Retargeting is a rescue, not a strategy.
Retargeting works, but its reported numbers overstate its contribution, because it advertises to people who already visited and some of them would have returned anyway. Treat it as a rescue for a leaky funnel rather than a growth channel, keep it to a small share of spend on most accounts, and test its real effect by turning it off for a period and watching total sales rather than its own reported return. If retargeting is the majority of your budget, you do not have a retargeting strategy. You have a demand problem hidden by good looking reporting.
Why the numbers always look good
Retargeting reaches people who have already visited your site. They know who you are, some of them were mid decision, and a portion were going to come back regardless.
When one of those people converts after seeing a retargeting advertisement, the platform records the sale against it. Some of that credit is real and some of it is bookkeeping, and nothing in the report separates the two.
The test that tells you the truth
Turn it off for a fortnight and watch total sales, not the retargeting campaign's own numbers. If total sales barely move, most of what it was reporting was already going to happen.
That is uncomfortable to run and it is the only honest measurement available without a formal incrementality study. Do it in a period that is otherwise unremarkable, so the comparison is not contaminated by a promotion or a seasonal peak.
It cannot grow a business on its own
Retargeting can only reach people who already arrived. The size of the audience is set entirely by how many new people you brought in, which means it scales with prospecting rather than independently of it.
If retargeting is most of your spend, the account has stopped acquiring and started recycling. The reports will look excellent while the business gets smaller, which is the most convincing failure mode in paid social.
Where it genuinely earns its place
Some situations are exactly what retargeting is for.
- Abandoned baskets and abandoned checkouts, where a specific decision was interrupted.
- Considered purchases with a long decision window, where staying visible for a few weeks is the whole job.
- A specific objection you can answer, such as delivery time or a guarantee, shown to people who saw the product and left.
- Existing customers, for a genuinely relevant next purchase rather than the thing they already bought.
The most common mistake is repetition
Showing the same advertisement to the same small audience for weeks is how retargeting becomes annoying, and annoyance is expensive: engagement falls, delivery costs rise, and brand damage does not appear in any report.
Cap the frequency, exclude people who have already bought, and set a membership window that matches how long your purchase decision actually takes. A thirty day window on a three day decision is four weeks of advertising to people who already decided. If you are unsure where your own ceiling is, the ad frequency and fatigue calculator will show you where results start falling away.
What share of your budget is rescue
Add it up. If retargeting is a large part of the account, the next question is where new people are supposed to come from, and whether anything in the plan answers that.
If the reporting looks strong but the business is not growing, that gap is worth investigating properly rather than quarterly. Send us the account through the enquiry form and we will tell you how much of the reported return is actually new.
Questions we get asked.
Is retargeting worth it?
- Yes, in its place. It rescues interrupted decisions and keeps you visible through long considerations. What it cannot do is create new demand, so treat it as a small share of spend rather than as the strategy.
Why does retargeting always report such good returns?
- Because it advertises to people who already visited, some of whom would have returned anyway. The platform credits those sales to the advertisement, and nothing in the report separates rescued sales from sales that were already coming.
How do I measure retargeting honestly?
- Turn it off for a couple of weeks in an otherwise ordinary period and watch total sales rather than the campaign's own numbers. If the total barely moves, most of what it reported was already going to happen.
What share of budget should retargeting take?
- On most accounts a minority, because its audience size is decided by how many new people prospecting brought in. When it becomes the majority, the account has stopped acquiring and started recycling.
How long should someone stay in a retargeting audience?
- Roughly as long as your purchase decision takes. A thirty day window on a three day decision means weeks of advertising to people who already chose, which raises cost and irritates customers.
