What the work actually taught me
Each of these is written method-first: the decision, why it worked, and where it stopped working. Numbers are included as context rather than as the headline, and where I don't have a number, I say so.
The Pixel Was Firing. It Was Just Firing Wrong.
Broken conversion tracking isn't a reporting problem, it's a media-buying problem. The algorithm can't optimise toward a signal it never receives.
Relaunching the Same Ad Didn't Reset Anything.
Copying a winning ad into a fresh campaign doesn't give Meta anything new to learn from. It restarts learning on the same ad, for the same audience, and you pay for it again.
The Course Was Already Unlocked. Nobody Had Told the Students.
A post-purchase message failure doesn't stay a support problem. On paid social it becomes an acquisition problem, your angriest customers are commenting on the ads your coldest prospects are reading.
The Rules Changed Before the Numbers Did.
In a regulated market, a falling cost per result can be the warning, not the win. The rules change before the numbers do.
The Followers Were Bought. So Was the Audience Meta Kept Finding.
Buying engagement doesn't just fail to help, it teaches the ad platform to go and find you more people who behave like the ones you bought.
We Renamed One Bot. Engagement Doubled. The Rest Never Shipped.
When adoption is mandated, engagement metrics measure compliance, not interest, and you cannot gamify your way out of a motivation that was never there.
Why We Sold an English-Speaking Course in Hinglish
Sell in the language of your buyer's current competence, not the language of their aspiration.
Project portfolios
Shorter build-and-ship work, grouped by service rather than written up individually.