Ecommerce conversion rate is calculated as purchases ÷ sessions × 100. Shopify is more specific: its online store conversion rate divides sessions containing a purchase by total online store sessions. Those calculations match when each purchasing session contains one purchase and both counts cover the same period and online-store activity. Shopify explains the definition in its behaviour reports documentation.
The arithmetic is straightforward. Working out what belongs in each count takes more care. We recommend Shopify’s standard report as a starting point, with an understanding of why its figures may differ from GA4.
The checks below are ranked by calculation dependency: define the purchase count first, establish the session count, then compare reports and interpret changes. This is a practical checking order, not a ranking based on measured impact.
| Priority | Check | Question to settle |
|---|---|---|
| 1 | Purchase count | Are you counting purchases or sessions containing a purchase? |
| 2 | Session count | What does the report include as a session? |
| 3 | GA4 metric | Have you selected a purchase-specific rate? |
| 4 | Reporting scope | Do dates, channels, filters and funnel settings match? |
| 5 | Comparison period | Does the comparison account for traffic volume, sales and seasonality? |
1. Define purchases before doing the calculation
For Shopify’s online store conversion rate, use:
Purchasing sessions ÷ total online store sessions × 100
A clearly labelled worked example:
- Total online store sessions: 1,000
- Sessions containing at least one purchase: 30
- Calculation: 30 ÷ 1,000 × 100 = 3%
That is an illustrative calculation, not a client result or a target conversion rate. Both counts cover the same reporting period and online store. Source: Shopify’s conversion rate definition.
The distinction between purchases and purchasing sessions matters when someone buys more than once during a session. Two orders placed in one session still represent one purchasing session. Dividing total orders by sessions is a different measure and can produce a different percentage from Shopify’s reported conversion rate. Shopify documents multiple purchases within a session.
If your spreadsheet disagrees with the dashboard, check the definitions before blaming the spreadsheet. It may be doing exactly what you asked.
2. Check what counts as a session
The denominator deserves just as much attention. Shopify’s calculation uses sessions, not unique people, so replacing sessions with a visitor or user count changes the measure. Source: Shopify behaviour reports.
Session measurement can also change within a platform. Shopify documents a rollout on 21–23 September 2026 that allows sessions to continue across midnight UTC, includes some sessions without pageviews and excludes identified bots by default. Shopify says these changes can move conversion rate without changing orders or sales. It recommends using post-update data as a new baseline rather than treating earlier session counts as directly comparable, even with matching bot filters. Source: Shopify’s session measurement update.
You do not need to make session accounting a daily occupation. But if a comparison crosses a measurement change, keep that limitation beside the result. Otherwise, a change in counting can look like a change in customer behaviour.
3. Select the right conversion measure in GA4
For a session-based purchase conversion rate in GA4, use the session key event rate with purchase selected. This measures sessions containing the selected key event divided by total sessions. Google defines session key event rate here.
The event selection matters because a key event can represent an action other than buying. An all-key-events rate therefore answers a broader question than a purchase-only rate. Source: Google’s key event documentation.
Also check the metric name before exporting figures:
| GA4 metric | What it counts |
|---|---|
| Ecommerce purchases | Purchase events |
| Items purchased | Product quantities |
Neither count automatically gives you the number of purchasing sessions. Dividing product quantities by sessions would measure units per session, not purchase conversion rate. Source: Google’s ecommerce metric definitions.
Even session totals can vary by reporting method. Google says standard reports and explorations estimate distinct session counts, which can produce small differences from precise counts of exported records. That does not establish why a particular store’s GA4 and Shopify figures differ. Source: Google’s session documentation.
4. Match the reporting scope and funnel settings
Before comparing percentages, write down the date range, reporting timezone, website or sales-channel scope, selected event and filters for each report. Treat this as a comparison checklist, not evidence that any one setting caused the mismatch.
Shopify’s funnel settings need particular care. An open funnel includes sessions that skip earlier steps. A closed funnel requires shoppers to add to cart, reach checkout and complete checkout in sequence. A purchase that skips an earlier step can therefore be absent from the closed funnel without the order being missing. Source: Shopify behaviour reports.
Shopify also uses total sessions as the denominator for each displayed funnel percentage. Its reached-checkout percentage is not the share of cart sessions that reached checkout. Check the denominator before interpreting a funnel percentage as a step-to-step completion rate. Source: Shopify’s funnel definitions.
5. Give the comparison enough business context
Once the calculation is clear, decide how you will monitor it. Our recommendation is to use weekly, monthly, previous-period and year-on-year comparisons where useful, with the reporting window guided by how many sessions the store receives.
Daily movement needs context. Account for seasonality and sale periods before deciding how much weight to give a rise or fall. A sale period should be labelled as such in the comparison; an ordinary trading week is a different context.
We favour ongoing conversion rate optimisation rather than waiting for a worrying percentage. That can include checking site speed, making navigation and checkout easier, and running A/B tests where there is enough traffic. These are areas to assess, not promises of a particular increase.
Before commissioning CRO work, agree on the report, purchase definition, session definition and comparison period. Then use that shared measure to decide where to investigate and what to test. A percentage is much more useful when everyone knows what went into it.