...and then pivot
Hi all,
It is time for a pivot from what we were intending to do and what we have ended up doing - this has been a quicker turnaround using the experience we have gained from creating the original version and I will share some of my lessons.
Our original software was an SaaS to allow an end user to create cloud hosted data and suite of API's quickly allowing an end user to spin up data and use in their website, backends etc - think of baserow etc but focusing on the users who wanted a scalable API focused solution rather than the online spreadsheet approach.
Why have we pivoted?
We felt this was too ambiguous and too generic to gain any real traction and struggled to explain the concept. Although it was something we both felt would be useful especially to non developers we were sort of trying to target developers who I think were the wrong target audience.
The pivot....
So we decided to hone in on the area which we felt had a better use case which was concentrating on allowing users of Cosmos DB to manage and secure instant external access to all or some their containers across all subscriptions in one place and protect the integrity of the data with data validation and also allowing custom SQL to be created as an endpoint.
So the users are in full control of their Cosmos estate but we are just adding the top gateway layer so they can be confident in passing the data out to third parties, external developers or external companies etc.
The main lessons learned...
1 ) Don't get too hung up on getting everything perfect, we waited until the original product was bulletproof and stringent testing etc but this turned out to be a waste - so just get something that you are certain works and wont fall over and get it out there - I wish we had done this six months ago but.....
-
Focus on the features that matter and figure out how to get those in your product the simplest way you can - make it more complete with a better UI later when it is being used. We originally created a nice UI to allow users to create validation for the data but in this version we allow them to use a JSON schema, simple and effective, we will build a UI for that later.
-
Don't necessarily build for scale out of the box - we did this and spent a lot of time getting it production ready, yes it is bulletproof and can handle the high production load especially when compared to other solutions on the market - but those users are far off so we could have spent the time on other things.
I would be eternally grateful if you have a look at the website, it is not for everyone as you really need to be a user of Cosmos DB but any feedback welcome.
Thanks for reading
