Average order value dropped? How to find the change behind it

Why did my average order value drop?

Your average order value (AOV) drops when the typical order gets smaller: customers buy fewer items, cheaper items, or pay less because of a discount. Sometimes buying behaviour did not change at all, and the number moved because the definition, the tracking or the mix of customers changed.

AOV is revenue divided by the number of orders, so it can move from either side. A promotion that adds many small orders can lower AOV while total sales grow. A tracking bug that splits one order into two can halve it overnight. Before you react, answer three questions in this order:

1. Is the drop real, or did the way AOV is measured change?
2. Which part of the order got smaller: items per order, price per item or discount per order?
3. Which change, made by your team, a platform or the market, lines up with the first day of the drop?

This article goes through each question. If total revenue fell too, start with the wider check in why ecommerce sales drop, then come back here for the order value part.

Is the AOV drop real, or did the definition or tracking change?

Check measurement first, because a definition or tracking change can create a drop while customers spend exactly what they spent before. Compare the same days in your store backend and in GA4, and look for the date when the two started to disagree.

Tools calculate order value differently. Shopify changed its AOV definition in January 2023 to gross sales minus discounts, excluding adjustments made after the order was created, and applied it to historical data as well. Shopify's help also notes that its Orders over time report shows an order edited after the day it was placed as a separate order. In GA4, purchase revenue comes from the value parameter of the purchase event; in Google's own example, tax and shipping are sent as separate parameters.

If the backend order export shows the same drop as your analytics, the drop is real. If only one system shows it, you are looking at a measurement change, and you should fix that before you touch campaigns or prices.

Which part of the order got smaller: items, prices or discounts?

Split AOV into its parts before you look for a cause. An average order is roughly the number of items per order times the average price paid per item, minus the discount per order, and each part points to a different kind of change.

Then split the same numbers by channel, campaign, device, country and new versus returning customers. A drop that sits in one channel or one campaign usually has its cause in that channel. A drop spread evenly across every channel points to the site, the catalog, prices or a sitewide promotion.

Look at the distribution, not only the average. A few very large orders last month can make this month look worse even if the typical order is the same. Compare the median order value too: if the median held and only the average fell, check for missing large orders (wholesale, B2B, one big customer) before you hunt for a change.

Which changes usually lower average order value?

Most AOV drops trace back to an offer, pricing, catalog or targeting change in the days before the drop. Go through these groups and write down each change with its date and owner.

Offers and pricing

• A new discount code, sitewide sale or automatic discount, especially one without a minimum order value.
• A lower free-shipping threshold, so customers no longer add an item to reach it.
• Removed bundles, quantity breaks or multi-buy offers.
• Price cuts on best-sellers or the launch of a cheaper product line.

Site and catalog

• Cross-sell or upsell blocks removed in a theme or app update.
• High-price best-sellers out of stock or hidden from collections.
• A new cart or checkout flow that skips a recommendation step.

Campaigns and targeting

• Budget moved to campaigns that sell entry-level products.
• A lower target ROAS. Google says lowering the target lets the strategy enter more auctions to grow conversion volume, so it can bring in more orders, some of them smaller.
• Broader audiences, a new market or a prospecting push. Check whether first orders from these buyers are smaller than orders from returning customers.
• Feed or listing group changes in Shopping and Performance Max that shift spend to cheaper items.

Outside your accounts

• A competitor sale, the end of a season, or the end of gift-buying weeks.

Keep the platforms' own changes on the list. Auto-applied recommendations, bid strategy changes and feed updates count as changes even if nobody on your team clicked a button. In Google Ads you find them under Campaigns > Change history, which shows the past two years. The Google Ads change history guide explains how to filter it.

How do I isolate which change caused the AOV drop?

Put every change from the two weeks before the drop on one timeline, rule out tracking, then test one hypothesis at a time against the part of AOV that fell. The cause is usually the change whose date and scope both match the drop.

1. Rule out tracking
2. Split AOV into its parts
3. Line up the changes
4. Test one hypothesis

Start from the first day of the drop, not the day someone noticed it. Collect changes from every source: your team's edits in Google Ads and Meta Ads, platform automations, store changes (discounts, shipping rules, theme and app updates, prices, stock), tracking releases and outside events. Then match each change against the scope of the drop: a code sent only by email cannot explain a drop in paid search orders.

With every change on one timeline, the answer is often visible at a glance: the free-shipping threshold was lowered the same morning AOV started to slide, two days before the bid change everyone suspected. A shared marketing change log turns this into a short check instead of an afternoon spent in admin panels. In Changeline, Google Ads edits can be imported automatically, and the team logs store, offer and tracking changes next to them.

Example: on Monday, the owner of a home-goods store asks why average order value fell from €82 to €68 last week while orders went up. The agency lines up the changes from the ten days before: a 15% sitewide code went live on Wednesday, the free-shipping threshold was cut from €60 to €40 the same day, and target ROAS on the Shopping campaign was lowered on Thursday. Splitting orders by code shows coded orders averaging €55, while orders without a code held at €81. The code, not the bid change, explains most of the drop.

When two changes fall on the same day, look for a segment that only one of them touched: orders with and without a code, one country, one campaign. The change whose segment holds the drop becomes your lead hypothesis.

How do I confirm the cause before I tell the client?

Confirm it by showing that the drop sits only where the change applied and that the timing matches. Then, where the business allows it, reverse or narrow the change and watch whether AOV comes back.

1. Compare affected and unaffected segments over the same days: coded and non-coded orders, the changed campaign and the others, one market and the rest.
2. Check the timing: the drop starts on or after the change date, never before it.
3. Check the size: multiply the share of affected orders by how much smaller they are, and compare that with the total fall in AOV.
4. Reverse or narrow the change: restore the shipping threshold, add a minimum order to the code, or raise the target ROAS back.
5. Write it down: what changed, when, who changed it, the evidence and the decision.

Report the trade-off, not only the drop. A promotion that cut AOV by 15% but brought, for example, 30% more orders may still have been a good decision; the client needs both numbers and the margin to judge it. If you have to explain the drop in a meeting, the guide to explaining a conversion drop to a client has a structure that works for AOV too.

Then keep logging changes as they happen, in a marketing change log, so the next drop takes minutes to explain instead of days.