Hi Reader -

A while back I told you about the ugliest ad of my career. The pink one. The one that was worth $170 million. ​

Quick recap if you missed it: 2013, I was running user acquisition for Bingo Bash on a $1.75 million monthly budget, and one pink monstrosity starts eating most of it.

I wanted to kill it - and kill myself.

It scaled anyway, crushed everything else, and the company sold for $170 million not long after.

Here's where most marketing teams stop. "Wow. What a freak ad. Sometimes the algorithm just blesses you," they say.

They call it luck. They call it lightning in a bottle. Somebody says "you can't reverse-engineer magic" in the tone of a man who has thought about it deeply, for seconds.

Then everyone high-fives and goes back to brainstorming fresh concepts.

Warren Buffett had this exact argument in 1984. An economist named Michael Jensen stands up and makes the case for luck. He'd already put it in writing: "I believe there is no other proposition in economics which has more solid empirical evidence supporting it than the Efficient Market Hypothesis."

Translation: nobody beats the market, and the ones who do are lucky. Let enough analysts flip coins, he argued, and a few will toss ten heads in a row. It doesn't mean they're any good at flipping coins. Winners are noise. Nothing to see here. Go away.

Buffett takes the coin and flips it right back at him. Run your contest with orangutans engaged in a coin flipping contest, he says, and you'd still end up with "215 egotistical orangutans with 20 straight winning flips."

Quoting him exactly:

Feeding. Exercises. Books. That is not a man admiring a zoo. That's a man planning to build one.

In our world, this has a less charming name: you deconstruct the ad.

And here's the part nobody tells you. Almost no team actually does this. The same team that will debate shades of green for forty-five minutes will not spend ten taking apart a winner.

Why? Because brainstorming fresh ideas feels like art, and studying an old winner feels like homework.

Because the winner usually offended everyone's taste on the way up, and nobody enjoys staring at the thing that beat them. And because "I deconstructed the narrative spine of the janky pink one - and built 6 more based on it" has never once impressed anyone at a dinner party.

The luck story survives because it's flattering. If winners are lightning, then ignoring them isn't negligence. It's just weather.

We take the other road. We deconstruct our winners using our internal software BruteForce AI(and we deconstruct other brands: here is one we did of Wispr Flow today).

Hook. Format. Emotion. The exact person on the other end. The narrative arc that made the winner win. That's not a story about one lucky ad. That's a parts list that you can - and should remix.

Most teams look at their best ad and call it luck. The top 1% ask what the zookeeper's been feeding it - and feed their orangutans the same thing.

Anywayy, if you're spending $50k+/month and want help taking your winners apart and making more of them, reply "ZOO" and we'll talk.

Or don't. I'm sure the next blue sky brainstorming is exactly what will unlock your next winner.

Later,

Shamanth

P.S. I contributed to Singular's new Marketer's Guide to Turning AI into Real Growth, a practical playbook on making AI actually move the numbers. It's free. Grab it here.