Most paid accounts I audit are not badly optimised. They are badly measured, which makes optimisation impossible, because nobody can agree on which campaigns are working.
Platform-reported conversions overlap. Meta claims a sale, Google claims the same sale, TikTok claims it too, and the three numbers add up to more revenue than the business made. Until that is resolved, every budget conversation is a negotiation between three dashboards that each have an incentive to look good.
Where I start
Measurement
One source of truth, GA4 plus your own back end, that all platforms are reconciled against. Server-side tracking where iOS and browser restrictions have made pixel data unreliable.
Unit economics
What a customer is worth, what you can afford to pay for one, and what margin survives. Without this number, ROAS targets are made up.
Channel fit
Which platforms suit your product, price point and buying cycle. A SAR 4,000 B2B service and a SAR 90 impulse product do not belong on the same platforms.
Allocation
Budget split by what the data supports, with an explicit test budget for what it does not yet.
Scale or stop
Monthly review against cost per acquisition, not platform ROAS. An honest call on what to turn off.
The platforms
Google Ads
Search, Shopping, Performance Max and YouTube. Highest-intent traffic available, and the channel where poor account structure wastes money fastest.
Meta Ads
Facebook and Instagram. The broadest reach in Egypt, and still the most reliable prospecting engine for e-commerce across the region.
TikTok Ads
Where attention in Saudi and the UAE has moved. Creative-led, cheap to test, and punishing to anyone who uploads a repurposed Meta video.
Snapchat Ads
Consistently underrated outside the Gulf and consistently dominant inside it. Saudi Arabia has among the highest Snapchat penetration in the world.
X Ads
Narrow but real: B2B, tech, finance and event-driven campaigns in the Gulf. Small budgets, specific use cases, rarely a primary channel.
Budget and blended targets
The number that actually matters is blended cost per acquisition across all channels, not any single platform’s self-reported figure.
How the region changes the answer
- Saudi Arabia: Snapchat and TikTok carry reach that Meta does not, especially outside the 30+ bracket.
- Egypt: Meta remains dominant and cost per result is low enough that testing is cheap, but so is wasting money at scale.
- UAE: the most expensive of the three, the most competitive, and the market where creative quality separates results most sharply.
- Arabic creative: translated ad copy consistently underperforms natively written copy. This is measurable, not a matter of taste.
- Ramadan and Eid: auction dynamics shift enough that annual budgets planned without them are wrong by a wide margin.
What I will not do
- Report platform-attributed ROAS as if it were business ROAS.
- Recommend a channel because it is fashionable rather than because your unit economics support it.
- Scale spend on a campaign whose conversion tracking I have not personally verified.
- Keep a channel running because it is in the retainer. If it is not paying, I will say so.
Frequently asked
What budget do I need to start?
Enough to reach statistical significance before you run out of patience. Practically, that usually means a minimum of two to four weeks of spend at a level that produces at least 30–50 conversions, otherwise you are reading noise. What that costs depends entirely on your cost per acquisition, which is the first thing we work out.
Do you manage the day-to-day, or set it up?
Audit, strategy, tracking and account structure are where I add the most value. Ongoing daily optimisation is often better handled in-house or by a dedicated buyer. I will help you brief them and evaluate their work honestly.
Our ROAS looks great but revenue is flat. Why?
Usually attribution overlap, incrementality, or both. The platforms are claiming credit for sales that would have happened anyway. A geo holdout test or a simple spend-versus-revenue correlation over time settles it quickly, and the answer is often uncomfortable.
Can you run ads in Arabic?
Yes, and this is where a large part of the advantage sits. Arabic creative written natively, per market rather than one generic MSA version, consistently outperforms translated creative on cost per result. I write the briefs and direction and work with native copywriters per dialect.
Should we use Performance Max?
Sometimes. It works well for e-commerce with a clean product feed and enough conversion volume to train on. It works badly for lead generation with weak conversion signals, where it will happily spend your budget on junk leads that look like conversions. The feed and the conversion definition decide it, not the campaign type.