Discovering what to build
In the last post I talked in general terms about the app I’m going to build in the series. In this post I wanna get a little more specific and also start pointing where to start the development.
There is a whole area of Product Discovery — of which I am just an admirer — so I do not intend to provide a tutorial of what you should do in order to build an app, or create any product for that matter. I just want to outline the steps I’ve made in order to narrow down the possibilities and get a clearer idea of what my app should be like.
The app
So, if you didn’t read the previous post, here’s a summary for you. The idea is to help prevent the development of Alzheimer’s Disease. The app should help in two main fronts: (1) prevent the concentration of amyloid-β, by living a healthy lifestyle, i.e. eating well, sleeping well and exercising; and also (2) create a cognitive reserve, by engaging in cognitive stimulating activities, such as reading a book, learning a new language, playing a musical instrument, etc.
With that in mind, there are two features that are mandatory for the app:
- The user should be able to inform their activities
- The user should be able to see their progress in the two fronts
Discovering
Still there is room for an infinite number of possibilities on how to implement an app that does those things. Also, there is still the question whether those two features are sufficient or am I missing something. Will users engage in the app with those features or should I add something more?
In order to make it a little less blurry, I decided to ask people a bunch of things.
Form #1
Firstly, I created a form to find out about people’s habits. Specially on how they use their phones (what they like, what they don’t, which difficulties they have, etc.) and about their lifestyle (are they already living a lifestyle that prevents Alzheimer’s or is the app gonna help them?).
The most important informations I got from the answers to that form were:
- People spend a lot of time using their cellphones (57% of the respondents spend more than 3 hours in their phone everyday).
- When asked about their difficulties with their phone, the most frequent answer (after "None" with 32%) was the advertising in the apps (8%). Other common complaints were related to the battery, storage and connection.
- 15% would rather record an audio than type the text. However, only 6% prefer listening to an audio rather than reading a message.
- 17.6% consider the internet connection a problem when they use the phone.
- 47% do not exercise, except for their daily activities, such as walking or climbing stairs.
- 34% will only read one book (at most) during the year.
From those results, I got to a few conclusions about the app. Firstly, it shouldn’t have too many ads (if any); it should not depend on internet connection; it shouldn’t spend too much battery or occupy a lot of storage; and also, for some people, it would be nice to be able to input informations as audio instead of typing it. From the answers it seems like the app would be helpful too, since most respondents are not living exactly a brain protecting lifestyle.
Form #2
With the informations I gathered in the form, and also in order to validate the hypothesis that the app would really help people, I decided to create the first prototype. As a developer, I gotta confess that when I hear “prototype” I tend to think of myself opening Xcode and start coding something. It’s almost a reflex. But what I’ve been learning from some very competent Product people is that you shouldn’t spend that much time on a prototype. Thiago Lioy mentioned it in one of his posts. You should try something very simple that already adds value to the users and, very importantly, gives you information on what to build.
So I decided to create a form that would work like the app. The respondent would inform how their day was, what they ate, whether they exercised or not, which cognitive activities they engaged on, and so on. I selected a few people I know to start trying it and I sent them the form.
I would work as a notification. Everyday at 7h45PM I would send them a message reminding them to fill the form.
That experiment gave me some informations I wouldn’t have thought of and it also confirmed some hypotheses I had.
Accountability
The first thing I noticed was that the very fact of answering a form at night is sufficient to make people improve their lifestyle. In the very first days there was a clear difference on the percentage of people that did exercise for at least 30 minutes.
Notifications are important
It was very clear that if for some reason I missed a day and did not send the message, some people wouldn’t remember to answer the form. That one I already had in mind. I know I wouldn’t remember to answer it myself!
Customized notifications are more
After analyzing the data, I decided to test whether sending a message that specifically instructs people on what were the areas they needed to improve would get them a better result. No surprise: people did improve after receiving custom "notifications". So the app should be able to tell people what exactly they need to improve.
Timing
A few days after I started, one of the people that were filling the form told me 7h45PM was too soon for him. It seems obvious now, but I did not consider the fact that people have dinner at different times, some people go to the gym late at night, and so on.
The app should be smart enough to send the notifications in the right time to different people.
Challenges and gamification
One day I decided to send them a notification in the morning reminding them of the activities they should perform during the day and I also sent a link to a public domain book. The book wasn't long so I challenged them to read it in 4 days. The results after the challenge were great. 20% of the people completed the challenge and read the whole book in last than three days.
After one week, I gathered the results and sent them a badge according to how they went in the two areas (preventing amyloid-β and creating a cognitive reserve).
The branches would represent how well they went with the cognitive engaging activities and the soil would represent how well they went with the physical exercises, food and sleep.
People got excited about making their trees look nicer! So I decided to implement that (or some other gamification/reward mechanism) in the app.
Time to build
Well, after gathering those informations, it's time to write some code.
I know I need notifications, I need a way for people to input their information, I need to let them track their progress and it would be nice for them to have some reward if they improve their lifestyle. Also, if some informations could be retrieved without the need for the user to manually input it, that would be great.
In the next post I will talk about the development of the first version of the app. We will talk about HealthKit, CoreData and Push Notifications with CloudKit.
Thank you for reading!
If you have any suggestions or comments, let me know!
