What is a Google Ads attribution window?

It's the time period after a click (or impression) during which Google attributes a conversion to that ad interaction. The default click window is 30 days – any conversion within 30 days of clicking your ad gets attributed to that click.

Should I trust Google Ads ROAS numbers?

Use them directionally – they show which campaigns perform better relative to each other. But don't treat them as absolute truth for total revenue. Compare against your backend revenue for the real picture.

Why your Google Ads ROAS is lying to you

27 March 2026·5 min read

AdRoaster Team Conversational ad analysis

Your Google Ads dashboard says ROAS is 6x. Your finance team says total revenue doesn't support that number. Someone is wrong – and it's not your finance team.

Google Ads ROAS isn't fabricated, but it is constructed from a set of assumptions that almost always favour a higher number. Understanding those assumptions is the first step to making better budget decisions.

The attribution window problem

Google Ads defaults to a 30-day click attribution window. If someone clicks your ad today and buys 29 days later, Google claims that conversion. Even if the customer spent those 29 days researching on your site, reading reviews, and coming back via organic search.

Compare this with Meta's default 7-day click window. The same conversion might show in Google but not in Meta – making Google look disproportionately better.

View-through conversions: the invisible inflator

If you're running Display or YouTube campaigns, Google can count "view-through" conversions – where someone saw your ad (but didn't click) and later converted. The default view-through window is 1 day for Display campaigns.

View-through conversions can be legitimate. Someone sees your brand ad, remembers your name, and searches for you later. But they can also be coincidental – the person was going to buy anyway and happened to see your display ad in passing.

Check how much of your reported conversions are view-through. In some accounts, they make up 20–30% of total conversions. That's a significant chunk of potentially inflated ROAS.

Conversion counting: one vs many

Google Ads lets you choose between counting "one conversion per click" or "every conversion per click". For e-commerce, counting every conversion makes sense – one click can lead to multiple purchases. For lead generation, counting one conversion per click is usually more accurate.

If your conversion action is set to "every" and a customer buys three items in one session, Google reports three conversions from one click. Your ROAS looks three times better than reality.

Smart Bidding and the feedback loop

Google's Smart Bidding algorithms optimise towards the conversion data they receive. If that data is inflated (through generous attribution windows or view-through counting), the algorithm optimises for a phantom goal. It bids more aggressively on audiences that appear to convert well – but "converting well" might just mean "happened to see a display ad and would have bought anyway".

This creates a feedback loop: inflated data → aggressive bidding → higher spend → more inflated data.

How to get closer to the truth

Shorten your attribution window. Try 7-day click for a week and compare the numbers. The drop in reported conversions tells you how much of your ROAS depends on long-tail attribution.
Separate view-through from click-through. Create custom columns in Google Ads to see click-through conversions and view-through conversions separately. Make budget decisions based on click-through only.
Compare against backend revenue. Pull your actual revenue from your e-commerce platform or CRM for the same period. Divide by Google Ads spend. That's your real Google Ads ROAS.
Use blended ROAS. If you're running ads on multiple platforms, calculate blended ROAS across all platforms for the clearest picture.
Run holdout tests. Pause Google Ads in one market and measure the revenue impact. If revenue barely drops, your ads were taking credit for organic conversions.

What Google Ads ROAS is good for

Despite its flaws, Google Ads ROAS is useful for relative comparison. If Campaign A shows 4x ROAS and Campaign B shows 2x, Campaign A is likely performing better – they're both measured with the same methodology. Just don't compare the absolute number against your real revenue and expect it to match.

Key takeaway: Google Ads ROAS tells you which campaigns are performing relative to each other. It doesn't tell you how much money you're actually making. For that, you need an independent source of truth – like your backend revenue, or a tool like AdRoaster that gives you the honest numbers.