29 Jun 2016

The state of Exist, 2016

We recently sent out our annual Exist user survey to all our users. Thanks to everyone who took the time to complete it! I wanted to recap some of the interesting results we found, and then talk about what we're going to do in response.

Let's take a look at some numbers from various areas in the survey. We'll assume our 110 survey results are representative of the larger user base.

Survey results

Goals and insights

  • 38% of users pay some attention to Exist's smart goal tracking
  • 55% of users look at their insights at least weekly
  • Only 10% pay attention to their real-time step goal insight (so that wasn't a very useful idea), but 84% would like to try getting insights and suggestions as push notifications

Mood tracking

  • 70% of users find mood tracking useful
  • 56% of users rate their mood most days or daily
  • 40% of users add a note most days or daily
  • 84% of users find the Trends pages at least "moderately useful" as a summary
  • 25% of users find correlations useful as is, 41% want them to be more useful or actionable

Everything else

  • 55% of users have our iOS app, 34% use our Android app, 11% only use the web app
  • 75% of users think the weekly email is at least "pretty useful"
  • 65% of users also use a habit tracker app, 51% use a personal data visualisation app, 45% manually track things, 43% use a journalling app
  • 85% of users think Exist works better for longer-term trends than as a day-to-day tracker
  • When we asked users what one thing we should work on this year, 30% wanted smarter, more informative correlations, 26% wanted more subjective (manual) data tracking, 19% want more proactive feedback and real-time suggestions
  • When we asked users what one area of data they wanted to add, 39% wanted custom tracking, 19% wanted more productivity data, 18% wanted media, 17% food
  • When we asked about alternative Exist niches, 61% said they'd use Exist for Mental Health, 60% Exist "Smart Assistant" for managing their devices and Internet of Things, 30% wanted Exist for Chronic Illness (this one was multiple choice)
  • 54% would pay a little more if it helped us move faster, 16% would pay double

What it means

It was quite interesting to see the results roll in and find that we'd misjudged many users' priorities, having guessed at trends for some things that turned out to be quite off the mark. We knew that mood tracking was a popular feature, but we underestimated how much users want custom tracking, given that Exist is all about importing your data automatically! Users also overwhelmingly prefer using Exist to check on how they're doing long-term, rather than on a day-to-day basis. Our "smart goals" based on users' average activity aren't used much, so we assume most users prefer to track their own static goals in their activity tracker. (That's a shame, because I love those and use them every day.) The overwhelming trend was towards making Exist more useful for finding long-term trends and understanding correlations — eg. showing how whole seasons affect your activity, removing obvious correlations, explaining what correlations might mean, and maybe even using them to give suggestions on behavioural changes. If most people don't use Exist to track how they're going daily, then our job is to make it more useful for long-term reporting.

So based on those results and some specific written feedback, we've already implemented some changes you might've seen:

  • We've removed public profiles, so there's no more worrying about public and private attribute management — all Exist data is private only
  • We've made a blacklist of common "obvious correlations" so these are never displayed
  • We've analysed how often correlation combinations occur, and added explanations and related links for many that are common. We'll keep working on this — it's quite time consuming and there are over a thousand combinations!
  • We've implemented a notifications tab in the web app, which we'll use to give users updates on our progress between big, email-worthy feature releases
  • We've made a bigger fuss of the roadmap and the API at the bottom of the services page, so new users are aware of their options
  • We've made a couple of other small tweaks to the web app and the weekly email

Belle and I also sat down and had several big discussions on what we should be working on, long-term. This stuff really matters, it's our baby, so we want to get it right. We've been pretty public about the fact that Exist only pays for me to work on it full-time, and Belle spends most of her time on her freelance writing work out of necessity. Because Belle is in charge of our iOS development, but we're constrained by her lack of time, we have to limit the amount of work we plan that needs her involvement. Unfortunately, manual tracking will involve huge additions to the iOS app, so we had to take that into account.

Belle also brought up the fact that, aside from our survey feedback from existing users, we receive a steady trickle of data from users who delete or suspend their accounts and leave us feedback on what was missing. A huge number of these users are still telling us that they don't use enough of our connected services to find Exist useful, and wish we had more integrations. This is something we have to take into account too.

We were all set to go all-in on manual tracking, but we've decided to change course. To get more of Belle's time so we can make big changes to our iOS app, we need to make more money. To make more money, we need to keep growing! And to keep growing, we need to get, and keep, more users. So we're putting manual tracking on hold for a little while, and I'm going to knuckle down and try to get some more integrations completed in the next few months. We've already planned Google Fit and Todoist, so these will be up first, and after that we'll re-evaluate. After a certain number of integrations, I'll move back to working on manual tracking, and hope that Exist can pay for more of Belle's time by that point.

