Like custom tracking, this one requires architectural rejiggering and so while it will happen because we want to do it, it’s still a while off.

This would be very, very useful.

very important, key moments are:

after wake-up before lunch mid afternoon after work before sleep

would not track more than 2 at a time, depending on the intended outcome I’d choose the timeslot

I agree that this is useful, and I would add that it would be useful to keep a separate “overall” category to rate a day, and then have separate ratings by moment. For example, I noticed that the things that happen towards the end of the day tend to have a large influence on the overall day rating that I do at night. Further, research shows that the morning mood is an important factor in productivity: it would be nice to see if this is true for individual users.

i like the idea of some “default” time of days but with customization options

This idea is actually how I found Exist… because I wanted a way to track my mood over time, with tags. Particularly, I would like to record how I’m feeling at the beginning of each day, because I think there will be some important trends in that data, related to motivation and mental health — eg. “what gets you out of bed in the morning?”. Also, I have a hypothesis about “Monday Morning” syndrome that I’d like to test.

I’ll vote for this as well. I’m using a standalone app called iMoodJournal to bug me thrice a day for my mood. Being able to have one app for everything is better, in my book.

I manually track mental health and physical health - I know this is beyond the scope of this consideration, but I’m just throwing it out there as an idea.

I am a nutritionist and it is important for me to track what my patients are experiencing multiple times during the day to understand what triggers their food choices

I would also like a larger scale (1-10 if possible) as many of the other mood apps use

Changing the mood scale would be a big hassle, probably disproportionate to the benefit received. There are pros and cons to both approaches but we value the simplicity of 1-5.

I really need a lifestyle/health tracking app that measures my moods at various times of the day. My mood affects and is affected by sleep, work, exercise, food, social activity, etc. And I’d like to suggest a compromise on the mood scale. I agree with Joan’s comment on June 29 that a larger scale would be very useful. Those 1-5 things are almost useless for someone seriously trying to track their moods. But if 1-10 is a big hassle, could you make it 1-7? From my experience, that would help quite a bit.

Any change to the scale would be the same amount of hassle 😊

Makes sense that changes here are difficult to make, especially when you start thinking about how historical data would be affected… But I really think that having this functionality is a game-changer, and makes Exist more appealing for people that are using other apps to handle it right now

Would be great to enter mood multiple times per day and on demand since mood can change a lot during the day, also would be wonderful if you could make the range of evaluation 1 to 7 (or user defined) for better evaluation.

multiple moods per day more accurately reflects the day and the insights we read. e.g. in one day I may be angry about work and happy about family - and overall rate the day a 2 b/c of work being so bad but in the trend data it shows that family comes up associated with the 2 when actually family should be the 4 work should be the 2

could just use a simple average for each day to get the overall mood for the day

but at least for the insights of what drives each mood rating from the notes, the notes will be more accurately associated with the ratings

Have you looked into integrating with MoodTrek (which is owned by Cerner, so who knows what kind of infrastructure it might use)? MoodTrek also uses a 1-5 scale, and it would be AWESOME to import the day’s average of all moods recorded. I’m basically doing this when I record my mood in Exist. Even better would be to be able to import or track mood at the time recorded.

Unfortunately MoodTrek doesn’t offer an API, so there’s no way for us to get to that data. Feel free to suggest to the MoodTrek developers that they offer a public API or integrate with ours (http://developer.exist.io).

This really is an issue for me too. I never know when to rate the day and then in the evening I often forget doing it. Also in the evening I’m often unsure how my mood was “in total” over the day. It would be really great to just be able to create multiple entries per day and just get an average of those entries for the day.

I also do it right before bed. I don’t think there’s one “right way” to do it, but I think if you struggle to categorise your day you should try to go with your gut feeling, don’t overthink it. Over the long term the data will bear out, even if you rated today a 3 but it could’ve been a 2.

Another way of reflecting is to think about the most significant feeling of the day, so if you had a 2 period and 4 period, which one affected your day more? Did you move on from the bad thing quickly, or were you still dwelling on it even when the good stuff happened?

