I plan, build and run paid campaigns on Google, Meta and TikTok for businesses in Dubai and across the GCC. The work is equal parts strategy and daily maintenance: an account structure you can actually read, disciplined creative testing, and conversion tracking accurate enough for the bidding algorithms to learn from.
Most accounts are not broken in some exotic way. They are simply spending in places nobody is watching, on a structure built for a much smaller budget.
Almost every ad account I am asked to look at has the same three things going on at once. Money is going to placements, search terms and audiences nobody deliberately chose. The reporting is producing numbers nobody fully trusts, because the platform, the analytics and the finance spreadsheet never quite agree. And the account is still running on a structure that made sense at a fraction of the budget it now carries.
None of it looks dramatic on any single day, which is exactly why it survives. The waste compounds quietly, month after month, until the channel gets written off as something that does not work for this business — when what did not work was the setup underneath it. A campaign nobody can read is a campaign nobody can diagnose, and an account nobody can diagnose is one where every decision ends up being made on instinct.
Four failure patterns account for most of it. Each one is fixable, and each one is considerably easier to fix once it has been named out loud rather than left as a vague sense that the ads are underperforming.
Broad match soaking up irrelevant searches, placements nobody chose, audiences that overlap each other, and campaigns still live months after the promotion behind them ended.
The ad platform reports one number, analytics reports another, and the finance spreadsheet reports a third. Every performance conversation becomes an argument about whose data to believe.
Something works at a modest daily budget, the budget doubles, and efficiency falls apart. Scaling is a structural problem long before it is a bidding problem.
Creative, audience, landing page, bid strategy and offer all changed in the same week. Without a testing sequence, a good month teaches you nothing you can repeat.
A strategy nobody maintains drifts. Daily fiddling without a strategy is just noise with a login.
Paid advertising rewards two things that rarely sit in the same place: a plan that holds for a quarter, and someone in the account most days. I work across both ends, and I keep the reasoning visible so you always know why something changed and what it was expected to do.
The bias throughout is toward decisions you could audit. Structure is chosen so results stay attributable, tests are queued so one variable moves at a time, and the numbers that judge a campaign are the commercial ones rather than the ones that happen to look best. That occasionally means a slower start — a week spent on tracking before anything scales — but it is the difference between an account that teaches you something and one that only ever produces a monthly surprise.
Every quarter opens with a written plan: which offers get budget, which audiences the plan is trying to reach, what is being tested, and what would make me stop. Then the account is worked on a daily rhythm — pacing, search terms, creative fatigue, anomaly checks.
Each campaign has one job and each ad set answers one question. Prospecting is separated from retargeting, cold audiences do not compete with warm ones, and naming conventions make the account readable by anyone who opens it.
Tests are queued rather than launched all at once, each with a stated hypothesis, a defined budget and a point at which it gets called. The output is a growing list of things the account has proven about your market, not a folder of ads nobody can rank.
Spend is added in increments the algorithm can absorb, and only where there is creative depth and audience headroom behind it. Small budgets get concentrated on one platform and one offer so the data is conclusive.
Impressions, clicks and cost per click are diagnostics, not goals. What decides whether a campaign lives is acquisition cost and return on ad spend, read against your margins. Cheap traffic that never buys is still expensive.
Server-side tagging and the Conversions API send events from your server with deduplication and consistent parameters, so bidding systems optimise on accurate signals rather than a partial picture left over from browser pixels.
The measurement layer underneath all of this is the same one I use on every channel: which numbers define success, and where each one is sourced, is agreed before the work starts rather than afterwards. How an engagement actually runs — cadence, access, ownership of the accounts — is set out in about.
Each platform is good at something different. The mistake is treating them as interchangeable places to put the same ad.
Google is where you capture demand that already exists. Search puts you in front of someone actively looking for what you sell, which is why it usually carries the highest intent in the account. Performance Max extends that across Google inventory using your feed and assets, and needs careful asset-group discipline and exclusions to stay useful rather than opaque. Display is best treated as a retargeting and awareness support layer rather than a primary acquisition channel.
Google Ads scopeMeta is where you create demand rather than wait for it. It suits products that need to be shown, demonstrated or explained, and categories where people are not yet typing your solution into a search bar. Performance here is driven mostly by creative, so the account lives or dies on how many genuinely different angles you can put in front of the algorithm. It is also the strongest environment for retargeting and for lead generation in the UAE, including click-to-WhatsApp journeys.
Meta Ads scopeTikTok rewards content that looks like content, not like an advertisement dropped into a feed. It suits consumer brands with a broad audience and a real supply of native video — creator-style demonstrations, before-and-afters, founder-to-camera explanations. Production cadence matters more here than anywhere else, because creative fatigue arrives faster. It is a strong discovery channel and often works best when the follow-up happens on Meta or in email.
