Ad spend in.
Revenue out.

Managed the full e-commerce operation for a Dutch fashion store, remotely — product research, supplier negotiation, and paid social scaling across Meta and Snapchat. Result: $115K (€107,016) in sales and 3,844 orders in a single 90-day window (Sep 1 – Nov 30), with $101K (€94K) net after discounts and returns — including a single October delivering $22,672 in tracked purchase conversion value from the ad account.

High-volume MENA e-commerce operation: 233 ads running simultaneously, 18.5M impressions and 3.7M people reached on $17.3K spend in the captured window, with purchase CPAs as low as $1.16. In peak campaign windows, purchase ROAS ranged 8.5x–18.6x at CPCs of $0.08–0.15 — snapshot figures, not a sustained blended average. I ran everything from micro-budget ABO testing at $5/day to scaled winners. Part of a Business Manager where I managed $500K+ total spend from 2017–2022 (Meta's data retention has since purged the full history).

App-install campaigns at scale, including post-iOS14 SKAdNetwork attribution. Flagship ad set delivered 24,209 installs at $0.03 per install (1.67M reach, 1.83 frequency); scaled structures held $0.19 CPI at $150/day. Campaign totals: 102K+ installs, 9.5M reach, 14K in-app conversions.

Scaled a modest-fashion brand past $50K in profitable spend at 8.40x blended lifetime ROAS — the sustained, API-verified average, not a spike. Advantage+ carried $21.3K at 9.84x, an ABO structure hit 12.41x, and a UGC creative test peaked at 16.02x in a single campaign window — creative testing, not budget, is the scaling lever.

A Moroccan food brand selling premium products cash-on-delivery. I ran the growth system end to end — campaigns plus the WhatsApp order-confirmation automation that protects delivery rates. Documented: 3,400+ website purchases across the captured campaigns, with the hero ad set delivering 2,467 purchases at $0.67 each. COD economics reward tight CPAs and disciplined kill/scale decisions — exactly how I treat every budget.

I didn't just buy ads — I built the machine that makes them. FrameGen turns a product photo into a ready-to-post TikTok / Instagram / Facebook video ad in minutes: AI-generated scenes and avatars, script approval, captions and delivery — a full pipeline I architected and run on my own infrastructure. Its promise is right in the tagline: AI video ads that actually convert. For clients that means one thing — creative testing velocity: more UGC-style variations, faster, at a fraction of production cost.

Two years of profitable growth for a specialty coffee brand on repeat-purchase economics: 9.8M impressions bought at $0.57 CPM, sustaining 2.51x purchase ROAS (verified live via API). The documented ad-set view shows the granular testing behind it — purchase costs mapped against a $46 (460 MAD) average order value.

Lead gen is a different sport — volume means nothing without confirmation rate. Best documented run: 4,616 website leads at $0.11 each from 923K impressions. For a laser-therapy clinic network: 2,200+ documented leads at $1.41–2.72 CPL, converting into 1,800+ confirmed appointments at a 72% confirmation rate. For a French solar installer: full-stack delivery (landing page, Pixel + CAPI, lead capture) at $44 (€41) per high-ticket lead — API-verified.

A natural health & wellness e-commerce brand running purchase campaigns on Meta. The account's structure follows my standard system: continuous creative testing feeding proven winners into scaling ad sets, with daily kill/scale decisions.

Results documented directly from Ads Manager: top-performing ad set delivered 268 website purchases at $6.62 per purchase, with acquisition costs across scaled ad sets ranging from $5.34 to $14.98 per purchase. Single ad sets reached over 471,000 people while frequency stayed in the healthy 2–3 range — scale without audience burnout.

Funnel efficiency held at the top too: view-content costs as low as $0.06, keeping the pipeline full while purchase campaigns converted the demand.