There’s a particular kind of monthly report that looks great and means nothing. Impressions up. Clicks up. Click-through rate holding steady. Cost per click down slightly.
Every number moves in the right direction, and the business hasn’t grown.
This happens because most reporting measures the ad account rather than the business. Those are different things, and the gap between them is where budgets quietly disappear.
Not “how many clicks did we get” but: what did a customer cost, and what is a customer worth?
If you can’t answer both, no amount of campaign optimisation will help — you’re refining a machine without knowing what it’s supposed to produce.
Start here:
Cost per acquisition (CPA). Total spend divided by actual customers, not leads. If your form fills doubled but your customers didn’t, you didn’t improve anything — you just made it easier for the wrong people to fill in a form.
Customer lifetime value (LTV). What an average customer is worth over the whole relationship, not the first transaction. This number is what tells you whether a €200 CPA is a bargain or a disaster.
The ratio between them. If LTV is comfortably above CPA, spend more. If it isn’t, no clever bidding strategy is going to save the campaign — the problem is upstream.
For context on what’s realistic: businesses generate roughly $2 in profit for every $1 spent on Google Ads on average, and paid search captures around 65% of clicks on keywords with clear buying intent. Those are real returns. But they’re averages across well-run accounts, not a guarantee that any given campaign is working.
Bidding on your own brand name without checking. Sometimes necessary, often not. If nobody is bidding against you and you rank first organically, you may be paying for clicks you’d have received free. Test it — pause brand campaigns for two weeks and watch total traffic, not just paid traffic.
Broad match without tight negatives. Broad match will find you volume. It will also find you people searching for free versions, jobs at your company, and your competitors’ products. A negative keyword list is not optional maintenance.
Sending every ad to the homepage. The ad promised something specific. The homepage delivers everything generally. That gap is where conversion rates die. Match the landing page to the promise.
Optimising for the wrong conversion. If your account counts “newsletter signup” as a conversion alongside “requested a quote”, the algorithm will chase whichever is cheaper. It will get very good at generating newsletter signups.
A useful report tells you three things: what a customer cost this month, how that compares to what a customer is worth, and what specifically you’re changing next month as a result.
If your report doesn’t say what’s changing next, it isn’t a report. It’s a receipt.
The reason this matters isn’t philosophical. Once you’re measuring the right thing, decisions get easy — you scale what returns, cut what doesn’t, and stop optimising metrics that were never connected to revenue in the first place.