Insights
Notes from running
acquisition programmes.
Specific, opinionated observations from operating affiliate and digital campaigns. Where we think the conventional wisdom is wrong, we say so.
Attribution
If the numbers add up, they are wrong
When platform-reported conversions sum to more than your actual customers, at least one channel is being paid for someone else's work. Reconcile to one model, and expect the reconciliation to be uncomfortable.
Lead quality
Cost per lead is the easiest metric to game
Any partner can halve your cost per lead by lowering the quality bar. Unless you track leads to a downstream outcome, you will reward exactly that.
Partners
Undisclosed sub-affiliates are the real risk
The partner you vetted may not be the party promoting you. Requiring disclosure of the full chain is the single highest-value control in an affiliate programme.
Creative
Message match beats clever
The gap between what an ad promises and what the landing page delivers predicts conversion better than most creative variables. Close the gap before testing headlines.
Permission
An unprovable list is a liability
Lead data with no evidenced permission behind it has negative value in a due diligence process. It is worth less than no data, because it carries obligations.
Economics
Scale exposes the model, it does not fix it
If the unit economics do not work at small volume, more spend simply produces the same loss faster. Test the economics before scaling the media.
Longer note
Why "more partners" is usually the wrong answer.
When an affiliate programme plateaus, the instinctive fix is to recruit more partners. It feels like progress, and the partner count is easy to report.
What usually follows is a long tail of low-quality traffic, a rise in undisclosed sub-affiliate activity, more incremental cost with less incremental revenue, and a partner list nobody can vouch for when a compliance question arrives.
The more productive question is why the existing good partners are not doing more. Usually the answer is commercial terms, creative support, or the destination experience. Those are fixable, and fixing them compounds. Adding partners does not.