Why our best-performing clients stopped optimising purely for conversions
REPORT LAUNCH: The Future of Meta is Full Funnel
Over the last 18 months, we have introduced awareness spend and upper funnel thinking into some of our clients strategies. We did it because every one of these clients was at a ceiling where if we pushed spend, CPA went up.
The results have been pretty powerful – and also for me, incredibly surprising.
I used to be pretty cynical on awareness spend.
I started my career very much as a ‘performance marketer’ (although we didn’t call it that then). I was someone obsessed with numbers and deterministic thinking. I followed the David Ogilvy view that if it doesn’t sell, it isn’t advertising. And none of the businesses I worked with or for had the luxury of investing today to make money tomorrow.
And so I’ve been really excited to share this new report with you. I’ve spent the last six months working on this report, and the last 18 months experimenting with this strategy.
It’s also the first report we’ve published at Ballpoint so far, that has been stylistically much closer to this Substack.
Today’s post is a summary of those findings with links to the full report at the bottom.
If you find this interesting, then I would really love for you to forward this email on to one or two people who you think might find it relevant.
And if you want to discuss this stuff, either leave a comment below or say hi over email.
I hope you enjoy
Josh
The ceiling
There is a moment in the life of every growth-stage brand when the maths stops working. You had a great run. You got Meta working, you knew your customer, you scaled. And for months, maybe years, that worked.
Then CPA started to creep. You tried pushing and it went to £45, then £52, then £61. Each time you pulled back, adjusted and pushed again to see the same result. You told yourself it was seasonality. You questioned what updates Meta had pushed live. You convinced yourself it was because of the offer your competitors were running.
If you’re the founder, this is the moment your board conversation changes. Six months ago you were showing a chart that goes up and to the right. Now you’re explaining that CPA is pushing up as you scale. You mention creative diversity or audience saturation. Phrases that sound like reasons but feel like excuses.
So you tuned harder. You tested more creative. Eight concept tests becomes 10. I heard of one brand go from 100 ads to 500 in one month to try to combat the issue.
A new angle might bring CPA down for a week, but then you’re back. You restructured campaigns. You diversified channels and started testing on TikTok and Google and influencer. Each takes time to learn, and all the while Meta’s performance continues to decline.
None of that is wrong. Creative velocity matters. Campaign structure matters. Diversification matters. The problem, we hypothesised, is you’ve reached the ceiling of what performance-only can deliver.
Why? Two things.
The ROAS trap.
Brands heavily prioritise ROAS or CPA. But Meta is a scale channel, not an efficiency one. The question should not be “how do we improve ROAS” but “how do we maintain this efficiency and scale two-to-three times?” Optimising towards ROAS encourages you to optimise towards customers already in market. You can’t scale infinitely by getting better at converting existing demand. At some point, you need to create new demand.
The audience you’ve exhausted.
Everything in your business goes through diminishing returns. Every product, market, channel, campaign, ad. Conversions-campaigns are no different, and what we were seeing is a flatlining of this curve: spend goes up and so does CPA for little marginal gain.
The research was already there
All of the marketing evidence was sitting in front of us.
→ Byron Sharp’s How Brands Grow identified that brands grow not through loyal super-sticky customers, but through high volumes of light-touch customers.
→ The Ehrenberg-Bass Institute created the 95:5 rule: 95% of your audience isn’t in market at any given time. Mental availability is how you reach that 95%.
→ The Long and Short of It suggests that 60% of budget should go to brand.
→ Haus published data from over 640 incrementality experiments across Meta and found that upper funnel has more marginal gain on new customers.
The question I came back to was: will this actually work for a DTC or ecom spending £50k per month on Meta?
What we did
In September 2024, with five clients all experiencing the same pain — reasonable levels of spend but really struggling to scale — we ran an experiment. With three of them, we introduced upper funnel campaign buys. With two, we didn’t.
We committed to 5-10% of spend on upper funnel, with the agreement we’d review it in three-to-six months. We put the line in the dashboard but didn’t talk about it for some time.
We weren’t running lift studies at that point. We were just watching the before and afters. Implied metrics, partial data, and gut instinct.
Today, we’re publishing everything we learned.
What happened
All three brands have seen dramatic increases in scale. This as a starting point was our most important metric: can we increase spend without an increase in CPA?
After introducing awareness campaigns:
Spend scaled 2-5x. Average monthly spend went from the £30-70k range to £300k per month in under a year.
CPA actually improved. The more data Meta had, the better it could optimise. CPA improved until around £125k/month, then stayed stable.
CPMs dropped year over year in Q3, and grew much slower in Q4.
Revenue and CM3 grew for all clients
How we validated it
While the ‘before and after’ and comparisons are useful, they only provide part of the picture. And before/after comparisons are deeply flawed.
Marketing mix models
Across each MMM with each of the full funnel clients, we saw awareness return sit at an average of 1.8x, with the highest of 3x. Importantly, this was marginal and incremental.
Conversion Lift Studies
With the core technique proven out, we’ve been evaluating the right % of spend for awareness.
With one client, we tested an 8% vs a 20% range and saw a 16% improvement in incremental CPA with the 20% level. That client now runs 20% of spend through awareness spend.
What the full report covers
The report goes deep on everything above, plus the playbook:
Budget allocation: how much to commit, and how to phase it in
Campaign setup: starting with Estimated Ad Recall, scaling winners with ThruPlay or Reach
Audience targeting: this was the biggest surprise. In the old days, you ran broad on upper funnel, tight on lower funnel. Now we run broad on conversions and tight on awareness.
Creative strategy – where to start and how we’re experimenting in 2026
Metrics to track in short term and long term
Thank you for reading this through. This has been one of my favourite reports to put together and one I first discussed back in summer of last year as we were in the midsts of this growth.
I’d love to hear your experiences with this. Please leave a comment below, and if you know someone else who will find it useful, then please forward it on to them.