3
3 Comments

What's the % of your time that you spend on testing/reliability of your product?

Most of the projects approach a time when you start shifting from fast feature development and start spending more time on unit test coverage, end-to-end testing, high-availability infrastructure (spawning deployments in new zones, spend some extra $ on those multi-zone database replication settings, etc).

With my projects I personally start with a good unit test coverage as it helps in prototyping (faster than compile, run, try to do something manually) and basic e2e tests. Then, once I get some production users I start adding e2e tests for their use cases so I don't break something too bad.

Would be interesting to hear about other people strategies and how much (if any) of your time do you spend on these activities. Obviously they are not as fun as new features :|

on July 25, 2020
  1. 1

    Historically I spend almost no time testing in the early stages. But I recently built testingbee.io to make testing less painful and it's been amazing. I just use that from here on out.

  2. 1

    In the beginning, I tend to focus more on the product itself, marketing it, and getting data about how people are using it, so I can iterate forward.

    In this stage, I like using Sentry to monitor any errors, if I see a crash that is critical I'll contact the user apologizing and saying that It's fixed now -- it can help to build trust.

    As the product is getting more solid, I'll start to focus more on tests for features that set to stay the same way for the foreseeable future.

  3. 1

    I believe test coverage will save you time in the long run and give you some confidence to release new features. Especially since most indie hackers do not have a dedicated QA. It is hard to remember all testing scenarios if you are not writing them as you go. I’m a big fan of api test that test the application from request to database to response. I have test on all apis for my link shortener service T.LY.