Shopping formats now take the large majority of retail search spend: 76% of retail search ad budget, and 85% of clicks on retail ads, go to Shopping rather than to text ads. That concentration means the mechanism those formats use to decide what you match against is worth understanding precisely.
The mechanism is your product feed. Not your landing page, not your ad copy, not your bidding strategy. A query is matched against the attributes in the feed, and a field you never recorded is a set of queries you cannot be matched to.
Where retail search money and clicks actually go
industry reporting on Google Ads retail spendBoth are matched on feed attributes rather than on ad copy or landing page.
View as table
| Value | |
|---|---|
| Retail search spend on Shopping | 76% |
| Retail ad clicks from Shopping | 85% |
The expensive half is the one that looks fine
There are two outcomes when a high-intent query names an attribute your feed does not carry. The first is that you are not matched at all, which costs you the impression and is invisible. The second is that you get matched loosely against a broader query, and you pay for a click from somebody looking for something else.
The second outcome is the problem, and campaign reporting cannot distinguish it from ordinary underperformance. It arrives as a click that did not convert. The natural response is to adjust bids and tighten negatives, neither of which touches the cause.
A missing feed attribute does not look like a data problem in your reporting. It looks like a click that did not convert.
What the platforms say about identifiers
Google reports that retailers who added correct GTINs to their product data saw a 20% increase in clicks on average. Agency analyses of feed optimisation programmes report larger effects, including a roughly 70% ROAS improvement and 4.4% CTR growth in one published case, though those come from practitioners describing their own work and should be read with that in mind.
The point is not the specific numbers. It is that the lever sits upstream of everything a paid media team normally controls.
Disapprovals first, then the quiet gaps
Feed disapprovals and limited-performance warnings should be cleared first, because they are binary and the platform tells you about them. What the platform cannot tell you is which of the attributes it permits to be empty are the ones buyers in your category actually decide on.
A feed can be fully approved, carry no warnings, and still be unable to answer the queries worth winning. Ranking the optional fields by what they would be worth, rather than by whether they are flagged, is the part nobody's dashboard does.