Shopify Just Reset Your Conversion Rate. Rebuild the Baseline Before BFCM, Not During It.

Shopify rolled out a change to how Shopify Analytics measures sessions, running from September 21 to 23, 2026. Sessions, conversion rate, add-to-cart rate, reached checkout rate, checkout conversion rate, bounce rate, pageviews per session, average session duration, and online store visitors can all shift. Orders, sales, and customer counts do not change. Shopify’s own guidance is to treat the update as a new baseline for every session-based metric.

The recommendation

Draw a hard line at September 21 in every report, dashboard, and KPI you run off Shopify session data. Do not compare pre and post conversion rate as if they measure the same thing. Use the next nine weeks to build a clean post-change baseline, so that when Black Friday lands on November 27 you are judging performance against real numbers, not a measurement artifact.

If you pay an agency, a CRO vendor, or an internal team against a conversion rate target, reset that target now. Otherwise you will either pay a bonus for a methodology change or fire someone for one.

What actually changed

Three things, per Shopify’s help center:

1. Sessions no longer end at midnight UTC. A session now runs on continued activity and ends after 30 minutes of inactivity. Previously, a visit that crossed midnight UTC was split into two sessions.
2. Some sessions without a pageview now count. The example Shopify gives is a customer who goes straight to checkout from a cart link.
3. Identified bot sessions are filtered out by default. Reports that support the “Human or bot session” filter let you add bots back in. The Home page and Live View filter bots with no option to change it.

Your historical data is not deleted or reprocessed. It just is not measured the same way as what comes after.

Why US stores should pay closer attention

The midnight UTC split sounds like a rounding error. For a US brand it is not. Midnight UTC is 8:00 p.m. Eastern and 5:00 p.m. Pacific during daylight time. That lands right in the evening mobile window where a big share of DTC browsing happens. Every evening shopper whose visit crossed that line was being counted twice. That fix pushes your session count down and your conversion rate up.

The cart-link change pushes the other way. If you run SMS or email flows that deep-link to a prefilled cart or straight to checkout, those visits now count as sessions when they may not have before. Brands heavy on abandoned-cart SMS and one-click reorder links could see sessions rise and conversion rate fall.

Which effect wins depends on your traffic mix. That is exactly why you cannot assume the direction, and why you need to look.

The math on a flat store that suddenly looks better

Take a store doing 200,000 sessions a month and 3,000 orders. Reported conversion rate: 1.50%.

Now assume 12% of those sessions were bots, and another 3% were evening visits double-counted across midnight UTC. Strip those out and you are at 170,000 sessions with the same 3,000 orders. Reported conversion rate: 1.76%.

That is roughly an 18% relative lift in conversion rate. Revenue did not move. Nothing on the site changed. These figures are illustrative, not measured, but the mechanism is straight from Shopify’s documentation.

Now picture what happens on your team the week after:

• A landing page redesign shipped on September 18 gets credit for a conversion bump it did not earn.
• An agency on a CVR-linked bonus hits its target.
• Someone models BFCM revenue on the new conversion rate against last year’s session volume and over-forecasts inventory.

Flip the scenario for a brand with heavy cart-link traffic and you get the mirror image: a conversion rate “drop” that triggers a panicked site overhaul in October, the worst possible month to change your checkout.

What to do this week

1. Annotate September 21 everywhere. Shopify dashboards, Looker or Sheets reports, weekly business reviews. Anyone reading a trend line needs to see the break.
2. Sanity check against orders and revenue. Pull orders, sales, and customer counts for the same date ranges. If session metrics moved and those did not, it is the measurement change, not your store.
3. Reset CVR-based targets and contracts. Any KPI, bonus, or SLA built on Shopify conversion rate needs a new baseline from post-change data. Put it in writing now.
4. Check any test that spans the date. If a running A/B test reads Shopify session data, only trust the post-September 21 window, or restart it. Tools that count their own visitors are less exposed, but confirm with the vendor.
5. Ask your apps. Shopify says apps that pull its session data will see the change too. That includes some reporting, attribution, and CRO tools. One email per vendor: does your data source change on September 21?
6. Fix your BFCM comparison now. Bot classification only applies to sessions from October 7, 2025 onward. BFCM 2025 falls after that date, so you can run both periods with the same bot filter setting. But Shopify is explicit that even with matching filters, pre and post sessions are not directly comparable. For year-over-year BFCM reads, lead with orders, revenue, AOV, and new customer count. Treat conversion rate as a within-season metric only.
7. Headless and Hydrogen stores, check your admin. Shopify says most merchants need no action, but stores with custom headless setups or _shopify_s cookie rotation issues may get a storefront message. If you got one, hand it to your developer this week, not in November.

What not to do

Do not try to back out the change and “adjust” historical numbers with a fudge factor. You do not know your bot share or your cart-link share precisely, and a made-up multiplier creates a second wrong number. Accept the break, lean on orders and revenue across it, and let the new baseline build.

And do not over-read the other direction. Filtering bots out of the default view is a better number. Bot traffic was inflating your denominator and hiding your true conversion rate. Once you have four to six weeks of clean data, you will be making CRO and ad spend decisions on a more honest read than you had before.

The bottom line

This is not a performance change. It is a ruler change, and it landed nine weeks before the most important trading period of the year. The brands that annotate it and re-baseline now will read Q4 correctly. The ones that do not will make October decisions on a number that moved for reasons that have nothing to do with their store.

If you want help rebuilding your Shopify reporting and Q4 KPIs on the new baseline, book a call and we will walk through it with you.