If sessions, conversion rate, or reached checkout rate moved on September 21, Shopify changed how Analytics counts storefront visits. The changelog is about sessions. It does not rewrite checkouts, orders, or the abandoned checkout list.
Shopify’s line is the one to keep: this does not necessarily mean your traffic, orders, or customer behavior changed. Orders, sales, and customer counts are outside the update. When a report crosses September 21, 2026, treat the session numbers as a new baseline.
What the session update actually changes
Starting September 21, 2026, the default sessions experience focuses on human visitors where the Human or bot filter is available. Shopify is also classifying bot and system traffic more consistently, and keeping valid shopper journeys visible even when the visit never included a standard pageview.
A session is a storefront visit. Every metric Shopify listed is a session metric:
- Sessions
- Conversion rate
- Add-to-cart rate
- Reached checkout rate
- Checkout conversion rate
- Bounce rate
- Pageviews per session
- Online store visitors
- Searches
Reached checkout rate and checkout conversion rate are funnel rates on those visits. They describe how sessions move through checkout inside Analytics. They are not the row count under Orders → Abandoned checkouts, and they are not your order count.
The move can go either way. Focusing the default view on human visitors can shrink the session count that bot and system traffic was padding. Counting real journeys that never fired a normal pageview can add visits that were missing before. Read a chart that crosses September 21 as two periods.
The filter itself is older. Shopify added Human or bot session in October 2025 as an optional lens on sessions reports. This update is the default sessions experience leaning on that classification, plus tighter bot and system labeling. The help article still limits the classifier: sessions-related metrics only, new sessions from October 7, 2025, and not on Headless or Hydrogen. The changelog’s own limit is the same idea: “where the Human or bot filter is available.”
Sessions are not checkouts
A cleaner session report can sit next to a dirty checkout list. Those are different records.
A bot session is a storefront visit Shopify labeled Bot: a crawler, a scraper, a monitor that loaded the theme. It can inflate sessions and flatten conversion rate, then leave. It never has to open a checkout, create a customer, or land in Klaviyo.
A bot checkout is a checkout a script created. Many of those never render the theme, so there is no storefront session for Analytics to classify. Others look Human because the client was a real browser. You still get a row in Orders → Abandoned checkouts, a customer profile, and a recovery send. A bot session and a bot abandoned checkout are separate events. One can exist without the other.
This September update stays on the session side. It does not score a checkout, delete an abandoned checkout, tag a customer, or keep a junk address out of an email platform. Shopify already said orders, sales, and customer counts stay put. Abandoned checkouts belong with that group. They are created by a checkout, not by a session label.
Shopify’s June 2026 abandoned-checkout update is the checkout-side change, and it only covers card-testing bots that probe payment and never complete it. Scripts that submit a name and email and leave are still abandoned checkouts. The September session baseline does not extend that filter.
So three things can be true on the same store: session charts look different after September 21, orders are flat, and Abandoned checkouts is still full of fabricated names. The first is the new baseline. The third is a checkout problem the baseline does not touch.
What to use this for
Use the new baseline when the question is how many real visits you had, and what those visits did. Mark September 21 on any dashboard that charts sessions, conversion rate, add-to-cart rate, bounce rate, or online store visitors. A move in those charts is not a move in orders.
Use the checkout list when the question is whether a specific checkout came from a person. Fake abandoned checkouts still show up as made-up names, disposable emails, and direct cart hits. A session label does not remove the customer profile or the recovery email that already synced.
CartWatch is for that checkout question. It scores the checkout Shopify already wrote. It does not replace Analytics, and it does not change how Shopify counts sessions. Detection runs on every plan. On Starter and above it tags likely bots and keeps them out of Klaviyo, Omnisend, Mailchimp, Drip, and Brevo.
If the session report looks cleaner and Abandoned checkouts does not, that split is what this update does. Install CartWatch from the Shopify App Store when the list you care about is the checkout list.
FAQ
Did the September 21 update change my orders or sales? No. Shopify says orders, sales, and customer counts are not changed by this session measurement update. The metrics that can look different are the session-based ones in the list above.
Why did conversion rate or reached checkout rate move? Those rates are built from sessions. The default sessions view now focuses on human visitors where the Human or bot filter is available, bot and system traffic is classified more consistently, and some real journeys without a standard pageview stay in the count. Compare reports across September 21, 2026 as a new baseline.
Does this remove bot abandoned checkouts? No. The update measures Analytics sessions. An abandoned checkout, a customer profile, and an email-platform contact come from a checkout. Filtering or rebaselining sessions does not delete those records.
Is this the same thing as the Human or bot filter? The filter has been on sessions reports since October 2025. On September 21 the default sessions experience started focusing on human visitors where that filter is available.
Does CartWatch change Shopify’s session counts? No. CartWatch scores checkout webhooks after Shopify creates the checkout. Session measurement stays in Shopify Analytics.