Blog › Campaign Case Studies

From Under KES 10,000 to KES 5.3 Million: A Meta Ads Turnaround

By Joseph Maina · · 6 min read

Makeup brushes, lipstick, mascara and powder compacts laid out on a beige surface

Before we started managing this account, Meta ads were bringing in next to nothing for this business, under KES 10,000 a month in anything you could actually trace back to an ad. Three months later, the same account had produced over KES 5.3 million in tracked revenue. Here's what changed, with the real numbers, not a rounded-up summary.

The starting point: ads running, nothing to show for it

The business, a beauty and skincare retailer in Nairobi, was already boosting posts and running occasional promotions on Meta before we took over the account. The problem wasn't a lack of activity, it was a lack of structure: no consistent conversion tracking, no clear campaign objective, and no real distinction between what was working and what was quietly burning budget.

Attributable monthly revenue from Meta sat under KES 10,000, which for a retail business spending anything meaningful on ads is close to a rounding error. The ads were technically running. They weren't doing a job.

What we actually changed first

The first fix was conversion tracking, properly installing and verifying the Meta Pixel so purchases were actually being recorded against the campaigns that drove them. Without this, every decision after it would have been a guess.

The second was campaign structure. Instead of one broad 'boost everything' approach, we split spend into clearly defined Sales campaigns built around specific offers, restocks, a lipstick-focused push, a mid-year clearance sale, rather than one undifferentiated stream of ads competing with each other for the same budget.

The numbers, over roughly three months

Across the full tracked period, 3 July to 8 October, the account spent KES 129,150 on Meta ads and Meta attributed KES 5,328,094 in revenue to them: 953 purchases at an overall return of 41.3x on ad spend. The sales campaigns on their own returned 43.7x, at KES 127 per purchase and a 4.7% purchase rate on clicks.

Monthly revenue held and grew rather than spiking once and fading: KES 1,375,513 in August and KES 1,518,617 in September, with ROAS moving from 39.1x to 42.2x, then 51.5x across the first eight days of October. Steady month-on-month improvement, not a single lucky campaign, is what built that total.

One caveat worth stating plainly: these are revenue figures as Meta reports them, on a 7-day click and 1-day view window. Platform-reported revenue usually runs higher than what lands in the bank, so I treat it as a consistent yardstick for comparing campaigns, not as an audited sales figure.

What worked best, specifically

The single strongest run in the account was a product-day promotion built around one product range. It returned 76.8x on ad spend, spending just KES 4,607 to generate KES 354,050 in revenue. Tightly scoped, offer-specific campaigns like this consistently outperformed broader, less focused ones.

Short, repeatable restock campaigns, re-announcing a specific product back in stock to an audience already familiar with the brand, became one of the most reliable formats in the account. The three one-week runs returned between 50x and 98x on modest budgets.

What didn't work, stated honestly

Not every campaign performed. A broader video-led sales campaign spent KES 18,077 and returned 15.4x, the weakest result of any Sales campaign in the account, and a cost per purchase nearly three times the account average. We flagged it, reduced its budget, and shifted spend toward the formats already proving themselves.

Four reach and awareness campaigns ran alongside the sales campaigns and spent KES 8,255 between them for just 4 tracked purchases. That was expected given their objective, but it's a reminder that not every campaign in an account should be judged by the same yardstick, covered in more depth in my piece on Meta's Advantage+ automation and how much to trust it.

The actual takeaway

None of this came from a secret targeting trick or a single breakout ad. It came from fixing the measurement first, structuring campaigns around specific, trackable offers second, and then doing the unglamorous work of checking the numbers weekly and cutting what wasn't earning its budget.

The account that was producing under KES 10,000 a month wasn't short on product or audience, it was short on structure. That gap is the same one I see in most Meta accounts I audit, and it's almost always fixable faster than business owners expect.

Related articles