Apple Health vs. Fitbit

A new feature of iOS 8 is Apple’s Health App. It’s a way for users to view any health data that has been collected by in-built sensors in the device itself (such as step counts from the phone’s specialised accelerometers), along with data that can been added by third party apps (such as your weight, as recorded by a set of smart scales).

The dashboard that Apple supplies is deliberately basic. Everything that can be graphed is graphed, albeit in a very stripped-down way; no scrolling, no trendlines, no regular axis labels, and so on. They’re so spartan that I’m not really sure why Apple included them at all. But the true purpose of Apple Health is not as a pretty dashboard, but rather as a way for apps to share health data with one another.

In Apple’s world, users of a running app wouldn’t have to manually enter their weight in order to calculate an accurate ‘calories burned’ figure; instead, they would authorise the running app to access their Health data so it always has the most up-to-date weight information. Likewise, the running app would synchronise its calorie burn information back to Health so that (for example) a dieting app can have a better view of calories out vs. in.

None of this happens automatically. Developers must specifically build in ‘Healthkit’ functionality, and while Apple may have hoped that everyone would eagerly jump onto their bandwagon, many of the most popular apps have been dragging their heels. The reason is that while adding Healthkit functionality isn’t particularly difficult from a technical perspective, it poses troubling business issues for some companies. Take Fitbit, for example. Why would they contribute the data their pedometer collects – step count, distance count, floors climbed, and soon, heart rate information – to Apple Health when it could result in their customers using a non-Fitbit app to view that data?

Not only do they lose control of the customer experience; not only do they lose the ability to sell their Fitbit Premium subscription; but worst of all, they become commoditised. They’re just the same as any other cheap pedometer, because as far as the customer is concerned, all they are is a bit of plastic that sends bits to a phone.

Fitbit has even more cause to worry with the iPhone 5s and iPhone 6, since both phones include specialised accelerometers that allow them to record steps with effectively zero battery consumption. Indeed, the iPhone 6 includes a barometer that allows it to record floor counts. Theoretically, this means that anyone who owns those phones has absolutely no need of any dedicated pedometer device, Fitbit or not.

In practice, Fitbit is still doing fine. People still buy their devices, and I still use my own Fitbit. Firstly, the data appears to be more accurate:

All the screenshots in this post were taken at the same time; you can see that on Saturday 25th October, I walked a lot of steps and climbed a lot of flight-equivalents

Now, while most health professionals will tell you that consistency is more important than precision when it comes to step counts (i.e. it’s more important to know that you’re doing 20% more steps than yesterday, rather than knowing you did precisely 1000 more steps), it’s still nice to see your steps tick up reliably. As for floor climbing, Apple’s sensors woefully underestimate the true count, which is disappointing. But two things are even more important than precision.

One: The Fitbit is almost always with me, clipped to my belt, while (amazingly) I don’t always carry my phone with me; hence more complete records.

Two: Viewing my step count on my Fitbit takes about three seconds. On Apple Health, it takes more like ten seconds (although I could probably get an app that might accelerate that). So I look at my Fitbit more frequently.

Having said all of that, the Apple Watch will eliminate all of Fitbit’s advantages in terms of accuracy and accessibility (due to its fixed position on my wrist) and I suspect that will be the end of my Fitbit-using days.