Are you wasting money on Google Ads products that never convert? (a 5-minute check)

The problem

If you run Shopping or Performance Max campaigns, the number you probably look at most often is the one right at the top — your overall return on ad spend. And when you look at the numbers, everything usually looks fine. The return is roughly where you want it, the conversions are steady, and there is nothing obvious to worry about, so you leave the account to get on with it.

The trouble is that this top-line figure, reassuring as it is, tends to be the number that hides the most, because it is only an average. An account made up of a few products doing very well and a long tail of products doing almost nothing will still, taken as a whole, look perfectly healthy.

So underneath a respectable blended ROAS, it is very common to find that a small number of strong products are quietly carrying the cost of a much larger group — products that take spend month after month and rarely, if ever, return a sale. You are not losing money across the account as a whole. You are losing it in particular places, and making more than enough elsewhere to cover it, which is exactly why it never shows up when you only look at the top-line number.

Most advertisers we talk to are suprised to see how much budget they are actually spending on low or non-converting products.

This check takes about five minutes. It finds the products that are costing you with nothing to show for it — and, just as importantly, it stops you from killing the ones that only look like they are.

Why the overall number hides it

In Shopping and Performance Max, even a well-segmented and carefully prioritised account can hold hundreds or thousands of individual products. However tightly you structure the campaigns, Google still has a lot of say as to how the budget is shared out across those products — leaning spend towards the ones it expects to convert and easing off the rest. In a well-run account that is broadly what you want it to do. The difficulty is that it makes the products which spend without converting very easy to miss: they sit quietly behind the good performance of everything else, and because the interface so often defaults to sorting by conversion value first, you rarely scroll far enough down the list to notice them.

Because it all rolls up into one blended figure, none of it shows unless you go looking. A product with £600 of spend and no conversions does not announce itself. It just sits inside a number that, on the whole, looks acceptable.

The Products view is where the average comes apart into individual products. In most of the accounts we audit, it is a report nobody has opened.

The five-minute check

You are looking for products with meaningful spend and little or nothing to show for it.

1. Go to Campaigns > Products. This works at account level for Shopping and Performance Max, or filter to a single campaign if you want to narrow it down.
2. Set the date range to the last 90 days. Thirty days is too short — individual products are low-volume, and you will not have enough data to trust.
3. Add the columns that matter: Cost, Conversions, Conv. value, and ROAS (or cost per conversion). Drop in Clicks too.
4. Sort by Cost, highest to lowest, so the products taking the most budget sit at the top.
5. Now filter the list down, because on a large account it can run to hundreds or even thousands of products and you do not want to read all of them. Add a filter on Conv. value (revenue) less than £5 — or whatever figure makes sense for your business and your margins. The number itself is not important; the point is to strip out everything that is quietly earning its keep and leave only the products that have taken real money and returned little or nothing. That is the list worth your attention.

What you are left with is a much shorter set of products that have spent money over the last three months and given very little, or nothing, back. It is worth reading through them slowly, because a product with no conversions is not necessarily a product to switch off — as we will come to in a moment.

A tip if the list is still long, or you want to make the case to someone else: rather than reading it in the interface, export it to a spreadsheet using the download button at the top right of the table. In Excel or Google Sheets you can sort and total it properly — group the products that convert against those that do not, and add up the spend sitting in each group. That gives you a single, rather uncomfortable figure: the total amount going, over ninety days, to products that have never returned a sale. That number tends to be far more persuasive than any individual row, whether you are convincing yourself or a client.

One note on data, because it changed recently. Until mid-2026, product-level cost and conversion metrics were patchy — you only saw them for some networks. Since June 2026, Google expanded product reporting to cover all networks in Performance Max, plus Video, App and Demand Gen campaigns. So the picture in the Products tab is more complete now than it was even a few months ago. If you looked at this a year ago and gave up because the data was thin, look again.

Before you cut anything — read this bit

Here is where most people get it wrong, and it is the whole reason this is a five-minute check and not a five-second one.

A product with no conversions in Google Ads is not automatically a product to switch off. The campaign only sees its own slice of the world. It does not see the rest of your business.

“It’s not converting — surely I just turn it off?”

Sometimes, yes. But before you exclude a product, cross-check it against how it performs everywhere else. Ask four questions:

• Does it sell fine off the back of other channels? A product might convert poorly on paid but do well organically, through email, or in store. Open your analytics or your back-end sales and look at that product’s total performance, not just its Ads performance. Killing an ad for a product that sells well elsewhere can quietly cut off the top of that funnel — people who discover it on an ad and buy it later, through a route Google never gets credit for.
• Is it a first purchase that leads to more? Some products lose money on the first sale and make it all back on the second, third and tenth. If a product has strong repeat-purchase or lifetime value, judging it on a single-transaction ROAS will talk you into cutting your best customer-acquisition tool.
• Does it have a long consideration window? Expensive or considered purchases convert weeks after the click. If your conversion window is short, or you are looking at a 90-day range that is mostly recent traffic, the sales may simply not have landed yet.
• Is there actually enough data to judge? A product with 30 clicks and no sale has told you almost nothing. That is not really an underperforming product, it is just a product you do not yet have enough information about. Rather than making a decision on it, the sensible thing is to leave it and keep an eye on it.

If the answer to all four is no — it does not sell elsewhere, there is no repeat value, the window has passed, and it has had plenty of clicks — then you have a genuine underperformer, and you can act.

When it genuinely doesn’t convert

Say you have done the cross-check and the product really is dead weight. The instinct is to exclude it and move on. That does work, but it only treats the symptom rather than the cause. It is worth thirty more seconds to ask why it is not converting, because the answer often applies to more than one product:

• The price is wrong. People clicked, saw you were dearer than the shop next to you, and left. Google is telling you exactly where you are losing on price.
• The landing page or feed is weak. A poor image, a thin product page, or a vague, keyword-stuffed title. Google served it, someone clicked, and the page did not close the deal.
• It is out of stock or barely available. Ads for products you cannot reliably fulfil are pure waste.
• It does not belong in the campaign. Different products with different margins and intent sharing one budget means Google optimises for the average, not for this product.

Fixing one of those can rescue several products at once. Excluding one product fixes only that product.

It may also be worth, when implementing these fixes, to segmented some of the worst products into their own campaign, ad group or asset groups so that you can monitor performance closely.

The bottom line

Run the check. Sort by cost, find the products spending real money for no return, and read the top of the list.

Then — and this is the part that matters — do not act on the Ads data alone. Cross-check each candidate against how it performs across your whole business before you cut it. The goal is to stop paying for products that genuinely go nowhere, without switching off a product that sells perfectly well by a route Google cannot see.

A healthy overall ROAS is not proof that every pound is working. It is an average, and averages are very good at keeping quiet about the places you are losing money. Five minutes in the Products tab is usually enough to hear them.