flowchart LR
acquisition --> taskratchet --> churn
taskratchet --> new_task --> taskratchet
new_task --> email --> taskratchet
beeminder --> acquisition
beeminder --> taskratchet
new_task --> beeminder
- The addressable market is too small to fully support myself financially.
- I develop features that fail to provide value to users.
- I can build a product that provides value to users.
- The addressable market is enough to support myself financially. To make $1,500 / month, I'd need 3000 new tasks to be created per month on average. This is 100 tasks per day.
- Cohort analysis
- Split testing
- Tracking prodcut-market fit
- How many users never create a task?
- What's my churn rate?
- What are my most effective acquisition channels?
- Why do users stop using TaskRatchet?
- What use cases do users use TaskRatchet to solve?
- What changes to TaskRatchet would create the most value for users?