How I Built this Vol. 2: The Pretty Pivot

If you are building a company from scratch, there will be things that stick, break, take off, and fall down. I am no exception. Here’s what I built, what broke, and what I finally understood.

When I started hi-BB in 2023, AI was still a sparkly, lean-in-and-listen kind of topic. There was so much curiosity, a little bit of awe, and this shared sense that something big was about to happen. So I built it into that. An intelligent, thoughtful, rational system that could cut through thousands of products and tell you, clearly, what was right for you.

From a technical point of view, it worked really well. But it wasn’t as sticky as I thought, and as a determined entrepreneur, I was going to get to the bottom of it. We tested everything. Different landing pages. Different entry points. Quiz, chat, search. Our Meta ads got clicks. Curiosity brought people in. And then, quietly but consistently, people dropped off. Follow-ups told us it wasn’t because we got the answers wrong. In fact, we got a lot of oohs and aahs. But the problem was deeper, and for a while, I was dumbfounded.

Then, after one of our investors asked me the hard question, right after I rattled off all the data and insights, “What did you learn, Bina?” everything clicked. In a moment of real clarity, I said, “I think I built the perfect rational solution for a problem that is deeply emotional and often irrational.” Saying it out loud lifted a weight off my shoulders. I actually felt lighter.

The AI hook wore off as bad PR was only growing. So what I learned, and probably always knew, is that beauty decisions are made on impulse, self-identity, expression, mood, boredom, hope, insecurity, and self-reward. It is why influencers work. Why do you buy something in Sephora that you did not plan to buy? Why do you trust a friend’s recommendation even when you cannot fully explain why? Even the most “rational” places, like Reddit threads or ingredient deep dives, are still driven by emotion underneath. You are not just looking for the best product. You are looking for the feeling of finding the needle in the haystack, something you discovered yourself. Bottom line, the first version of hi-BB made you work too hard to hit joy.

While we were experimenting, AI recommendation infrastructure caught up, and maybe even sped past us. ChatGPT announced commerce. Sephora aligned with ChatGPT. And all of a sudden, my value proposition felt like table stakes.

Also, from a business model perspective, I was too far away from the money.

Then Claude Code has transformed everything. What used to take months to build can now be built in minutes with vibes. The moat evaporated, and the original code’s value plummeted. The real work ahead of me is making people feel something again. What we need right now is not a better recommendation engine, something that meets you where you are, pulls you in, makes you feel something first, and only then, quietly, does the rational work of closing the loop.

That is what the next version of hi-BB is about. Emotion first. Rational second.

Stay tuned for the unveiling of hi-BB 2.0. Grab a tissue, because I hope you weep with happiness. We are turning the volume down on AI, and turning the volume up on people, play, and emotion.

Bina