Sleep logs via IFTTT contradict app data

[DanielAustin]

I use the IFTTT applet to log the sleep data from my Charge 2 in a Google sheet. However, the data being logged in the spreadsheet does not correspond with that being displayed in the Android app and online dashboard.

The sleep start/end times are always identical, but the Time Awake and Total Time Slept are always significantly different, and the Number of Times Awake is always drastically higher in the app than the log. It's as if the data generated for the log is using a different algorithm from that in the app/dashboard.

For example, the information from my last two sleeps:

FROM THE LOG VIA IFTTT:
Monday: Slept 01:06 to 09:34; time awake 29m; time sleeping 7h59m; awake count 1
Sunday: Slept 02:02 to 08:40; time awake 23m, time sleeping 6h15m; awake count 0

FROM THE APP AND DASHBOARD:
Monday: Slept 01:06 to 09:34; time awake 1h11m; time sleeping 7h17m; awake count 44
Sunday: Slept 02:02 to 08:40; time awake 49m, time sleeping 5h49m; awake count 29

Why is there this big discrepancy?

[RobertoME]

Hello @DanielAustin, welcome aboard to the Fitbit Community! I was reading your post and this is an interesting question.

Apparently the difference by the data you have shared here, is located in the awake count. It is possible the algorithm or the recipe created in IFTTT is not recognizing this time; which by consequence might be causing this discrepancy. I presume this is related to your sleep stages; given the fact you have a Charge 2.

However, I'm wondering how you are seeing your sleep logs in your Fitbit Dashboard? Do they look correct or is the information inaccurate too? It is possible that your sleep logs are incorrect which will explain why the data triggered in IFTTT is different as well.

In either case, I would recommend to contact directly IFFFT customer support for further review. As Fitbit, we do not have access to their data to find the source of this issue.

Stick around if you have more questions or if your sleep records in Fitbit are not correct.

[DanielAustin]

Thanks for your reply @RobertoME. I have made some advance in understanding the issue here. It does indeed seem that the two sets of processed data (ie: the stats viewed through Fitbit's phone app and computer dashboard vs the stats logged via IFTTT) result from two different algorithms that differ in the way they count the amount of time and number of times awake during the night.

There is a Fitbit help article called "What should I know about sleep stages?" which says:

"If you previously tracked sleep with Fitbit, your log included the time spent awake, restless, and asleep. We’ve fine-tuned the mechanism we use to track sleep in order to estimate your sleep stages by including heart rate and other data. As a result, we’ve combined the time you spend awake and restless into total awake minutes to give you a better sense of your sleep cycles. You may notice more awake minutes in sleep stages than in your previous sleep data due to this change."

It does not say how recently this was written or when that tweak was made, but it seems clear now that the problem is that the app/dashboard shows the new algorithm stats while IFTTT is still logging stats produced by the old algorithm.

The question, then, is has Fitbit already updated their backend API to output the new-version algorithm stats? If so, then we need to get on to IFTTT to get them to update their applet.

[arovik]

This is still not fixed:(

[SunsetRunner]

Has anyone by now managed to get correct sleep data from IFTTT? I am hoping to use to keep a sleep log in the DayOne diary app. But it is posting wrong numbers.

[enigma17591]

Just found this thread in 2022. Unfortunately fitbit still hasn't updated their backend API to give us the new stats, so the IFTTT data isn't very useful.

[JoelMalissa]

I also just found this thread.  Same problem.  I switched from the Amazon Halo to Fitbit purely because of the ability to more easily export sleep data.