Premise
We know our daily habits and choices are closely-linked to our quality of life and goal achievement, yet struggle to understand the relationships between them; they are largely hidden, and differ by individual. Experiments, self-monitoring, and statistical analysis can help you discover the hidden causes behind your state and performance, giving us the information needed to take actions that help us achieve our peak mental and physical potential. Experiments, combined with observational journaling, yield unexpected insights into simple behavioral changes that can improve your life – causal links between choices and goals.
Approach
Total Optimal is an app that helps you achieve personal goals by discovering the relationship between daily choices and outcomes. It assists with the tracking of choices made and experiments performed, along with performance outcomes (via self-evaluation, journaling, and fitness trackers). When causal links are suspected between choices and goals, experiments help discover possible relationships. Users actively improve their lives through better choices, informed by machine-learning-powered feedback.
Experiment examples:
- does coffee help my productivity / sleep / athletic performance over the short / long term?
- does the duration, timing or intensity of my cardio workout affect my sleep, focus, or motivation?
- is there any noticeable impact of the nutritional supplements I’m taking?
Lessons Learned
Personal experiments and optimizations:
- over short time periods, supplements had negligible effects, except for valerian root, which led to non-trivial improvement in sleep, with an optimal dosage of 2 capsules
Technical:
- The original project was developed for Android; a reboot would use a web-native approach for emphasis on more rapid iteration.
- The original tech stack utilized Firebase, which was easier for our developer but made data analysis cumbersome and more expensive to maintain due to the ETL processes required
Product:
- LLMs now significantly increase the value of text-based journaling and input
Going Forward
If you are interested in a reboot of this project, write to: contact@qpub.ai