05 Mar 2020

Exist annual survey 2019

As part of our values of working transparently, we send an annual survey to Exist users to get their feedback on how we're doing and what we should work on next. We've been doing this since 2015, the year we launched.

This year we broke from tradition and didn't include questions on what we should do next. We've already decided our plan for 2020 is to focus on adding manual tracking features, expanding the usefulness of Exist so users can track whatever kinds of data they want.

So with that out of the way, let's go through the survey questions and your responses. This year we received 250 responses.

Survey questions

This year we started with a new question, "How much does Exist matter to you". We wanted to set a baseline of interest so we could break down later answers based on whether the respondents felt invested or not. 15% said they were "our biggest fans", while 60% said they were "fairly invested". 24% said they like Exist but aren't invested, and 2 users "don't really care". Thanks for filling out the survey despite that level of interest! This is about what we expected, as those who can be bothered to fill in the survey are generally those who are more invested.

The most popular way respondents first heard of Exist was "article or blog post" with 37%, then social media with 10%, and then podcasts and related services (like RescueTime) tied on 8%. 24% of respondents picked "I don't know", however, which is fair enough if you've been using Exist for a long time. We expected podcasts to be higher here, given our previous shout-outs from people like Merlin Mann, so this was surprising. If you'd like to help us grow, it seems that writing a blog post about Exist would be the biggest help :)

This year 44% of respondents use the Android app, and 51% use the iOS app. Last year was 41% and 53%, respectively, so it's nice to see a further evening out. Years ago iOS users were much more in the majority.

Mood tracking is used by 94% of respondents, down from 96% last year. 46% rate their mood every day, and 40% catch up on multiple days at once at least half the time.

91% of respondents have custom tracking turned on, up from 88% the previous year. It's not surprising to see this climb as custom data has become more integral to the value people see in Exist.

42% of respondents don't use the web app at all for editing mood or tags. The web app declines in use every year, which is the result of a positive feedback cycle where people use the mobile apps more, so we add more features to the apps, and so on.

The main reason respondents use Exist is "seeking interesting or useful insights into their behaviour", with 57% picking this option. This is up slightly from last year's 54%, and remains the clear winner here. "Keeping data in one place" was next at 19%, down from 23% the prior year.

70% of respondents picked at least 4/5 when asked how well Exist was providing what they were after. 26% felt ambivalent, on 3/5, and only 4% picked 2/5. There's always room to grow, of course, but this is an encouraging result.

The biggest problem respondents have with Exist is "too much effort to track mood and tags" at 21%. This was the biggest problem last year too, on 25%, but it's declined this year. We've made some changes to make the process quicker and easier since the last survey, such as the quick tags interface and background saving, so I'm glad to see that reflected in the outcome. Next was "not enough integrations I use" on 20%, which seems high to me. My guess would be that now we're working down "the long tail" of data types, rather than mostly focusing on activity data like we did some years ago, we're giving the appearance of being open to track any sort of data you like, when of course we only have a finite number of integrations. (This will remain a problem for the lifetime of Exist as developers continue to make apps for the purposes of tracking data, and then don't give their users any ownership over that data or way to export it. Sigh.) Finally, the free-form answers mostly fell into "too much effort", "not enough integrations I use", and "I want manual tracking", which, again, we are working on this year.

To ruthlessly force people into showing their priorities, we asked what four types of data people would keep in Exist if that was the maximum allowed. The most popular data types were health (a nebulous group of attributes like weight and heart rate) on 86%, sleep on 85%, activity on 77%, and productivity on 66%. Social media was the lowest on 7%. I'm surprised to see health at the top but otherwise I think this is a sensible group that reflects what we consider the most useful data as well.

We then followed up with a question about which four features respondents would keep as well. There was much less consensus here, which I think reflects multiple groups using Exist in different ways. Long-term trends graphs were the most popular feature, on 66%. Next were correlations on 58%, custom tags on 56%, and mood review on 52%. I find the popularity of the trend graphs baffling, and wonder if people were confused about this answer. Nonetheless, as a group and when combined with the previous answer, I can see that you'd have the core of a useful app, so I think broadly this result aligns with what we expected. Phew. Interestingly, today's progress (the web dashboard, and the mobile Progress tab) was second last on 22%, with the weekly report emails last on 18%. The dashboard was historically the core of Exist, and the weekly summary email has been a favourite based on other feedback, but neither counts as a core feature here.

The most popular speculative addition to Exist was habit tracking, at 71% (multiple answers allowed). This is almost exactly the same as last year's result, and is the reason why Belle has been working on an iOS app called "Exist Habits" to explore what that might look like. (Given the enduring popularity of this answer in the survey, I'm surprised that beta tester feedback has been fairly tepid. Either we have the wrong approach there or it's something people think they want but don't actually want to put the effort into.) The next popular answers were "time of day tracking" on 49%, automatic location/venue tracking on 46%, and "more proactive feedback to change your behaviour" on 41%. Location tracking has fallen a little from 50% since last year, and I hope it will continue to decline in popularity as people realise how hard it is to get right for what they get out of it.

Finally, respondents wanted to read more on "Quantified Self and apps for tracking your life" at 72%, and "general research on productivity and building habits" at 56%. Last year "how to use Exist and interviews with other users" was equal first, but this year it's last. Users are fickle! It's a good thing we haven't acted on this question at all because we're terrible at writing things.

Thank you to everyone who wrote something positive in the final free-form feedback question, we get a real kick out of reading these :)

What's next

I'm getting fewer questions about this these days, so I guess I've finally been able to put the right info into all the right places. But one more time, just to be clear, we are spending this year working on manual tracking. It's a big change, we don't have an ETA, and we won't be quickly adding your favourite feature first. If you want to hear more about what this means, please refer to the announcement blog post which goes into more detail.

Thanks to everyone who answered this year! We hope you've been heartened by the chance to participate, because we've been heartened to hear from you. Although we didn't ask you directly what we should work on this year, your feedback is forever helpful to understand which parts of Exist are most useful to you and what we can improve on.