Skip to content
Digital Choice
Performance Marketing· 6 min read

Inside a 16x ROAS month: what we changed in an e-commerce ad account

Between May and August 2024, an Egyptian e-commerce brand turned EGP 263K of Meta ad spend into EGP 4.2M in revenue. Here is exactly what we changed.

Between May and August 2024, one of our e-commerce clients in Egypt spent EGP 262,834 on Meta Ads and got EGP 4,225,454 back in tracked revenue. That is a 16.08x return on ad spend, from 9,465 purchases at an average of EGP 27.77 per purchase.

None of that came from a secret trick. It came from fixing the basics, then doing the boring work every week. Here is what we changed.

1. We cleaned up the account structure

When we started, prospecting, retargeting and old test campaigns were all running side by side, often aimed at the same people. Budgets competed with each other and nobody could say which campaign actually drove sales.

We rebuilt the account around the funnel:

  • Prospecting: broad audiences, purchase-optimised, where most of the budget goes.
  • Retargeting: people who viewed products or added to cart, with a smaller budget and offer-led creatives.
  • Testing: a separate space for new creatives, so experiments never disturb what is already working.

Fewer campaigns meant more data per campaign, and Meta’s delivery system learns faster with more data.

2. We fixed tracking before spending more

A campaign can only optimise for purchases if Meta actually sees the purchases. We checked the pixel and conversion events end to end, removed duplicate events and confirmed that the purchase value matched the order value in the store. Only then did we raise budgets.

3. We treated creative as the main lever

With broad targeting, the creative is the targeting: it decides who stops scrolling. So we tested constantly:

  • New hooks in the first two seconds of every video.
  • Real product shots next to polished ones.
  • Offer angles (bundles, free delivery thresholds) against product-benefit angles.

Winners moved into the main prospecting campaign. Losers were switched off quickly, before they burned budget.

4. We optimised weekly, on the right numbers

Every week we looked at two numbers first: cost per purchase and ROAS. Clicks, likes and reach are useful for diagnosis, but they do not pay the bills. Budgets moved towards ad sets with the best cost per purchase, and away from the rest.

5. We scaled slowly

Doubling a budget overnight usually resets learning and pushes costs up. We raised budgets in steady steps on campaigns that had been stable for several days, which kept the cost per purchase close to EGP 28 even as spend grew.

What you can take from this

Every account is different, and a 16x month depends on the product, the margins and the season. But the order of work is always the same: structure, tracking, creative, weekly optimisation, then scale. If any of those is missing, more budget only makes the problem bigger.

Want us to look at your ad account? Book a free consultation call and we will tell you honestly what we would change first.