Paid social that
pays for itself.
I plan, launch and scale Meta ads for brands that want more sales, not more reports.
Budgets up to 10x, return held steady.
Every account runs the same way: more of the right people see the ad, more of them click, more of them buy.
- $15Mad spend a quarter
- 10xbudgets scaled, return held steady
- 40+accounts managed
From the first audit to scaling what works.
Pick one, or run them in order. Every engagement starts with a look at your account, so nothing gets built on a guess.
-
01
Account audit
A full read of your ad account, tracking and funnel. You get a ranked list: what is wasting money, and what to fix first.
- Campaign and audience structure
- Tracking and conversion events
- Creative fatigue and overlap
-
02
Full-funnel build
Campaigns for every stage: people who have never heard of you, people who visited, people ready to buy.
- Prospecting and retargeting
- Advantage+ and catalog ads
- Budgets split by stage
-
03
Scale and manage
Ongoing management. Budgets go up in steps, and only while the return holds.
- Weekly optimisation
- Scaling rules agreed up front
- A monthly report in plain words
-
04
Creative testing
A steady test plan for hooks, formats and offers, so you always know which ad deserves the money.
- Test rounds with one change each
- Briefs your designers can use
- Winners moved into scaling
What changed, and the numbers after.
Client names stay private. The numbers are real.
-
Problem
Every time the budget went up, the return fell, so spend stayed stuck.
What changedFewer, broader campaigns, fresh creative every week, and budget raised in small steps only after the numbers held.
ResultBudget up ten times over, return on ad spend steady the whole way.
-
Problem
A hundred accounts, each set up a little differently, and no clear view of which products paid off.
What changedOne catalog setup for all of them, Advantage+ shopping campaigns, and the same weekly check on every account.
Result$6M in revenue a quarter across the group, with return on ad spend up to 3x.
-
Problem
Forty-plus advertisers, different goals, and a lot of money that had to be spent well every single day.
What changedA shared playbook: the same audit, the same test rules and the same scaling steps for every account.
Result$15M in ad spend a quarter, run without surprises.
Policy first, then ads. Learned inside Meta.
I started on Meta's side, reviewing ads against policy, then moved to managing advertiser accounts. So I know why ads get stuck in review, and how the system decides who sees them.
- Week 1
01Audit
I read the account, the tracking and your numbers. You get the list of what to fix.
- Week 2
02Plan
One page: who we reach, what we test, the budget, and what counts as a win.
- Weeks 3 to 6
03Launch and test
Campaigns go live, checked against policy first so nothing sits in review. Creative is tested in rounds.
- Ongoing
04Scale
Money moves to what works, in steps, while the return holds. A monthly report in plain words.
From Meta's policy desk to its ad accounts.
9years since my first Meta project in 2017
2sides of Meta policy first, then the ad accounts
- 2017
Content moderator, Facebook project
Arvato Bertelsmann, BerlinRuled on policy cases every day: hate speech, false claims, violence. This is where I learned why ads get rejected.
- 2019
Technical agent, PlayStation project
Sykes, Berlin50+ customer cases a day by phone and email. Explaining tech in plain words, under time pressure.
- 2023
Meta account manager
Concentrix, Berlin100 ad accounts and $6M in revenue a quarter. Return on ad spend up to 3x. Trained advertisers on Advantage+ and catalog ads.
- 2024
Senior Meta account manager
Concentrix, Berlin40+ e-commerce accounts, $15M ad spend a quarter. Custom funnels, quarterly audits, budgets scaled 10x.
- 2025
Independent
BerlinFreelance paid social, and building my own data and AI tools on the side.
Data and AI tools I build for my own work.
28 tools so far, all on my own GitHub. Building them keeps me close to the data and AI tools that are changing how ads get made.
- 01CORVID Research workbench. Maps people, companies and the links between them from open sources.Research
- 02Prompster A prompt library that lives in the browser, for repeatable AI work.AI tooling
- 03Talk Aid Live German-to-English subtitles with suggested replies.AI and language
- 04Cutroom Turns a folder of clips into one edited video.Creative
- 05Model Library Sorts a folder of AI-made images and video by the face in them.AI and media
- 06AllesRoger A free, open-source DJ prep app: key, tempo and energy for a whole music library.Music data
- 07Waveroom Audio workbench: trim, split voice from music, and line tracks up for a set.Audio
Arabic native, English, German. MENA and EU campaigns.
I read the ad copy, the comments and the landing page in the language your customers use. I trained as an English-Arabic translator at Damascus University.
- العربيةArabic, nativeMENA and the Gulf
- EnglishEnglish, C2UK, US and global
- DeutschGerman, based in BerlinGermany, Austria, Switzerland
After hours I DJ and make techno.
Berlin techno, made in Ableton. I even built my own app to prep my tracks. Same habit as the ad work: try it, listen, cut what doesn't land. No audio on this site, promise.
Good ads don't
look like ads.
Let's make your ads pay for themselves.
- 01Book a short call, or send an email.
- 02I look at your account before we talk.
- 03We agree on a plan and the first step.