With practice you can become your own overall mood averager :)

Issue for me as well. I want to track time of day/activity in association with my moods to see if there’s a pattern. I don’t want an average mood for the day, it’s too general. Thanks!

Exist looked like a killer tracking app except for this feature. Then I read Josh’s comments. ‘With practice you can become your own overall mood averager’ Yeah, no. Look into some of the standard psychology tracking protocols especially for anxiety/depression, or bipolar… ‘Changing the mood scale would be a big hassle, probably disproportionate to the benefit received. There are pros and cons to both approaches but we value the simplicity of 1-5.’ Also, no, for the same reasons. Just cuz you guys didn’t care to look into it when you added the feature does not justify your take that the increased gradations are not worth it. There is already lots of data on how to do mood tracking well, might be wise to use it. “not everything that can be counted counts, and not everything that counts can be counted.”

Hey Zach, that’s fair. We don’t have backgrounds in psychology and when we started Exist, many years ago, we looked at what other apps were doing but not what was recommended in academia. Unfortunately I’m not just being lazy when I say it’s a pain to change both the scale and tracking multiple moods a day — I can’t show you the source code so you’ll have to take my word for it! With just me to do all the work across the web app, API, servers, android app, all the data analysis, etc., there’s lots to do and different tradeoffs to make. I could try and make this work, but then everyone waiting for new integrations or caring about different data would be out in the cold for a long time. Whatever I pick, someone is going to be disappointed. I totally get that the current mood rating system isn’t enough for you though. That sucks, because it’s not something we can easily change, so there’s no easy solution. I’m sorry for that.

Just to add in my comments here on why I voted on this feature. I don’t have a 24 hour day and so most of the tracking apps that depend on having a set schedule fail for me. Worse, my actual day length fluctuates between 14 hours and 36 hours in an unpredictable way that depends on some kind of complex interplay between my stress levels an diet.

Not trying to complain, but hoping to shed some light on some of the motivation on why I found Exist to begin with.

My impression is that exist’s original direction was meant more to be an aid for habit building and daily improvement, but others like me are seeing the inference engine as the main feature and want to apply that an arbitrary stream of events to get insight into our struggles.

The kind of person who gravitates toward self-tracking seems like one who is trying to solve problems. It might help to reframe this multi-entry request in terms of those problems as I suspect there may be other creative ways of approaching some of them that might not need an entire rearchitecturing of the app.

I have yet to find a tool that does what exist does but in a time-flexible way. But having looked at this space at length, I fully appreciate the technical challenges that this would create - especially since most of the other 3rd party APIs operate under the same time segmentation paradigm.

Good luck.

I think the problem here is the “it’s just me” thing you mentioned. This topic has had active discussion since 2015, so that should be a STRONG signal that it’s important to your audience. You’re charging for software and expecting people to be ok with the “there’s only so much one person can do” story indefinitely. Like when you were rebuilding the Exist website, that was at the cost of actual functional improvements to the app that everyone is paying for.

I can’t speak for everyone, but I certainly know that when I PAY for software, I expect that comes with a commitment to continuous improvement from the developer. Tradeoffs within application functionality are one thing, but tradeoffs between the app and the website? No one using your existing app should be funding something that only helps you drive new paying customers. Also, when you continue to charge for software year after year after year, there’s an unspoken expectation that you’re reinvesting that money, not into your personal bills, but back into your own capability to improve the software. If that means growing your team to keep up with the demand, then so be it.

There are plenty of active feature discussions, but I think the strong concensus here proves that the use base has a wide-spread concern with the existing feature and a reasonable expectation that it meet the needs better than it does. Just because something is “hard for you to do”, shouldn’t mean “tough luck”–it’s supposed to mean that you find a way to make it happen, even if it means finding help to rearchitect something you built in a way that was far too limited for growth.

That’s the web software business. Either get on board with what the world expects from meaningful, timely continuous improvement in 2018 or stop charging for the software.

