How to Evaluate Campaign Performance
Jun 06, 2026Most people look at the wrong metrics. They see a decent CTR and think the campaign is working. Then they look at conversions and think the landing page is the problem. Then they look at spend and go broke. None of those metrics tell you if you're actually winning. Only one does.
The metric hierarchy
Here's how to evaluate campaign performance, ranked by importance:
One: CAC below target? That's it. Everything else is supporting data. If your CAC is below target, the campaign wins. Full stop. If your CAC is above target, the campaign loses. Full stop.
A PT ran Facebook ads. 8 percent conversion rate on the landing page. Beautiful. But her CAC was $340. Her LTV was $360. She's making $20 per customer. Technically it wins. Practically it's useless. One refund kills a month of profit.
CTR was good. Conversion rate was good. CAC was the problem. Most people would've optimised the landing page. She needed to optimize the offer or positioning.
Two: Conversion rate trend. Is it improving or declining? Week one 6 percent, week two 7 percent, week three 8 percent, week four 8.5 percent. Trend is positive. Your optimisations are working. It's declining? Stop. Diagnose.
Three: Cost per click or impression. This is supporting data. It tells you about platform saturation and audience fatigue. Rising fast? Could be a signal to pause or pivot creative. But it only matters if CAC is moving the same direction.
Four: Everything else. Landing page CTR. Audience size. Impressions. These are nice to know but they're noise if CAC is wrong.
How to diagnose a broken campaign
CAC is above target. Now what? You look at the chain.
Is traffic quality down? Check audience targeting. Are you reaching the right person? If impressions are high but CTR is low, you're reaching the wrong audience. Fix targeting.
Is the landing page not converting? Check conversion rate. If traffic is good but conversion is low, your landing page isn't resonating. Fix the page or the offer.
Is the sales process weak? Check show rate. If people are filling out forms but not showing up, your messaging lied or your qualification was off. Fix the call process.
Is the offer not closing? Check close rate. If shows are good but closes are bad, your offer is the problem. Fix the offer.
You test one link at a time until you find it. Don't change everything at once. You won't know what worked.
The evaluation timeline
Never evaluate at one week. Week one data is random. Evaluate at 30 days. That's four weeks of volume. 20-30 conversions on most channels. Enough for a real trend. Not so much that you've wasted a month of budget.
At 30 days you know if CAC is below target. If it is, you scale. If it's not, you diagnose and test a change. Then run another 30 days. Repeat.
This is why impatience kills campaigns. You evaluate at week two, see noise, panic, change everything. The change wasn't needed. You just killed something that needed four weeks to stabilize.
PS: If you're optimising something other than CAC, you're optimising the wrong metric.