How to Read a Google Ads Report Without Being Fooled by Vanity Metrics

Why a Google Ads report can look great and still be losing you money

Most Google Ads reports are built to look good, not to tell you the truth. That’s not necessarily dishonesty — it’s a structural bias. Impressions, clicks and CTR are the easiest numbers to grow, the easiest to put at the top of a dashboard, and the easiest to feel good about in a meeting. None of them tell you whether the campaign made you money.

I’ve looked at accounts where every headline number was trending up — impressions rising, CTR healthy, clicks pouring in — while the business behind the account was barely breaking even on the sales those clicks produced. The report wasn’t lying. It just wasn’t answering the only question that actually matters: is this profitable, and would we have got the sale anyway?

Three metrics that flatter without informing

• Impressions — how many times your ad was shown. Impressions cost nothing to inflate: broader targeting, more Display or YouTube inventory, a wider match type all push this number up, regardless of whether anyone who saw the ad wanted what you sell.
• Clicks — someone clicked. That’s the entire story a click tells you. Without conversion tracking, qualification, or context behind it, a click is just spend leaving your account with a maybe attached.
• CTR in isolation — click-through rate tells you your ad text and audience match reasonably well. It says nothing about lead quality, sale value, or margin. A high CTR on the wrong audience is worse than a modest CTR on genuine buyers.

CPA against a margin-aware target — not CPA against thin air

Cost per acquisition is a genuinely useful number, but only once it’s measured against the right target — and the right target isn’t a round figure someone picked because it felt achievable. It should come from your actual margin: what you can afford to pay for a customer and still make money, given your product margin, average order value, and how much repeat business that customer is likely to bring.

A CPA of £40 can be excellent for a business with a high average order value and healthy margin, and unworkable for a business selling a low-margin product. Reported on its own, both look identical: ‘CPA: £40.’ The number only means something once it’s held against what your business can actually afford to pay for a customer — which is a conversation about your margins, not just your ad account.

The two other numbers worth actually reading

• Incremental conversion volume — not total conversions, but conversions you wouldn’t have got anyway. Branded search terms, remarketing to warm visitors, and automated campaign types can all report conversions that would likely have happened without that ad spend — branded search especially. The real question for any conversion is: did the ad cause this, or just get the credit for it?
• ROAS vs POAS — return on ad spend divides revenue by ad spend; profit on ad spend divides gross profit by ad spend. ROAS treats a sale with 10% margin the same as a sale with 60% margin. If your product mix, discounting, or margins vary by product or campaign, ROAS can show a ‘winning’ campaign that’s actually one of your least profitable.

The 'so what' test — run this on your next report

1. Read the top-line numberWhatever the headline metric is — clicks, impressions, CTR, even total conversions — note it down without judging it yet.
2. Ask 'so what?'What does this number actually mean for revenue or profit? If you can’t answer in one sentence, it’s probably a vanity metric dressed up as a KPI.
3. Ask 'compared to what?'A number with no benchmark is meaningless on its own. CPA needs your margin-aware target; ROAS needs POAS alongside it; conversions need a sense-check against what would likely have happened anyway.
4. Ask 'would this exist without the ad?'For every reported conversion, consider whether it’s genuinely attributable to the ad, or whether it’s a branded search, an existing customer, or someone who was going to buy regardless.
5. Only then look at the trendA trend on a vanity metric — impressions or clicks climbing — is only useful once you know the underlying number is one that actually matters. Otherwise you’re watching noise go up.