Retargeting Strategy: Re-Engage the 97% Who Don't Convert
97% of visitors leave without converting. Retargeting brings them back — but only if your sequence is structured correctly.
Retargeting is the most profitable line in the account and the easiest to ruin. It gets ruined one way: showing the same banner to everyone, forever.
Split by what they did, not by how long ago
A visitor who read the blog and one who abandoned a cart aren't the same person. The segments worth separating:
Viewed a product page, didn't add to cart. They're missing a reason. Show the main benefit or social proof.
Added to cart, didn't buy. Something concrete is blocking them: shipping cost, an uncertainty, a comparison in progress. Address exactly that.
Bought. Remove them from acquisition campaigns immediately. Every impression is wasted money and, worse, an irritant.
Read content, never reached a product. It's early. Don't ask for the sale, give them the next step.
How often and for how long
Cap frequency. Three or four impressions a week is enough. Above that, click-through falls and brand perception degrades.
The window matters. For e-commerce: 7–14 days for abandoned carts, 30 for product viewers. For B2B with long cycles: 60–90 days.
After the window, stop. Someone followed for three months over a £40 product isn't going to buy, they're just going to get annoyed.
What to show
Not the same creative as acquisition. They've already seen the introduction — repeating it says you have nothing new.
What works on the second touch: reviews, an answer to a common objection, a comparison with the alternative, or simply the exact product they looked at.
What changed technically
Cookie windows have shortened and pixel-based audiences have shrunk considerably. What compensates: your own email lists uploaded as audiences, server-side tracking, and offline conversions sent back.
The trap in the report
Retargeting always looks good in reports, because last-click credit goes to people who were already close to buying. The right question isn't "what's the return", it's "how many of them would have bought anyway".
You only find that out by switching it off for a week and looking at total sales.
How many hours a week do you lose to repetitive work?
In 20 minutes on Zoom we work out which of them can run on their own. You leave with the answer, whether we work together or not.
Free call, no obligations. You leave with a concrete list of problems and automation opportunities, whether or not you choose to work with us.