| Check | The pattern | Which Shopify report |
|---|---|---|
| 1. Did orders drop too? | The rate fell, orders didn't | Orders over time next to Conversion rate over time |
| 2. Did the traffic mix change? | One source grew and converts near zero | Sessions by referrer, before and after the drop |
| 3. Did something ship? | The drop starts on one date, on every source | Theme and app changes that week, your own phone |
| 4. Is a best seller sold out? | People still land on a product nobody can buy | Sessions by landing page against inventory |
| 5. Is it the season? | Orders fell about as much as last year | Orders, same weeks last year |
| 6. Is checkout failing? | Reached checkout holds, completed checkout falls | Conversion rate breakdown, Abandoned checkouts, Shopify Status |
What to check, in order
Shopify's conversion rate is the share of sessions that ended in a purchase (Shopify Help). So it falls in two very different ways: fewer orders, or more sessions with nothing behind them. The first is a problem. The second often isn't. The first two checks tell you which one you have, and they cost nothing.
1. Did orders drop, or only the rate?
Open Orders over time and Conversion rate over time for the same dates (order reports). If orders held while the rate fell, no buyer left. The session count moved. Three things do that.
Shopify changed how it counts. Between 21 and 23 September 2026, Shopify started counting sessions without a pageview, stopped ending sessions at midnight UTC and began filtering identified bots by default. Shopify says a higher or lower rate after the update "isn't automatically good or bad", and that orders and sales aren't affected (Shopify Help). If your line steps down on those dates, that's the ruler changing. I covered what moved and why in what's a good Shopify conversion rate.
Bots got through. Look for a burst of sessions from a country you don't ship to, one page each, no orders. The "Human or bot session" filter separates them (Shopify changelog).
A tag broke. If the drop shows in GA4 or Meta but not in Shopify's orders, it's measurement. TrackBee, a tracking app, lists a consent banner change, an iOS update and a disabled app embed among the usual causes (TrackBee). Shopify's orders are the referee. If they held, fix the plumbing and leave the store alone.
2. Did the traffic mix change?
Your rate is an average of sources that convert very differently. In Contentsquare's 2026 benchmark (6,500+ sites, Q4 2025), paid traffic converted at 1.83% against 2.66% for organic, and new visitors at 1.7% against 2.9% for returning ones (Contentsquare). Launch a prospecting campaign or have a post take off, and you add a lot of cold sessions at once. The rate falls while the site works exactly as it did last month.
Open Sessions by referrer for the two weeks before the drop and the two weeks after. If one source grew and its visitors barely buy, take it out of the comparison and see whether everything else held.
A single source can be very lopsided. One Meta campaign of ours brought hundreds of cheap visits to anry.io that almost never scrolled or tapped (the numbers); on a store, that's a pile of sessions with no orders, and a lower rate with nothing broken on the site.
3. Did something ship on the site?
When the drop starts on a single date and hits every source at once, look at what changed that week. A theme update or a theme switch. A new app. A price rise. A new shipping rate or a higher free-shipping threshold. Price and shipping changes tend to show up late, as fewer completed checkouts while add to carts hold.
The changes don't need to be big. One merchant on the Shopify Community went from about 10% to 1% within two weeks of moving to the Dawn theme, with traffic steady at around 100 visits a day (thread). Nobody in the thread found the cause.
The fastest test is your own phone, on cellular data: tap one of your ads, add the best seller to cart and pay. The ROAS drop guide has the longer list of site changes that hurt results after the click.
4. Is a best seller sold out?
Sold-out products keep pulling visits. Emails, Meta ads, old links and Google results still send people to the page, and by default Shopify won't let anyone buy a tracked item at zero stock (Shopify Help).
A catch from the same help page: if the only stock sits at a location that doesn't fulfil online orders, the product still shows as sold out online. It looks fine in your inventory and nobody can buy it.
