Thinking · Newsletter
How to prove a new channel adds demand instead of re-counting it
Every brand diversifying off Meta asks whether the new channel brings new customers or just re-labels the old ones. It is answerable with a bake-off, and most teams never run one.
This piece is now part of a longer guide: Landing page vs product page for paid ads: the full guide.
This is the question that stalls channel diversification, and it usually gets settled by argument rather than evidence.
The question, stated properly
Not "did the new channel perform". Did it reach people the existing channel was not reaching. Those are different, and platform-reported ROAS answers neither.
The design that answers it
A bake-off. Run your existing repurposed creative against creative built natively for the new channel, at the same time, on the same channel, under the same conditions.
The logic is simple. If the native creative is merely reaching the same audience the old creative already reached, it performs roughly like the baseline. When it clearly beats the baseline, the channel is reaching people the old creative was not, and that is incremental demand rather than reallocation.
| Repurposed baseline | Native concepts | |
|---|---|---|
| Built for | The original channel | This channel |
| Tells you | Whether the channel works at all | Whether it adds demand |
| Runs | Same window | Same window |
What to hold constant
Budget split, window, audience settings, and the destination. Change only the creative. If you change the landing page at the same time you have learned nothing about either.
We ran this
Three native concepts for HealthyCell against their repurposed Meta baseline. The native set outperformed on both interaction and purchase, which is the evidence for treating the channel as genuine diversification rather than a different way of billing the same customers.
The honest caveat
A bake-off tells you about creative fit. It does not replace a proper incrementality test with a holdout, and anyone telling you it does is overselling. Use it to decide whether to invest further, then run the holdout when the spend justifies it.
Questions people ask
How long? Long enough to clear the learning period and accumulate real purchase volume, not clicks.
How many native concepts? Three is the practical minimum, each testing a distinct hook, so a null result is informative rather than ambiguous.