Chances are that you discovered by now that health care apps users won’t pay you a red cent to use the software. This is fine, because I’m sure you know that Larry, Sergey, Zuckerberg and that cute boy from Tumblr made billions of dollars at very short notice without charging anybody for using their software either, so why not you? The modern software paradigm has nothing to do with pricing products. It has to do with getting as many users on your platform, as fast as you can, and the money will follow. Unfortunately, unlike the olden days of social media, your innovation is competing against thousands of others that look just as good on the surface, 99.9% of which will die an uneventful death. When the dust settles, the spoils will go to those thoughtful entrepreneurs who had the wisdom to prepare for success. So make sure you set aside some of the prize money you win at the next health care app fair to get your invention ready for the day when its name becomes a verb, and here are some suggestions:
- The Cloud – Yes, cloud is getting a bad rap right now, but don’t worry. People have short memories and even shorter attention spans, and by the time you are ready to rake in the cash, nobody will remember the NSA storm in a teacup. Make sure yours is a truly multi-tenant application with one database that can scale to hold the millions of profiles you will soon have. Whatever you decide to store on a user device should be replicated to the cloud religiously.
- Open Standards – Even if you rigged a couple of things here and there, now is the time to replace all your data definitions with standards based frameworks. Health care, thank goodness, has an ample supply of those. Use SNOMED, LOINC, RxNorm and keep tabs on national standards development. Even if your app is about fixing lifestyles, you should use standards for your data. No, there are no standards (yet) for eggplant parmesan at lunch, but there are standards for weight, height, BMI, cholesterol levels, smoking status, gender preference, and of course, most importantly, last name, first name, DOB, address, zip, phone numbers, email, marital status, and whatever else you can get without creating suspicion (more below). You probably don’t need all the standard libraries, but make sure your metadata dictionaries (i.e. unique database IDs) are standards based.
- Data Completeness – Whatever your app does, you should always strive to complete your datasets. If your app is providing eating habits modification tools, you should try to obtain all information pertaining to any possible research regarding food intake, and also any possible research where diet may affect results. If you think about it, this is a very tall order, and you may need to do a little research yourself, but a simple diet app could provide you with opportunities to amass data on people’s race, ethnicity, mental status, family situation, occupation, education, income, leisure activities, medications, acute and chronic conditions, etc. To keep your users calm, you should not make any of these items required fields, but instead gently remind them at various points in the lifestyle modification workflow that, for example, evidence shows that optimal caloric intake is affected by say, ethnicity, and give them an opportunity to update their profile to obtain more accurate evidence-based advice from the app.
- Data Validation – Data that is misspelled and inconclusive is useless, so make sure your app knows exactly what the user meant. For example, when you baseline your user’s dietary habits and she says that she is consuming one candy bar at 3 pm every day, ask which candy bar. Snickers? Twix? Kit Kat? Have some pictures display and let her click on her favorite. Is she eating that because that’s what the office machine usually stocks, or because her kids make her buy lots of these products? If your app is a trusted confidant, you can get all sorts of information that will make your profiles well-rounded, and let you provide truly useful advice. Maybe she would be open to switching to PayDay because it has less sugar, or a nice fat protein bar.
- Open Architecture – Good software is built with growth in mind, and nothing grows software faster than a good set of APIs that you can hook to additional functionality developed by others, or by different teams in your own shop. You don’t have to build those extensions now, but you should have the infrastructure to easily add them in the rosy future. So for our Kit Kat lady, you may have APIs available for candy vendors to provide her a healthier alternative right then and there, such as a coupon for Luna bars which are created specifically for her. You can take it one step further and if you have a pill box reminder app, you can suggest that your user may want to talk to her health care professional about certain new meds that evidence shows are helpful to others with similar drug regimens, and if she doesn’t have the money, AstraZeneca can help. So encapsulate your functionality really well, and use generic interfaces for all your class objects.
- Personalization – This is the age of personalized medicine, and the deluge of genomics is almost upon us. It is probably too early for most health care app developers to take advantage of that, but there are other smaller things you can do right away. The most important one is to have your users upload their picture. You can say it’s for safety and security, or you can say it’s for before and after progress tracking, depending on what your app does. Either way, I don’t need to tell you what can be accomplished with cross platform face recognition. Another way to personalize a health app is to surround people with friends and family. Give your users the opportunity to add family members to their profile and to discover close friends who may be using your app as well. It doesn’t have to be a social forum. Sometimes people feel better just knowing that others are in the same boat, and all you really need is the social connectivity map (degrees of separation). Be careful here so you don’t blow the trust relationship with your users, and make this 100% permission dependent.
- Usability – There is nothing more useless than an app that does only one thing, no matter how well. Even if you start with a specific functionality, such as say, tracking calories, you should quickly expand to other services. Sticking with a narrow scope, and refining it ad nauseam, puts your app in danger of becoming a module for someone else’s comprehensive platform. You want your app to eventually be THE platform, so add whatever you can add to your open architecture quickly. Social media links are always a hit. Letting your users tweet about their incredible Crab Shack dinner while also adding the outing to their calorie log, is an example of good usability. Maybe you can enable them to post a picture of the crabs on Facebook to support their active social lives. By the way, if your app has bad news to deliver (e.g. you just ate all the recommended calories for an entire week), don’t be a party pooper; save it for the morning after, when your users are already miserable, thus more receptive to reprimand, and constructive advice on alternative eating establishments.
- Mobility – This is definitely preaching to the choir, since almost all health apps are available for iPhones and most also for Android. If yours isn’t, fix it now. There is absolutely no point in developing solely for the old web any more. And don’t be shy about utilizing all the goodies that come with geo-location (with permission, of course). If you can capture the location and time where Kit Kats are consumed, you have something really unique, like everybody else.
- Gamification – Here too, you are probably ahead of the game (no pun intended), but if you don’t have stars, badges and happy sounds in your app, you’ll have to add them. Even 80 year olds in a nursing home enjoy a gold star here and there, not to mention the youths in their thirties and forties. People need to be reminded often that they are special and provided with multiple instant gratifications on the road to healthy lifestyles. An app that constantly tells you how great you are is more likely to be used than one nagging you to do better. Needless to say that you should skimp on text and lavish your users with pictures, sounds and infographics, because people who can actually read entire paragraphs of plain text are not your target market. If you created an open architecture as mentioned above, this is a good place to throw in an API for some personalized (with user name) electronic coupons and other exciting freebies.
- Taste Your Own Dog Food – The culmination of all your hard work is the creation of a big set of interlinked profiles that can be combined with other available information to provide insights into consumer needs, wants and possibilities thereof. To make sure you are producing marketable goods, you should simulate how others may consume your output. Once you have a little extra cash, go out and purchase readily available data sets, or obtain some that are already in the public domain, and mash them all up to see what happens. If you followed all the steps above, this should be like throwing Mentos into a bottle of Coke. Myriads of luscious patterns will emerge to guide your future development, sales & marketing efforts.