During my two years as a consultant, I kept a spreadsheet tracker of the behaviors I was focused on improving. The behaviors ranged from exercising and mental wellness state (1–10) to snooze counts and ensuring I talked to a person other than my coworkers.
The reason the latter was important to me — I was a traveling consultant in the middle of Texas and the only people I knew within several miles were my coworkers and my clients. And as much as I loved my coworkers, I realized I hadn’t spoken to anyone else in weeks.
You must be wondering why the column headers are so cryptic. There really is no reason to shorten “Hours” to “Hrs” and “Snooze” to “SNZ”—jeez, what an idiot, but I can explain!
Consultants spend 40% of their day in Excel and the other 60% of the day in PowerPoint. Consultants also use an obnoxious amount of acronyms. I kept this spreadsheet up at all times during the work day hoping that anyone who looked over my shoulder believed I was working. This way I never missed a beat when tracking — it was always present.
Identify, Hypothesize, Test
Besides allowing me to procrastinate from work, the tracker allowed me to reflect on my actions. After seeing weeks of sub-6 hours of sleep a night and my mental wellness state averaging at 5.5, I realized something had to give.
It’s easy to fall into a pattern and the tracker allowed me to see a red flag for what it was. The tracker also allowed me to see obvious correlations. I usually slept less than 5 hours on travel days and had tons of coffee. On the surrounding days, all my target behaviors suffered — I didn’t work out, I didn’t complete problem sets on online courses, and my snooze counts would be off the charts.
Once I became aware of these patterns, I put processes in place to correct them. Instead of traveling Monday morning, I traveled Sunday night to test my hypothesis. I ensured I worked out on Sunday, so I could have a rest day on Monday. By using a tracker and identifying issues, I was able to set commitment devices and corrective actions.
Not even me! Even though I was making great strides, there were pain points with the tracker. There were days I would forget to log data, days I would have log fatigue, and times in the day were I would forget earlier behavior (I logged behaviors while at work and at which point, promptly forget how much I slept). I decided to tackle these pain points and automate the process.
I set out to build a layer of automation to tackle my three pain points:
- Data lost due to forgetting to log at all
- Not logging due to log fatigue
- Even if I wanted to log, logging time was too far from actual activity
I spent a few hours articulating requirements of version 1 of this tool. It would allow users to track a set of questions (the equivalent of column headers in the tracker). Each question had a frequency attached to it, and at the frequency, an email would be sent to the user to prompt them for a value. So, at 6pm an email would be sent to me every day with “How many coffees did you drink at work today?” Values would be continuously tracked and the user could retrieve the values as they pleased for reflection and analysis.
I spent two and a half weekends figuring out what I needed to build the tool and executing on the smallest version of it. Luckily I was already in the middle of “Intro to Backend” from Udacity, so I decided to use Google App Engine (and by god, they have an amazing free quota). In another piece, I’ll go into detail of the reqs building and the build process with GAE.
What Does It Look Like?
Well, it’s not pretty, but it does all the things I want it to! It sends out emails at the appropriate times and allows the user to log and download their information when they need it. Sweet!
I’ve been using it for 2 weeks now and it’s worked well so far — no missed days!
Want to Try It?
If you’d like to try Better Me yourself, go for it! Let me know what you think and what can be better (heh heh).
Where to go From Here
There are numerous enhancements I want to make to Better Me. For starters, usability is nowhere to be found — the following will be updated this week:
- Deleting questions (what!)
- Timezone support (sorry Team UTC-is-life)
- Advanced frequencies — hourly, daily, weekly, some days, weekdays, etc
In addition, some longer term updates are in the works:
- Implementing a charting framework
- Simple correlation/regression functionality
- More entry methods (SMS/app)
- Mass export of question data
- Advanced analysis (buzzword alert: Machine-learning)
- Import data (like Fitbit and other IoT data)
If you end up using Better Me, please let me know what you think! Also, if you thought this was valuable to read (or straight up sucked), I’d love to hear that too. You can learn more about me and the project at my page: galeeb.com