We'll tell you what we won't do, and why. This is one of them. If conversion tracking isn't installed and proven, we won't switch on a Meta Ads campaign. Not the first day, not after a "we'll do it next month". The conversation either ends with the tracking sorted, or we don't take the work.
The unspoken default
Most agencies run blind because the client asked. The brief is "launch the campaign by Monday". Conversion tracking is "we'll get to it". Meta's algorithm starts optimising for clicks and engagement because that's all it can measure. The campaign reports good-looking numbers — high CTR, low CPM, lots of engagement. The phone never rings.
Three months later the client wonders why ad spend is up and revenue isn't. By that point, the algorithm has trained itself on the wrong signal. The campaign needs more than a fix; it needs a restart.
What "no tracking" costs over six months
For a service business spending $2,000/month on Meta with no tracking:
- The algorithm spend : $12,000 over six months going to clicks, not leads.
- The optimisation cost : every change is a guess. Best-case CPL ends up 60–100% higher than it should be.
- The audience tax : without conversions feeding lookalikes, you can't build the audiences that compound Meta's strength.
- The restart cost : when tracking finally goes live, the account often needs rebuilding from scratch.
Total bleed for one mid-size service business: $8,000–$15,000 of avoidable cost. That's why this isn't a soft preference. It's a hard line.
The three setups we get done in week one
Before a single dollar runs through the account, three things have to be working:
- The Meta Pixel : installed sitewide via Tag Manager, firing PageView on every page and the right event on every conversion action.
- Conversions API (CAPI) : server-side tracking that catches what iOS blocks. Around 25–40% more conversion volume on most accounts.
- Key events configured in Events Manager : Lead, Purchase, Contact mapped to your business outcomes. The wrong event gets optimised for the wrong thing.
Why iOS privacy made this harder, not optional
Apple's ATT framework cut Meta's pixel accuracy by 40% on iOS users. Some agencies took that as a reason to give up on tracking. The opposite is true. CAPI partly closes the iOS gap, but only when it's set up properly with hashed customer data and Enhanced Conversions feeding the right events.
If your current agency hasn't moved you onto CAPI by mid-2026, they're three years behind the platform. That's the same logic we hold to for every paid media account, not just Meta.
The conversation when the client refuses
About one in ten prospects pushes back. "It's too technical." "We'll add it later." "Our developer is too busy." The conversation we have: "We respect that, and we won't take the work without it. Three options — you find the time, we engage a developer (quoted separately), or we shake hands and move on." That's the same logic that runs through how we handle ad spend. The client who fits us is the one who values the discipline as much as the campaign.
The takeaway
Conversion tracking isn't optional in 2026. It's the difference between a paid media program and an expensive guess. If your current Meta account is running without it, the next month is the one to fix it — with or without us. The cost of waiting is real and measurable.