Blog › Meta Ads

Meta Retargeting on a Small Budget: Win Back People Who Almost Bought

By Joseph Maina · · 4 min read

Two women with shopping bags looking at their phones in a clothing shop

Someone watched your video, visited your site, added something to cart, and then their matatu reached the stage and they put their phone away. They were interested. They just got interrupted. Retargeting is how you show up again, and for a business with a small budget, these warm people are often the cheapest sales available.

The audiences worth building

Not all warm people are equally warm. Build them in layers so that your message can match how close they are.

AudienceHow warmWhat to show
Added to cart or started checkout, last 14 daysVery warmReminder, delivery details, a reason to finish now
Visited a product or service page, last 30 daysWarmReviews, results, answers to common objections
Sent you a message but did not buy, last 30 daysWarmThe offer again, plus proof
Watched most of a video or engaged with your Page, last 90 daysLukewarmYour best offer with a clear call to action
Past customersDifferent jobRestocks, new products and complementary items

The small-audience problem nobody warns you about

Retargeting needs enough people to retarget. If your site gets a few hundred visitors a month, a "cart abandoners, last 7 days" audience might be twenty people. Meta will struggle to deliver to it, and the same few people will see your ad again and again.

If that is you, widen it. Combine all your warm groups into one audience: site visitors, video viewers, Page and Instagram engagers, people who messaged you. Run one ad set against that group, then split it as it grows. Also, feed the top of the funnel. You cannot retarget people you never reached, so a reasonable share of your spend has to keep bringing in new people.

Do not show them the same ad again

They have already seen your introduction. They did not buy for a reason, so use retargeting to deal with that reason. These are the objections I would answer first in Kenya:

  • "Will it arrive?" Show delivery areas, times and costs.
  • "Is this real?" Show customer videos, reviews and your physical location.
  • "Can I trust paying first?" Explain pay on delivery or how your till number works.
  • "Is it worth the price?" Compare it with the alternative and show what is included.
  • "Can I wait?" Offer a real deadline if there is one. Never invent fake urgency.

How much to spend

Let audience size decide, not a fixed percentage. Watch your frequency, which is the average number of times each person has seen the ad. If it climbs past about 4 or 5 in a week and results are falling, you are saturating a small group. Lower the budget or change the creative rather than pushing harder.

Be aware too that Meta's automated campaigns, such as Advantage+, already show ads to warm people within their broader delivery. If you run one, exclude recent buyers, and compare results before you add a separate retargeting campaign. You may be paying twice to reach the same person.

Stop before it gets creepy

Remove people who have already bought, or they will keep seeing ads for the thing they own, which is both irritating and wasteful. Avoid ad copy that points out personal facts about them, such as "We saw you looking at weight-loss products". Meta's rules restrict this and customers rightly dislike it.

Retargeting does not fix a weak offer. If nobody buys from your warm audience either, look at the product, the price or the checkout before you spend more on ads.

Start simple: one audience of everyone warm from the last 30 to 90 days, one ad built on proof, and no recent buyers.

Related articles