1) It started out as just an image optimization API (so it was a little bit of a niche product back in 2013).
2) Our API covered all major image formats, and supported both lossy and lossless optimization modes, which again, was pretty rare at the time.
3) Listen to customers, fix things, and add sensible features such as image resizing. Always listen to customers and try to understand what they want even if they don't know how to explain it themselves.
4) With enough people using our platform to essentially replace the development work, R&D and infrastructural requirements needed for a decent imaging workflow, the app will essentially market itself.
5) Develop a stack which can be rapidly scaled up and simultaneously allows for costs to be kept as low as possible. Pass on the value to customers at every available opportunity.
6) I'll edit this post and add more detail once I get to a real computer.