Why did my ecommerce sales drop? How to find the change behind it

Why did my ecommerce sales drop?

Ecommerce sales drop for one of three reasons: fewer visitors, fewer of them buying, or smaller orders. Every cause you will find, from a paused campaign to an out-of-stock bestseller, moves one of those three numbers first, so start by working out which one fell.

The formula is simple: sales = sessions × conversion rate × average order value. If sales fell 25% and sessions are flat, the problem is on the site, in the offer or in checkout. If sessions fell and conversion rate held, look at where traffic comes from: ads, organic search, email and Shopping listings.

Before you blame the market, check two things that look like a sales drop but are not one: a reporting change and a tracking break. Then line up every change from the days before the drop and test the likeliest one. This article follows that order. When you need to present the result, the guide to explaining a conversion drop to a client covers the conversation itself.

Is the sales drop real, or did tracking or reporting change?

Compare the orders in your store backend with what your analytics tool reports for the same days. If the backend shows the same number of orders and the same revenue, your sales did not drop: your measurement did.

• Platform reporting changes. Shopify changed how Shopify Analytics measures sessions from September 21, 2026. Sessions, conversion rate, add-to-cart rate, checkout conversion rate and other session-based metrics might look different, while orders, sales and customer counts are not changed by the update. Shopify advises treating the numbers after that date as a new baseline (Shopify Changelog).
• GA4 purchase tracking. In GA4, open Reports > Life cycle > Monetization > Transactions and compare the Transaction ID list with your order export for a few days on each side of the drop (Google Analytics Help). Missing IDs point to a purchase event that stopped firing, often after a checkout, theme or consent change. Google's own sample code gives every purchase a distinct transaction_id so Analytics does not deduplicate them (Google for Developers), so check that your site still sends one per order.
• Ad platform conversions. If only Google Ads or Meta Ads report fewer purchases while the backend is stable, the problem is the conversion tag or attribution, not sales. The same logic applies in reverse to duplicate conversions in Google Ads.

Only when the backend confirms fewer orders or lower revenue should you move on to diagnosis. Note the date the drop started as precisely as you can, ideally to the day, because every later step compares changes against it.

Which part of sales fell: traffic, conversion rate or order value?

Split the drop into its three parts for two comparable periods: the week after the drop and the week before it, or the same week last year if you sell seasonally. The part that moved tells you where to look.

Then split the part that fell by channel and by product. A drop concentrated in one channel, such as Google Shopping, or in one product line almost always has a specific cause. A drop spread evenly across every channel and product is more likely seasonal or site-wide. To separate the two, see how to tell seasonality apart from your own changes.

If conversion rate is the part that fell, why conversion rate drops goes deeper into site and checkout causes.

What site, catalog and ad changes usually cut ecommerce sales?

Most sales drops follow a change someone made, or a change a platform made for them. These are the ones to check first.

Catalog and offer

• Bestsellers out of stock, or popular sizes and colours missing.
• Price changes, removed discounts or a higher free-shipping threshold.
• New shipping rates, longer delivery times or a stricter return policy.

Site and checkout

• Theme updates, new apps or scripts that slow pages or break the cart.
• Changes to payment methods or the checkout flow.
• A consent banner or tag change that broke tracking, which looks like a drop in reports but not in the backend.

Traffic sources

• Google Ads edits. Budget, bid strategy or target changes and paused campaigns. In Google Ads, Campaigns > Change history lists edits for the past 2 years (Google Ads Help); the Google Ads change history guide shows how to filter it.
• Merchant Center issues. Go to Products (under Products & store) and open the Needs attention tab. Disapproved products stop showing across Google until the issue is fixed and the product is re-approved; warnings can limit performance and can lead to disapprovals. The "Prioritized fixes" toggle hides issues with little impact, so turn it off to see everything (Merchant Center Help).
• Email and SMS. Fewer sends, a smaller segment or a broken automation can remove a steady share of orders without anyone noticing.

How do I isolate which change caused the sales drop?

Put every change from the 7 to 14 days before the drop on one timeline, rule out tracking first, then test one hypothesis at a time. The change that lines up with the start date, and with the channel or product that fell, is your lead suspect.

1. Confirm the drop in the backend
2. Find the part that fell
3. Line up every change
4. Test one hypothesis

Collect changes from every source: your team's edits in Google Ads and Meta Ads, platform automations, Merchant Center status, theme and app updates, price and stock changes, email sends and external events such as a competitor's sale. With every change on one timeline, a drop that seemed random usually has a clear neighbour: a price increase on Monday, disapproved products on Tuesday, sales down from Wednesday. In Changeline, Google Ads edits are imported automatically and the team logs the rest, so the timeline already exists when someone asks.

To prove it, reverse or test the suspect where you can: restore the old price on part of the range, fix the disapproval, re-enable the paused campaign. If sales recover in the affected channel and not elsewhere, you have your cause. If two changes landed on the same day, undo one at a time.

Example: on a Thursday, the founder of an online coffee shop asks why sales fell 30% that week. The store backend confirms the drop; sessions are down 28% and conversion rate is flat, so traffic is the problem. Organic and email traffic are stable, but Google Shopping clicks halved. Lining up the changes from the week before shows a feed update on Monday that edited 40 products, and Merchant Center lists those products as disapproved from Tuesday. The feed is fixed, the products are re-approved and Shopping traffic returns the following week.

How do I stop the next sales drop from becoming a mystery?

Log changes as they happen, not after the drop. A one-line entry with date, owner, what changed and why takes seconds and saves hours of guessing later.

1. Agree on what counts as a change: prices, stock of key products, shipping, theme, apps, tracking, campaigns, product feeds and emails.
2. Record each one with date, owner, platform and reason. A shared marketing change log works better than notes spread across chats and inboxes.
3. Review sales weekly against the log, split by channel and product, so you catch a drop in days rather than weeks.
4. Keep the history. Ad platform logs don't cover your site, catalog or emails, so year-over-year questions need your own record.

With that record in place, the first question after the next drop, "what changed?", already has an answer. For the paid side, why ROAS drops covers the ad metrics that usually move alongside sales.