Why your Shopify conversion rate changed in September

Open Shopify Analytics this week and your conversion rate may look different from last week. Nobody changed anything on the store. So was it the new banner? An app update? A bad day?

Most likely, none of those. On the 21st of September 2026, Shopify changed how Analytics measures sessions. It now focuses on human visitors and classifies bot traffic better. Session-based metrics, including conversion rate, will show different values from that date. Orders, sales and customer counts haven't changed. Shopify's advice is to treat this as a new baseline.

What did Shopify change?

Shopify set out the update in its changelog entry on session measurement. There are three parts to it.

• Session measurement focuses on human visitors, where the Human or bot filter is available.
• Bot traffic is classified more accurately.
• Valid shopping journeys that have no pageviews are kept.

The aim is a more accurate picture of real visitors. The side effect is that any number with sessions in its calculation now sits on a different footing from before the 21st.

Which numbers moved and which didn't?

This is the part to share with anyone who reads your reports.

These may look different from the 21st of September:

• Sessions and online store visitors.
• Conversion rate and checkout conversion rate.
• Add to cart rate and reached checkout rate.
• Bounce rate and pageviews per session.
• Searches.

These haven't changed:

• Orders.
• Sales.
• Customer counts.

So if revenue and order numbers look normal, the business is trading as it was. The ruler changed, not the store. If orders or sales did move, that's a trading question, and it's worth looking at promotions, stock and traffic sources.

Why comparisons across the date will mislead you

Conversion rate is the share of sessions that end in an order. If the way sessions are counted changes, the rate changes too, even with the same number of orders.

That affects any comparison that straddles the 21st. Week on week, month on month and year on year all mix two ways of counting. A dashboard showing conversion up or down against last year is now partly showing the measurement change.

It also affects tests. If you started an A/B test on a theme change before the 21st and it's still running, be careful reading session-based results across the date. We covered how theme tests work in our post on scheduling and testing theme changes with Rollouts.

How to reset your baseline

Shopify's guidance is direct. Treat the change as a new baseline for session-based metrics. In practice, we'd do this.

1. Note the date in your reports. Add a line or annotation at the 21st of September 2026 so nobody reads across it by mistake.
2. Start a fresh session-based baseline. Use the weeks after the 21st as the new reference point for conversion, add to cart and similar rates.
3. Lean on order and sales metrics for longer views. Revenue, orders and average order value are unaffected, so they still compare cleanly across the date.
4. Check any targets tied to conversion rate. If a team goal or agency KPI uses it, reset the target against the new baseline.
5. Tell the people who read the numbers. A two-line note saves a lot of worried messages.

If you use custom reports, this is also a good moment to tidy them. Our post on using metafields as filters in Shopify Analytics shows how to break results down by product attributes or customer groups. Our guide to measuring cross-border profit by market covers the market view.

What this means for conversion work

Conversion work depends on a trusted baseline. You measure where you are, make a change, and see what moved. A measurement change in the middle of that loop doesn't undo the work. It means the loop needs a new starting point.

On our Growth subscription, every store has a KPI baseline set during onboarding. That includes conversion rate, revenue per session, add to cart rate and mobile and desktop parity. For those stores, the change means marking the 21st of September in reports and resetting the session-based baselines from there. Order and revenue measures carry on as before.

The same applies to any store with its own reporting. Make the reset once, clearly, and the numbers become useful again.

If you report to a board, investors or a parent company, say so before they spot it. A short note explaining that session-based figures changed on the 21st, while orders and sales didn't, keeps the conversation on trading. It's much easier than explaining a sudden jump or dip after the fact.

Where parallel helps

We monitor, improve and report on live Shopify stores every month. When Shopify changes how something is measured, we explain it in plain English and adjust the reporting so decisions stay grounded. See our Shopify support packages.

If your numbers moved and you'd like a second opinion on what's real, book a call and bring your analytics.