3 ms·
Basically every application is "purely a UX + data" issue. ML is only sometimes a good solution. In this context, the "UX + data" issue is terribly hard. The r
by sidlls 8y ago
Basically every application is "purely a UX + data" issue. ML is only sometimes a good solution. In this context, the "UX + data" issue is terribly hard.
The reason why you don't have an all-in-one solution is because it's incredibly challenging and expensive. Where do you start? If you try to aggregate by provider you're looking at everything from modern applications for EHR to paper record keeping. If you try at the insurer level you're looking at entrenched players who have a profit motivated interest to explicitly not have such a product, long lags between service date and the data being available, and terrible data quality. Self reporting? Good luck with quality or consistency. This doesn't even begin to cover the regulatory issues and monetization.
- hsikka 8y agoWell said. What if you go to the consumer first, and build a beautiful, gamified experience that deals with non regulated data: i.e. fitness, sleep, and genetics. I imagine if you get to a critical enough mass of consumers, you could gain leverage over the incumbents and incentivize providers to help you integrate?
- hsikka 8y agoObviously these aren't easy things to do, but I'm trying to scope the broader problem down in order to make it actionable
- linesweep 8y agoThe experience you propose sounds like Fitbit + 23andMe, which sounds great, personally. Unfortunately, I don't think providers would want to dig their own grave. sidlls is also right - regulatory issues would be a huge headache.