Ask about TikTokSeven areas of work. Which of them leads in a given month depends on what the audit turned up and what the account has been saying since.
Channel mix, budget split, offer selection, and the audience logic behind each campaign — interest and behaviour pools, customer lists, lookalikes, keyword themes and match types, plus the exclusions that stop campaigns bidding against each other.
Concepts, scripts, hooks and headlines written for each platform rather than recycled across all three. I brief the shoot or edit, specify formats and aspect ratios, and keep a running queue of angles so there is always a next test ready when fatigue sets in.
Bid strategy selection and target setting, daily pacing against the monthly commitment, and adjustments for seasonality, promotions and auction pressure — so the month does not end with a third of the budget unspent or burned in the first ten days.
A structured test queue covering creative, audiences, placements, offers and bid strategies, each with a hypothesis and a decision point. Results are documented so the account keeps a memory of what has already been tried and what it showed.
Specific, prioritised notes on the pages your traffic lands on: message match with the ad, load speed on mobile, form friction, trust signals and how the offer is framed. I do not need to rebuild your site to tell you which element is costing you conversions.
Google Tag Manager, GA4 event schema, pixels and datasets, server-side tagging and Conversions API, offline or CRM conversion imports where the sales cycle needs them, plus reconciliation against your own order or lead records.
A monthly written read of what happened and why, what was tested and what it showed, where budget moved, and what the next month is planned around. Written to be read by a business owner, not to fill pages with platform screenshots.
Five phases, with the timings I plan against. They are expectations rather than promises.
An account with existing history and clean tracking moves faster than one being built from nothing — but you will always know which phase the work is in, what is being tested, and what comes next.
I go through the existing accounts, tracking setup and analytics, and sit down with you on the commercial side: margins, average order value or deal size, sales process, capacity, and what an acceptable acquisition cost actually is. The output is a plain list of what is working, what is leaking, and what is measured incorrectly. If you have never run ads before, this stage covers account and tracking readiness instead.
You and I agree the platform mix, the budget split, the offers being pushed, and the account architecture: campaign objectives, ad-set logic, audience segmentation, exclusions and naming. Alongside it I map the measurement plan — which events matter, how they are fired, and how platform numbers will be reconciled with your own records. You approve this before anything is built.
Campaigns are built, tracking is implemented and validated event by event, and the first creative sets go live. Launch is deliberately a starting position rather than a finished machine: enough variation to learn from, not so much that the budget is spread too thin to conclude anything. The first tests are queued before anything is switched on.
Day-to-day management: budget pacing, search term and placement review, creative rotation as fatigue appears, bid target adjustments, and running the test queue. Early-stage changes are deliberately restrained — moving targets every day while campaigns are still learning destroys the data you are trying to collect. You get a monthly written analysis, and I flag anything urgent when it happens rather than saving it for the report.
Once an offer holds at an acceptable acquisition cost with creative depth behind it, budget expands in controlled increments and the account widens: new angles, new formats, new audience pools, additional platforms where it makes sense. Scaling is a decision made from the data, on your timeline and your cash flow — not a milestone I promise in advance.
Indicative ranges rather than a package. Every scope is priced after the discovery conversation.
Paid advertising management typically runs from $3,000 to $15,000+ per month, depending on how much ad spend is being managed and how many platforms are in play. A single-platform account with one core offer sits at the lower end; a multi-platform setup with heavy creative volume, a large feed and complex tracking sits at the upper end. I quote against the actual scope after the discovery conversation rather than pushing every business into the same package.
The management fee is separate from ad spend. Your media budget is paid directly to Google, Meta and TikTok from your own accounts and payment methods — it never passes through me, and there is no percentage taken from it.
I recommend a minimum of three months. That is not a lock-in for its own sake: the first weeks go into tracking, structure and initial creative testing, and it takes a further cycle before optimisation decisions rest on enough data to be worth acting on. Judging the work on thirty days means judging it on the setup phase. If at any point the honest read is that your budget would do more elsewhere, I will say so.
See packages and pricingAds decide who arrives. What happens after the click decides whether the spend was worth it — the full list is on services.
Better landing pages and stronger ad copy. Message match between the ad and the page it leads to is one of the most common reasons paid traffic converts below what the creative deserves.
Explore content marketingFollow-up that costs nothing per send. Most people who click an ad are not ready that day, and email is where those visitors are recovered rather than paid for a second time.
Explore email marketingOrganic search builds a demand source you are not renting. It works on a slower clock than ads, which is exactly why the two belong together rather than as alternatives you choose between.
Explore SEO