With every experiment, we learn "what not to do".

For eg: I built a meme editor on the Facebook platform (canvas apps, remember Farmville?) about 10 years ago.

My first learning was: Users are lazy!

We changed the app into "Lucky number" generator, it went to ~250,000 users.

Then we built a Photo collage app that picked up pictures from Facebook albums and changed the cover of the profile with the collage.

It went viral, reached ~1 million users within a week, to only be banned by Facebook later for using their existing content (pictures).

Another learning: Prefer open platforms over closed ones.

E-commerce was booming back then, started a T-shirt company only to realize that the margin was small, the courier overhead was a big deal (costly) and selling to consumers is hard.

Another learning: Learnt to accept online payments, VAT, local laws and so many things.

Added a lot of skills on the way only to start a services company because it was comparatively the safer and easier option.

Learning: Skills are easier to sell for services compared to products.

Kept experimenting on the side and then entered React Native (out of curiosity) which led to NativeBase (14K Github stars).

Learning: Having an active R&D improves not only improves the skills but also works as a good marketing channel.

Kept working on the 70-30 ratio for Services & Experiments, scaled the company to 270 folks!

Learning: Bootstrapping takes time but it takes the pressure off and gives you the bandwidth to work on your favourite projects without having an external investor.