Meta advertising in Dubai, where creative is the targeting

AED 4,500 to AED 20,000 covers most of what we are asked to build. The number moves on how many markets and languages, whether creative production is included and at what volume, whether a product catalogue is involved, and whether Conversions API work is needed. Media spend is separate and paid to Meta. Anything quoted before those are known is a guess. We scope first, then price, and the scope document is yours whether or not you proceed. These are market ranges rather than a fixed rate card — the proposal carries the real figure.

Two separate numbers that should never be quoted as one. Media spend goes to Meta, and for most Dubai SMEs a serious test starts around AED 8,000 to 15,000 a month. Management is a separate fee. Anyone quoting one blended figure is hiding how much of your money actually reaches the auction.

For learning anything reliable, no. Meta's model needs roughly fifty conversions a week per ad set to leave the learning phase and perform predictably. At a cost per lead of AED 50, forty dirhams a day produces under one conversion a day, so the campaign never stabilises and every result you see is noise. A small budget is better spent on one tightly focused campaign than spread across several.

It is enough to check the mechanics work and not enough to judge the channel. You will learn whether the tracking fires, whether the creative gets engagement, and roughly what a click costs. You will not get a reliable cost per acquisition. Plan for AED 8,000 to 15,000 over four to six weeks before drawing a conclusion, and be sceptical of anyone who says otherwise to win the account.

It varies far too much by category, audience and season for a single number to be useful, and it rises sharply during Ramadan and the shopping seasons. More usefully: cost per thousand is the metric to watch least. A campaign with an expensive cost per thousand and strong creative regularly beats a cheap one, because it is reaching people worth reaching.

Yes, and they are bought together in the same platform. We do treat them as different surfaces though: Reels, Stories and the feed each need their own creative treatment rather than one asset resized three ways.

Yes, and on this platform we would argue you should not separate it. Meta's targeting has narrowed while its model has improved, which means the creative now does most of the targeting work. An agency that only buys media is missing the main lever.

On the UAE market, plan for meaningful fatigue within two to four weeks at reasonable spend, because the addressable audience is small enough that frequency climbs fast. That is why we work on a production rhythm rather than delivering a batch and running it until performance collapses.

Broad with strong creative usually wins now, and it is a genuine reversal of what worked five years ago. Meta's model finds the buyers if you give it clean conversion data and enough creative variety. We test it against your existing structure rather than asking you to take it on faith.

Because Meta counts a conversion when someone who saw or clicked an ad converts within its attribution window, including view-through, and it is naturally generous to itself. Your CRM counts what actually landed. Both are useful and neither is a lie. We report the platform number and the CRM number side by side and manage against the second.

It depends entirely on how the form is built. Instant forms with pre-filled fields produce cheap, high-volume, low-intent leads. Adding one or two qualifying questions raises cost per lead and usually lowers cost per customer substantially. We test the trade rather than assuming which side you want.

Four to six weeks at a serious budget. The first fortnight is the model learning and should not be read as performance. If a campaign is edited constantly it restarts learning each time, which is why some accounts never stabilise at all.

Usually yes, and it is normally the better option because the conversion history and audiences have real value. We audit first and tell you honestly if a rebuild is warranted rather than defaulting to one.

Yes, natively rather than by translation, with separate creative testing for each language. Translated ad copy underperforms and a native speaker notices immediately.

Advantage+ is Meta's automated campaign type, where you hand targeting, placement and budget distribution to the model and supply creative. It performs well when you have clean conversion data and enough of it, and badly when you do not, because it will spend confidently while learning from bad signal. The sequence that works is to fix event quality first, run a manual structure until conversion volume is healthy, then test Advantage+ against it rather than switching wholesale.

For a serious account on the UAE market, plan on somewhere between eight and twenty new assets a month, most of them variations rather than entirely new concepts. That sounds like a lot until you look at frequency: a small addressable audience means the same people see your ads repeatedly, and performance decays with each exposure. The production rhythm is the campaign.

For anything with a considered purchase, a landing page, because you need somewhere the argument can be made and the conversion measured. Sending paid traffic to a profile means you are paying for a follower who may never see another post, since organic reach on a business profile is a fraction of your following. Profile traffic makes sense for genuine brand-building campaigns and rarely for performance.

It varies enormously by category and we would be misleading you to quote one number. As a shape rather than a promise: consumer services in competitive categories such as aesthetics, fitness and property tend to sit in the tens of dirhams for a raw lead and climb sharply once you qualify. High-value B2B is materially higher. The useful figure is your own cost per closed customer, and until you have that, cost per lead is a vanity comparison between accounts that count leads differently.

Treat it as one input rather than as the answer. The default seven-day click and one-day view window credits Meta with conversions that other channels also touched, which is not dishonest so much as self-interested. The useful discipline is to compare the platform's reported conversions against what your CRM records and against total business, and to run an occasional holdout test where budget allows. If Meta claims more conversions than your business had in total, the answer is not to argue about windows.

In rough order of how often we find it: somebody edited a campaign and reset the learning phase, creative fatigued and frequency climbed, the pixel broke after a site release, a competitor entered the auction, or a seasonal shift moved the audience. The first two account for most cases. It is worth checking the change log and the frequency chart before concluding the platform stopped working.