Clickthrough rates don't show you if an ad will work
The final instalment from our Annual Report.
Three weeks ago, we launched our Annual Report, which you can download now. We are publishing key stories from this in the run up to Christmas.
If you’ve been reading this newsletter for a while, you’ll know I’ve been banging this drum since 2023.
Clickthrough rates don’t correlate with CPA.
I wrote about it in March 2023. I wrote about it again in September 2024. Meta showed me their own data proving it back in 2018.
And yet it remains one of the most common ways people judge ads.
So here’s the annual refresh. Fresh data from our 2026 report. Same conclusion. But this year, we found something new.
The data
Across every ad account we manage, we ran the same analysis. CTR against CPA, daily, at the ad level.
The correlation coefficients: 0.13 and -0.02.
Correlation coefficients track the relationship between two variables. A score of 1 means a perfect and positive correlation, i.e. variable A increases and variable B increases in line with it.
-1 means perfect negative, i.e. variable A goes up and variable B goes down.
What this means is:
High CTR doesn’t mean either good or bad CPA
Low CTR doesn’t mean either good or bad CPA
And so if you use CTR to be a proxy of how good a creative is doing, you’re not actually anchoring it in whether it works. And you are just as likely to kill a winner as you are a loser.
High spend and CTRs
As well as the usual CPA analysis, we also wanted to understand relationship with variables and spend.
Now here is where we did find a pattern for the first time.
When you look at the top 10 percentile, the ads there have higher clickthrough rates than those in lower percentiles – and it ladders down.
This means that high clickthrough rate does correlate with spend, but it still doesn’t correlate with purchase.
The spend-CTR correlation is 0.51 – i.e. there’s a strong positive correlation between CTR and spend.
What does this mean?
It means that Meta is spending in to the ads with high CTRs, even though that might not mean they have high purchase rate.
The paradox explained
There’s a few considerations here.
First is that CPA also tends to negatively correlate with spend – i.e. the higher spend is, the higher CPA is as well. This shows that Meta is ultimately a scale platform rather than an efficiency one. We know that CPA goes up over time, and we also know that diminishing returns are real. So this is to be expected to a certain extent.
But it also reveals that Meta still prioritises that data once ads become established.
The majority of ads are fails. And so in those early days, it implies that at low levels of spend, CTR and CPA don’t correlate, while both you the advertiser and the Meta the platform figure out what’s working and what isn’t.
But beyond the initial phase of testing, once you are in scaling mode, CTR does become reinforcing. More clicks leads to more spend, which is how you’ll judge performance.
High CTR, therefore, is perhaps a symptom of a winning ad, not the cause.
Why I keep writing about this
Even today, too many people reference CTR when evaluating whether a creative worked or not. “It’s got high CTR, but low CVR therefore the audience wasn’t right for landing page” – maybe, but also not necessarily.
Predictive models take all sorts of signals – positive and negative – to refine. So if Meta is serving your ads to certain people, it’s learning regardless of whether they convert or not. In early days of an ad launch, that signal will come frome verywhere.
The big issue is when in creative testing CTR is used as a benchmark for success.
The Gen Z factor
One more thing that’s only getting more relevant.
Younger audiences don’t click. For Gen Z, the first action after seeing an ad is often a TikTok search, a Google search, an Instagram search. Not a click.
If you’re optimising for CTR, you’re optimising for a behaviour that’s dying out with each passing cohort.
What’s your biggest report insight?
The report has now been out for a few weeks into the wild – what are your key learnings from it? Comment below. And if you’ve not read our Meta Ads in 2026 Report yet, take a read here.
Josh Lachkovic is the founder of Ballpoint, a growth marketing agency that helps brands scale from £1-10M ARR. Visit Ballpoint to learn more, or subscribe to Early Stage Growth for weekly insights on profitable scaling.
Note: Ballpoint works with brands looking to grow their spend from £30k per month to £300k per month.
Fantastic analysis with real data backing it up. The Gen Z point about search-first behavior is whats gonna change everything soon, I've seen this in my own campaigns where conversions come days after impressions with zero clicks tracked. The 0.51 correlation between CTR and spend basically means the algo is optimzing for engagement theater rather than conversion, which explains why so many campaigns scale into losses even wtih "winning" creatives.