Kristin,

You can see our income here. We have two developers and nothing to spare for another. We’re not just cruising along, raking it in and refusing to spend the money on hiring more people! I wish. The reality is that we’re small and probably going to stay that way for the foreseeable future. I can see how that’s frustrating if the additions you’re after aren’t happening — I’m often frustrated that there’s so much left to do as well, and so little time to do it. But it’s disingenuous to suggest we’re squandering some hidden income and not using what we have to fund further development. We make just enough to keep working on Exist.

The web site is not just something that brings in customers, it is a web app in its own right with core functionality. I think you misunderstand what I was working on — I made major changes like new insights, historical browsing, searching, and week and month views, and many smaller changes like renaming tags. Please give it a look sometime! Yes, I put off adding new integrations during that time, because at some point it makes sense to try to make existing data more useful rather than just continuing to add new ways to sync new data. I’m sure you understand this tradeoff.

I can see that, for you, you’d rather Belle and I spend, say, a year overhauling Exist’s architecture to handle intra-day events, making multiple ratings possible, and then adding that functionality to the web app, API, and mobile apps. But I’m not sure everyone else would make that tradeoff and be happy to see no other progress for that period.

We have the annual survey coming up where one question we ask is what your one priority is for development for the next year (and last year, this was not the winner) so perhaps in that survey the numbers will bear out what you’re saying, that this particular change is the top priority for the majority of users. I’m sorry you’re disappointed with our progress.

Kristin, I was really taken aback by the tone of this post and wanted to address it as another paying customer.

I disagree with the assertion that paying for a piece of software requires it to “improve” by whatever metric. When I pay for something I pay for what it currently does and not what I hope it will do later. I expect that the primary use of my money will be to pay for the time it takes to make sure the thing doesn’t break. This means updates and maintenance across devices and backend work on the website. That alone can be complicated and time consuming. Sometimes “continuous improvement” is simply doing this back-end work in order to prepare for a server upgrade or a new operating system.

I DO expect that some of the money that I pay to the software does go to the developers personal bills. 100% of the money paid for a product should not go back into the product because that creates a system where developers can’t actually live.

Josh recognized the need to expand and so he hired Belle (just over a year ago). That’s a significant investment in the company.

As a paying customer, I don’t feel that I can demand new features from a product. I can only demand that it continue working as it does. A portion of the cost of this subscription goes to the recurring costs of operation, including bandwidth and server space. That’s not an insignificant cost.

The way I see it there are two major reasons to want this integration (I know that there are more, but as I’m thinking through the problem these are the two big ones I see): 1. A more objective average of mood across the day 2. A timestamped mood that allows more granular correlations to be found.

The first option might be a more straightforward implementation, but that doesn’t mean it is easy. There would need to be some reworking to the behavior of the app to allow for the prompts and keeping track throughout the day. This would need to be done for the iOS, Android, and mood by email clients. Then the average would need to be taken and decisions would have to be made about what to do if not all entries are filled out each day. Does that make some entries “weigh” more? This could easily be a few months of dedicated work for prototyping code, refining the algorithms, etc.

The second use requires a huge infrastructure of time stamping. It would mean that all external connections would have to pass along the time stamps for when data was entered into the initial program (not when it was passed to Exist). I could easily see this taking a year or more of dedicated work and might not even be possible based on the API’s of the connections.

I also feel that I would benefit from multiple check-ins in a day, and I’m disappointed that it is not the next thing in the hopper, but what I’m reading from Josh is not “tough luck” but “not feasible at this time.” These are very different.

John

All of the big enterprise IT companies are making big investments into IOT and approaches to try and make sense of it all. Just today I heard about a large company building out a new office to do this for a few hundred million.

Josh and Belle could get jobs at a place like that and make more than double what they do now. I am quite happy they choose to work on exist instead.

Josh, sounds like you’ve got yourself a nice monolith going on. Perhaps this would be a great opportunity to break out into a separate service. Could have the service perform the daily aggregations (mean and variance seem valuable) and feed back to the main system for correlations.

Zach, do you have a direction to point towards for what would make a good mood tracker? Gyroscope takes a different approach but theirs feels like details which don’t matter, to your point.

I’d be interested specifically for mornings, just the exact same mood script I currently get but at wake up. I’d use to tag and label dreams to attempt to see how that would correlate.

