
Data is the new oil shit 💩
I stumbled upon an amazing thread on Reddit. A developer released an app in the body-tracking category (similar to Tuute service, which seriously tracks farting — yes, that exists too). Users had to photograph their, um, feces. The app gave them some basic analytics, while under the hood it stored everything in a database.
Result: the guy has a dataset of 150,000 labeled and classified (!) photos of shit from 25,000 unique users.
And now he's sitting on this pile (in every sense) and complaining that he can't find buyers for his dataset.
You need to understand that the entire Reddit, Github, and Wikipedia have already been vacuumed up by giants. You won't win the competition on texts, cats, or open source. Real added value lies in hard-to-reach, dirty, and specific niches.
A labeled medical dataset of 150k images in such a field is a goldmine. Manufacturers of smart toilets (already making cameras that analyze your stool on the fly) or pharmaceutical companies researching intestinal diseases spend years and millions of dollars on clinical collections of this volume.
But the guy collected gold but didn't think about a monetization method.
Big business will never buy a dataset from a nobody; they need legal clarity. And selling biometric/medical data (even supposedly anonymized) collected through a mobile app's user agreement is a direct ticket to court.
The only viable way out for him now is to sell not the dataset itself, but the entire company. Or train his own CV model and sell B2B clients access to the API.
Moral: if you're making some shit, first think about how you're going to sell that shit. ☝️
Comments
0No comments yet.
Sign in to join the discussion.