So what do we do about the fact that many users were willing to pay more for Exist? We went in circles about this one too. On one hand, it seems crazy not to accept users' help, because both parties get what we want — users get more progress, and Belle gets to dump her freelance clients (I mean, uh, let them down gently). But on the other hand, the amount of extra money we could make from charging a small portion of users a bit extra would probably still not be enough to get Belle on board full-time. We're not comfortable taking more of your money without being able to provide tangible benefits in return. Ultimately, our best course of action is to build a better product that attracts and retains more users. Everything else is what my dad would call "fiddle-farting around". It's just wasting time. (No, I don't know why he calls it that.)

Oh, and speaking of money, we've added a stats page so you can track how we're doing in between monthly reports. We're happy to keep being totally open about the business situation, and this page will help keep you up to date on our progress towards being totally legit and not poor.

So, to recap:

  • We've made some changes to correlations already to make them more useful
  • We're going to be adding more integrations in the short-term
  • Manual tracking will still happen, but it's been pushed back by a few months
  • We have some other plans for new ways to present trends and correlations, focusing Exist more on the value of long-term reporting
  • We'll be adding more of the trends and correlations bits into the mobile app so you need to visit the web app less often
  • We're not taking more of your money, but thanks!

I hope you can see the reasoning in our decisions, and Belle and I are both so grateful for the support of all our users. Even though our progress has been slow, we're not giving up just yet, and we hope you won't give up on us either.

Frequently Mentioned Issues

We saw some people mention things in the written feedback parts of the survey that are worth addressing here to correct misconceptions.

Why doesn't Exist for iOS work on iOS 8?

Good news, it already does! I just tested it again today to make sure. If you're having trouble running the app, that's probably a bug specific to your Exist account that we can fix. Send us an email and we'll look into it for you.

Why doesn't Exist work with my Apple Watch?

Well, it actually does, to some degree. You can send your steps, steps distance, and floors data from your Apple Watch to Exist (via Apple's Health app on your iPhone). For more info, check out this blog post on how to connect Apple Health to your Exist account.

Why don't recurring calendar events show up in Exist?

There are two main reasons for this: one being that anecdotally we noticed people used recurring events more like reminders (eg. Sunday at 9pm is the 'put out the rubbish bins' event), and they mostly didn't seem to count as something you're "really doing".

The other is that it's actually really hard. Behind the scenes, there's only one event created for the first instance, with rules on how it's meant to recur, and that means whenever we read your calendar (even if those events were created years ago) we have to keep applying that recurrence rule to the event until it gets up to the current day, then see if it matches.

I like how the word cloud of my mood notes is coloured to match the mood scores I chose.

Actually these colours are random. We picked them just to make it look pretty. Sorry if that was confusing! We changed them to be blues and pinks to avoid confusion.

Why is the mood reminder always set to 9pm?

Actually, we changed this quite a while ago. If you use Mood by Email, you can set your own mood reminder time for any hour from 5pm - 11pm from your mood service settings page. You can also find this setting in both our iOS and Android apps.

Why can't I use tags in my mood notes?

Although we don't have tags, we do correlate your word usage with your scores. On your mood trends page you can see which words you mention most for each score out of 5.

We're not going to include tags to use in correlations with other data, because eventually we'll have manual tracking, so instead of adding the tag "coffee" to your mood note, you can just track that you had one cup of coffee.

Why can't I rate my mood for days in the past?

You can! You can do this in both our mobile apps on the Timeline tab, and on the web in the mood tab.

I'd like to use custom tracking to track my workouts.

We already have workout tracking! If you use Fitbit, Jawbone, Misfit, or Runkeeper, you can track workouts in Exist. Most of these services let you manually track your workouts, too, so you don't need to have a physical Fitbit, for instance, to use it for tracking your workouts.

Why don't you have an API I can use?

You're in luck! We've had an API for quite a while. Check out the docs, and let us know if you have any questions.

You can also find the link to the API docs anytime from the footer of the Exist website.

I'd like more info about what you're working on.

Fair enough! We try to be transparent about what we're working on, and how the company is going overall. You can find out more by checking out our roadmap of upcoming integrations, our company blog, our company podcast, and our company stats page.

You can also follow Josh, Belle, or Exist on Littlelogs (either with a Littlelogs account, or using RSS) to keep up with day-to-day progress.