Why Performance Max reports more than you sold

The campaign had its best month on record, with more conversions than ever and a lower cost for each. The owner of the online store opened the order history for the same month. It looked a lot like the month before.

Nobody had made a mistake in the report. The numbers were real.

They just weren't measuring the same thing as the order history.

Performance Max can reach people across Search, YouTube, Maps and more from a single campaign, and for many businesses it does a lot of the work. It also reports its own results generously.

When Performance Max looks better in the report than it feels in the business, the gap is usually in one of five places.

Where the gap usually sits

• It's taking credit for customers who were already on their way.
• It's counting things that aren't sales or enquiries.
• A handful of products are carrying the rest.
• The results arrive as one total, with little sense of where they came from.
• It's competing with the account's other campaigns for the same people.

Customers who were already on their way

Someone hears about a business from a friend, searches its name and clicks the first result. A regular customer comes back to reorder. Both are welcome sales, and neither needed an ad to happen.

Performance Max can show for those searches and those people. When it does, the sale is counted as its result. The report fills with conversions that would mostly have arrived anyway, often at a cost that looks very good.

You can see this by looking at how much of what the campaign reports comes from people searching the business's own name, or from people who were already customers.

Counting things that aren't sales or enquiries

A campaign is judged by what the account counts as a result. If that includes page visits, add-to-carts, or the same sale recorded twice, the total climbs without the business seeing more of what it wanted.

Some results are recorded on Google's own platforms, like someone watching a video and visiting the site later, and are added to the total without being checked against a real order or enquiry.

You can check this by putting the campaign's reported results next to the orders or enquiries that actually came in over the same weeks.

A few products carrying the rest

In a store with hundreds of products, a small number usually sell far better than the others. When they're all grouped together, the campaign can look healthy overall while most of the range takes spend and brings very little back.

Products with missing or wrong details, like an old price or a missing size, can also show less often or not at all, with nothing obvious in the campaign to say so.

The same happens with services. A business that offers five kinds of work can find one of them bringing in almost everything while the other four spend quietly.

You can see this by breaking the results down by product or service, rather than reading the total.

One total, not much detail

Performance Max decides where each ad shows, and much of the detail stays inside the campaign. A good month and a bad month can look alike from the outside, because the total hides which channels, products or searches are doing the work.

That makes it hard to know what's worth keeping and what's being paid for without much coming back. It also means one part can quietly fall away for weeks while the total holds steady.

You can see how much is visible by checking what the reports show by channel, product and search, and how much is left unexplained.

Competing with the rest of the account

When Search or Shopping campaigns run alongside Performance Max, they can end up going after the same people for the same searches. Each one reports the sales it touched, so some sales get counted in more than one place.

Add those reports together and the account appears to have made more sales than the business did. It's the same problem as every platform claiming the same sale, just inside one account.

You can check this by comparing which searches each campaign is showing for.

What a good Shopping campaign looks like covers how the two can work together.

Where to start

The simplest test is the one the store owner did. Put the campaign's best month next to the business's own records for the same month, and see how much of the difference each of these explains. Most of the time, it isn't just one thing.

None of this makes Performance Max a poor choice. It means its results are worth reading against the business, not only against the campaign.

That's also where a review starts.

If you'd like someone to look at yours, you can request a review, or read more about Performance Max campaigns.