes. A smaller measured denominator can increase the rate even when purchase counts stay unchanged. Check counts and measurement changes before claiming growth.
Related research: see how refunds affect advertising revenue reports for a worked application of these measurement distinctions.