Case studies — real numbers, gross and delivered.
Every result here is reported the way I report all of them: the flattering platform number and the real collected number, with the gap named. Five engagements across education, e-commerce, a full store build, and my own SaaS.
The five case studies
121,330 AED in ad spend turned into ~912,550 AED collected. Then the site started getting cited inside Google's AI Overview across 8 course lines.
Read the case →Four ad platforms claimed SAR 14.2M; the store banked SAR 11.5M across ~27K orders. I reconcile dashboard credit to collected cash — and report the second number.
Read the case →A conversation engine that drove 10k+ chats. The dashboard said 4.1×; the money that actually cleared returns and refused deliveries said 1.9×.
Read the case →A complete store — site, UI and UX — designed in Claude Design, implemented with Claude Code, integrated with WordPress + WooCommerce, and shipped on Hostinger.
Read the case →I built the AI, then used it to acquire 1,230 leads at roughly $6.50 cost-per-lead — the system selling the system.
Read the case →Why every result here shows two numbers
Ad platforms are enthusiastic witnesses to their own performance. Each one counts the sale it thinks it caused, so four dashboards can claim more revenue than the business ever banked. The gross number is real — it just isn't the number that pays salaries.
So every engagement here is reported twice: the platform-reported figure and the money that actually cleared returns, refused deliveries and refunds — with the gap between them named, not buried. That gap is usually the most useful number in the account, because it's where the wasted spend and the fixable leaks live.
Request a systems diagnostic →