Your ROAS is 4. You need 5. You have tried everything.

You know the position. The campaign works. It has worked for months. But at a return of 4 it is not quite a machine yet, and at 5 it would be.

So you do the things you are supposed to do. New creatives. Tighter product groups. Negative keywords. A different bidding strategy. Then back again, because the old one was better.

None of it is wrong. It is just not where the last stretch is.

Where the money actually goes

Before your bid enters the Shopping auction, a fixed percentage is taken off it. Not from your budget. From the bid itself.

Google describes the mechanism in their own help pages:

“Google Shopping currently ensures profitability by deducting a fixed percentage margin from each merchant bid before entering it into the auction.”

That is under the heading “How do I know how much CSSs bid in the auction on my behalf?” in support.google.com/merchants/answer/7558973. Google confirms the margin exists. They never publish its size. Around 20% is the industry estimate and what we see on accounts.

So if you bid 1.00, roughly 0.80 competes.

The arithmetic, on your own numbers

Say you spend 10,000 a month on Shopping, your cost per click is 5.00 and you convert 2% of clicks at an average order of 1,000.

	Today	Through an independent CSS
Budget	10,000	10,000
Cost per click	5.00	4.00
Clicks	2,000	2,500
Sales at 2%	40	50
Revenue	40,000	50,000
Return	4.0	5.0

The budget did not move. The conversion rate did not move. The average order did not move. The only thing that changed is how much of your bid reaches the auction, and that is worth 25% more clicks for the same money.

Four becomes five. Three becomes 3.75. Four and a half becomes 5.6.

Why this is not a growth hack

It is arithmetic, not a trick. You are already paying for those clicks. Some of what you pay goes to a margin instead of to the auction, and an independent comparison service does not take it.

And that is also why it is the boring answer. There is no new angle to test, no creative to write, no audience to find. It is one setting in your Merchant Center, below the campaigns you have already spent months getting right.

Where it will not hold

We would rather say this than have you find out.

Smart bidding may spend the gain instead of showing it. If your strategy is chasing a target rather than a budget, it will often buy more volume at the same target rather than let your cost per click drop. You still get more, but the number that moves is impressions and clicks, not CPC.

Measure impression share on Shopping, before and after. Not average cost per click. That is the metric that shows what actually changed.

In Performance Max it applies to Shopping inventory only. Not Search text, not Display, not YouTube, not Gmail. And how much of your Performance Max budget goes to Shopping is not visible anywhere, which is why nobody who promises you a precise number can actually calculate it.

The extra clicks are not identical to the ones you already had. You are buying deeper into the same auction, so the marginal click is worth slightly less than your average. The table above assumes it is worth the same. Reality will be a little below it.

What it costs to find out

16.66 € per shop per month on the annual plan. Nothing changes in your advertising: same Merchant Center, same Google Ads account, same campaigns, same bids, same history, no relearning phase. We never get access to your Google Ads account, and we cannot.

You have 60 days to get your money back, for any reason at all.

If your return is already 4, you do not need a new idea. You need the last 25% of your bid.