You can already add tags whenever you like, Rich — sounds like what you’re after is just another reminder set to come first thing in the morning?

Exactly Josh, just twice at configurable times would be great

This suggestion might be a better fit then, as it’s about more reminders, not the larger change of being able to submit multiple mood notes each day.

Would it be any simpler is you linked mood with time entered? Then it wouldn’t be 2+ moods arbitrarily together per day

The suggestion of Alexandra McCoy must be taken into consideration. It would be amazing if you could add a timestamp to multiple mood ratings, but still allow the general mood assessment at the end of the day.

Wow I just wanted to add this feature on the suggestion list. Now it’s appears to be already under consideration. Yay! But since 2016 :-\ - created 2015.

Maybe this idea should be solved by implementing “Manual tracking, or adding numbers to custom tags” ?? There is no need for mood tracking if there is a generic manual tracking (eventually with different existing “templates” like “mood”).

Mood tracking already exists, so adding manual tracking doesn’t change that. This suggestion is for the ability to add multiple mood ratings each day, which would not be solved by manual tracking.

Why wouldnt it help/change that? So if adding tags would allow setting meta data, that would work.

How I see it:

Tags must have the following meta settings:

• description
• multiple flexible reminders settable (e.g. once per day at 23:30) -> would popup the list of predefined values for the label
• Predefined values (with label/value/color structure - then labels could be also icons representing a different value)
• Allow to set/add numeric custom value on relation creation
• Allow to set/add string value on relation creation
• Multi vs Single-Select (for values)
• Group (e.g. Workout; which can contain multiple assigned tags)

With these meta information you have all needed information available to track the mood like now, then you could replace the current hard coded mood (and just provide mood as a tag template).

Adding a relation to a tag for a specific date/time must allow the specification of a time (optional). This way its possible to have multiple trackings per day.

For compatibility you can replace the hardcoded current mood with the a new tag structure (that is created automatically), you “just” need to migrate existing mood data then into the new tag meta-data architecture.

Sorry, none of what you’ve described accounts for tracking multiple values for one attribute for one day. Manual tracking allows the ability to track things like an incrementing value (e.g. a count of coffees) or a duration (e.g. time spent watching TV) as a single total for the day. Multiple moods would need to be stored as separate mood rating events with timestamps which is orthogonal to manual tracking.

I’m currently using Daylio for mood tracking. Would love to be able to have the data from that integrated into Exist.

Dave, Daylio is a separate suggestion and isn’t possible. In general other mood tracking sources won’t be possible, even if they make the data available, as they all use other scales.

Instantaneous mood ratings seem much more reliable than an end-of-day guess, but only with a higher level of compliance from users and good data analysis.

Ideally, the day would be divided into, say, 3 periods, and at a random time within that period, the user would get a notification to enter their mood at that instant, and enter it with 100% compliance. Then we can calculate morning/midday/evening stats, and have overall stats from either an average of the 3, or a separate overall rating. Almost as good would be entering mood at exactly the same time, 3x per day.

In the real world, few users will do this. Can we guarantee the data is still free of confounding factors when there is e.g. low compliance with mood rating, or entering mood ratings at varying times? Until my own compliance with daily mood tracking was over about 80%, I didn’t trust the correlations. So it seems there will be additional data, but a high barrier to users and developers to make it useful.

Take a look at Moodnotes. While it is a separate application, it does this very well. I was really hoping Exist could be more adaptable and grow, I am realizing their solution is to remain simple and set for they want to deliver. Nothing wrong with that, but it might not fit with those looking for solutions like this.

Disagree. Here’s a blog post about how we’ll be spending all of 2020 working on being adaptable, rather than remaining simple 😉

Any news about this?

We’re working on the manual tracking updates this year. If anything changes, the task will be updated to reflect it.

Any update on being able to add multiple moods per day?

As Josh mentioned in his last comment, if there’s an update about this feature we’ll update the task to reflect that. We’re still working on manual tracking changes, so we have no plans to move forward with this suggestion at this stage.

I think if anyone asks again, I’ll lock the comments. No need to keep asking questions that have already been answered literally right above.