Building that Perfect Quantified Self App: Notes to Developers, Part 2
In my previous post, I started describing the quantified self app that I would have built myself, if I had access to necessary skills or resources. Originally, these notes were intended for software developers and enterpreneurs who contacted me after reading my post about my search for a good tracking app. After putting the notes together, I thought that it would be only fair if I shared them publically, so other developers, and Quantified Self practitioners like myself, could have a word in further discussion, should it ever ensue. This last portion of the notes will focus on analytical features of the app, data portability, and potential methods for monetization.
I already described in detail, while the app should be versatile, fully customizable, and enable users to log their everyday life data from different platforms. But our app is not just a tool for logging and accumulating information. At some point, users would want to look at their data and analyze it, and most apps have serious limitations when it comes to this point. You see, most of the Quantified-Self projects fall under one of the following categories:
- SELF-TRACKING PROJECTS
- Progress Measurement (e.g., tracking weight loss, improvement in attention)
- Habits Design/Behavior Modification (e.g., reinforcing habit of brushing teeth by recording it every time)
- Accountability (e.g., tracking expenses, or amount of fat in daily meals
- SELF-KNOWLEDGE PROJECTS: looking at the relationship between various life variables (e.g.,, comparing mood on Mondays and Fridays, or comparing acid reflux symptoms after consuming different types of meals)
- SELF-OPTIMIZATION PROJECTS: introducing new variables or manipulating existing variables in order to improve certain aspects in your life (e.g., self-experimenting with herbal remedies to improve quality of sleep)
Most of the apps that I have seen during my search were made with an intention to assist with Self-Tracking projects. As a result, they often offer some kind of visual feedback in form of trendline, so users could track their progress on the chart. What is missing in many cases, however, is the ability to retrieve data. After entering my data for a month, all I would get is just a couple of trendlines in pretty colors. But data portability is crucial when it comes to Self-Discovery and Self-Optimization projects: I want to be able to download my data in order to manipulate and analyze it myself. Preferably, I want my data back in comma separated values (CSV) format. The most convenient way to export data, at least in my personal opinion, is to simply email it to myself. This is where Track & Share actually beats rTracker. In order to export data from rTracker, I would need to use iTunes. TrackNshare, on the other hand, allows you to email your data right from the app.
Of course, the immediate feedback, in terms of tallies and graphs is still important, especially if I am using this app simply to shape a new habit, or track a simple variable like weight. The more advanced tools for cross-tabulation and more advanced analytics like comparing differences and computing correlations would be a huge plus. They could be made accessible via web interface, as opposed inside the app. I personally would partner with Statwing, or build my own analytic features.
How would I make money from this app?
Well, first, I would offer the app itself for a reasonable price. Considering that my app meets versatility, customizability and cross-platform/mobility criteria, a one-time download fee in the range of 4.99-9.99 seems reasonable to me. This fee would cover also ability to log data from the computer online. Please note that this is my personal opinion, and developers and users may deem a lower or higher fee more appropriate.
Second, I would never charge for data export! The data belongs to users, and they have a right to retrieve it when they need it. However, considering the costs associated with storing data on servers, I would only keep data on my servers for a limited time (let’s say, a month). After the time is up, users have a choice of downloading it and storing somewhere else, or paying a subscription fee for storage. Considering that most of personal data files are relatively small in size, this fee would be relatively low. Of course, storing images, video and audio data on my servers would cost more.
Third, I would offer advanced analytic features that would enable you to slice and dice data in search for patterns (e.g. compare differences across treatments, or compute correlations) . This feature would be offered for additional annual fee.
Fourth, I would offer data visualization and sharing features. Anyone can summarize their data in pretty charts and post them on Facebook, Twitter, etc. I could also offer nice pre-built reporting templates, similar to those produced by Nicholas Felton or Notch, as a PDF, presentation or Animoto-style videos. While some of these charts could be made free, some visualizations could cost .99 cents or could be “sponsored” by brands.
Finally, I would build an online community around the app, on my website. The users will be coming to website not just to log data and run reports, but also consume content from experts, exchange ideas, share their findings, and take advantage of offers and promotions from gadget-makers and other sponsors. This potentially means additional revenue from advertising. Don’t forget also about enabling users to SELL or exchange for products/services their data to outside vendors, like university researchers, pharmaceutical companies, etc.
And this is how I imagine the perfect quantified self app. Again, I am not a developer, and do not have necessary skills to build such an app. To the best of my knowledge, some of these features already exist, although are not yet offered under one service. But perhaps, some of them can not even be implemented in real life? Further, I myself will be perfectly willing to pay some of the fees mentioned above, in order to use this app and its services. But perhaps, my business perspective is naive, and the proposed fee structure won’t be able to cover all the expenses associated with developing and maintaining this app. These fees may also be too high for other users. Finally, I see myself as one of the most loyal users of this tool. But it is completely possible that other folks in Quantified Self community don’t share my enthusiasm, and do not see any need in such advanced services.
So if you mastered strength to finish reading this boring post to the end, let me ask you these two questions:
If you are a developer:
- is it POSSIBLE to build such an app?
- would YOU build it?
If you are a fellow self-quantifier:
- is such an app NECESSARY?
- would YOU personally use it?
And the final question to those who answered yes: can we build this app together? I would love to hear your opinion!