Compare your top pages in Sessions by landing page with what's in stock. Google Shopping stops showing a product marked out of stock (Google Merchant Center Help), but a normal Meta ad pointing at that page keeps spending until you pause it.
5. Is it the season?
Compare the same weeks last year, and compare orders, not the rate. Last October's conversion rate was measured before Shopify's September 2026 change, so the two rates aren't comparable. Shopify's own advice for older periods is to compare orders and sales alongside sessions (Shopify Help). If orders fell by about the same share as they did last year, it's the calendar.
There's also a slow drift underneath. Across Contentsquare's sites, conversion fell 5.1% from Q4 2024 to Q4 2025 (Contentsquare). A small slide that matches the market is the market. A cliff on one date isn't.
6. Is checkout or payment failing?
It's last because it takes the longest to test. It's also the most expensive one to miss. In Conversion rate breakdown, compare the last two steps: how many sessions got to checkout, and how many of those finished it (Shopify Help). If people still reach checkout but fewer finish, something there changed.
- Check Shopify Status for Checkout incidents on the days of the drop.
- Open Orders > Abandoned checkouts and look for a pattern: one country, one shipping method, one payment method (Shopify Help).
- Place a real order with every payment method you offer, on a phone and on a laptop.
Errors cost real orders. In Baymard's survey of US shoppers who abandoned a checkout, 17% left because of site errors or crashes, 10% because their card was declined and 9% because they didn't see enough payment methods (Baymard).
Is it the ads or the site?
Split the drop by source and by landing page. If one source fell and the rest held, the change is in who arrives: an ad, an audience, a placement, a channel. If every source fell on the same day, it's the site, the stock or the checkout. Your ad platform can't settle this for you, because its data stops at the click.
Where Anry fits
Checks 1, 4 and 5 take a few minutes in Shopify. Checks 3 and 6 come down to your own change log, Shopify Status and a test order; Anry doesn't track theme edits, app installs or checkout outages. Check 2 is the slow one, because you're hunting for one segment in a lot of data, and that's where Anry helps. It links every Meta or Google ad click with the session and any order behind it, and shows where money leaks by page, device or ad, with a price on each from your store's own orders instead of the ad platform's count. Deterministic code decides what's a leak. An LLM only drafts the write-up and the suggested fix, and your own agent applies it if you say yes.
If check 1 shows your orders held, you don't need Anry or anything else. Nothing is wrong with the store.
Frequently asked questions
Why did my conversion rate drop when sales went up?
Sessions grew faster than orders. That usually means a new source bringing colder visitors, a bot wave, or Shopify counting more sessions since September 2026. If orders and revenue are up, find which source added the sessions before changing anything on the site.
Did Shopify's September 2026 update lower my conversion rate?
It can. Counting sessions without a pageview pushes the rate down, and filtering bots pushes it up, so it depends on your traffic. Shopify says orders and sales aren't affected, and to treat the numbers after 21 to 23 September as a new baseline.
Can a theme update lower my conversion rate?
Yes. A theme update or an app conflict can break add to cart on mobile, hide a variant or change what happens after the click. Buy something on your own phone after every theme or app change.
Does an out-of-stock product lower my conversion rate?
Yes, when people keep landing on it. The visits count as sessions and nobody can buy. Check your top landing pages against stock, especially products you advertise or email about.
Sources
- Shopify Analytics session measurement update, Shopify Help Center
- Filter out bot traffic in your sessions reports, Shopify Changelog
- Analytics fields, order reports and behavior reports, Shopify Help Center
- Selling products when out of stock and abandoned checkouts, Shopify Help Center
- Shopify Status
- Availability attribute, Google Merchant Center Help
- 2026 Digital Experience Benchmark and its conversions chapter, Contentsquare
- Shopify conversion rate drop? Check your tracking, TrackBee, September 2026
- Cart abandonment rate statistics, Baymard Institute
- Conversion rate drop after changing themes, Shopify Community
See which ad clicks leave your store before they buy.
14-day free trial, no card. Works in Claude, Claude Code and Cursor.