Strava App Sensor Pairing Is Now Dead: My Last Workout Taking Out The Trash



As of today, the Strava app will no longer allow you to pair to sensors like heart rate straps or cycling cadence sensors. Instead, Strava is now actively recommending you download other apps instead of using Strava’s own app. Well, I suppose technically they started recommending using 3rd party apps instead of their own app last month. But today’s the day it matters. We all knew this was coming, but I thought it notable to get one last recorded workout with it late last night just 90 minutes before the feature disappeared. But more on that in a moment.

The reason for all this is that Strava says that when people used their own app with sensors paired, it crashed a lot. Too much apparently. In fact, it even crashed when people didn’t use sensors, because…Strava? I don’t know. Their words, not mine:

“Supporting Bluetooth devices was causing the app to crash during recording, whether or not a sensor was connected. Disabling this feature significantly improves recording stability for all athletes.  While this update will reduce crashes for millions of athletes, we understand that it will be disappointing for the athletes who do use the feature.”

Now, if you record your sensor data (like heart rate or cadence) on a Garmin or Wahoo or some other device – this won’t actually impact you. You’ll still be able to display all your data in all the same ways as before. So for example, if you have a Garmin Edge 500 paired to a Wahoo TICKR strap – that’ll still show up on Strava just fine. And if you record your running power on your multisport watch, that’ll still dork up your cycling power curves on Strava – just like before. See, Strava’s an equal opportunity sensor-killer here.

[Preemptive note before we continue: Yes, I’m being hard on Strava. As all of you have overwhelmingly told me – they deserve it. And I agree. But it’s not because I want them to fail. In fact, quite the opposite. There’s no fitness platform I actually enjoy using more than Strava – I’ve uploaded some 3,780 workouts to it over the last decade. I very much want them to succeed. I want them to find a way to becoming profitable so they’ll still be around. I think they’ve largely got great employees there, but it’s just hamstrung by poor executive vision and/or execution.]

How’d we get here?

Of course, this all started back on Aug 28th, when Strava told users that they would be cutting off new users from adding sensors to the app. At the time, it read (or, I interpreted the “will still be able use them “ part) that if you already had the sensors paired once, you were grandfathered in. Apparently that wasn’t the case – at least forever. Here was the original tidbits:

“As of August 28th, 2019, Bluetooth and ANT+ sensors will no longer pair directly with the Strava phone app (the Strava Apple Watch app can still record heart rate from sensors paired with the Apple Watch). This means that, for example, if you record activities using the Strava phone app, you will not be able to record your heart rate or other BT sensor data. Supporting Bluetooth devices was causing the app to crash during recording, whether or not a sensor was connected. Disabling this feature significantly improves recording stability for all athletes.

While this update will reduce crashes for millions of athletes, we understand that it will be disappointing for the athletes who do use the feature. Athletes who used sensors before 8/28 will still be able use them – we’re hoping a little extra time will give them a chance to try another way to pair their sensors or upgrade to a different device. For all other athletes, those who have never connected a sensor to the app or those who sign up for Strava in the future, Strava will no longer support these sensors after 8/28.”

Then, earlier this month (October) Strava decided it needed to clarify things – and with that, it would be cutting folks off entirely at the end of the month. Here’s the applicable part of that statement:

“When you record a workout on Strava, it’s our job to make sure your effort is captured accurately and reliably. We recently discovered that pairing Bluetooth heart rate monitors and power meters directly with the app is causing Strava to crash for millions of athletes — most of whom don’t even use these kinds of sensors. So to significantly improve the stability of the app, we’ll no longer support direct Bluetooth and ANT+ device pairing as of October 28th, 2019.”

This particular snippet is baffling on so many levels. Maybe it’s baffling to me because I used to work in a large corporation that carefully crafted how to communicate with millions of users. Maybe it’s baffling to me because it openly implies so many non-savory things. Maybe it’s baffling to me because my job these days is sifting through piles of PR statements from companies like Strava (Strava themselves stopped sending me PR bits last year, apparently upset about my coverage). Maybe it’s baffling to me because I worked at one of (if not the world’s largest) software development companies for over a decade – so the tech side I understand well. Who knows. Others with extensive BLE & ANT+ sensor experience in the industry agree.

But, let’s break it down a bit. First, this part:

“We recently discovered that pairing Bluetooth heart rate monitors and power meters directly with the app is causing Strava to crash for millions of athletes — most of whom don’t even use these kinds of sensors.”

It starts off saying they “recently discovered”, as in, they never bothered to look at the crash log data they’d undoubtedly been collecting for years. Why would anyone ever phrase it this way publicly? It goes on to say “…causing Strava to crash for millions of athletes”.

Holy cow – you actually just wrote that out and sent it to 46 million Strava users via e-mail? You actually just admitted your own app crashes millions of times due to poor coding? Did the PR firm get drunk that night, check-out, and then start a pool on creating the worst possible PR statement from a sports tech company this decade? I can’t actually think of any that top the nuance and self-admission failure in that one.

Here, for future reference – this is how you should properly write that sentence in bland PR-speak that minimizes your faults:

“As part of Strava’s ongoing commitment to app stability, we’ve determined that a small number of users may be seeing Strava mobile app issues due to Bluetooth Smart sensor pairing dependencies.”

See, that didn’t admit that you weren’t paying attention at the helm handlebars. It didn’t openly admit that you were causing millions of people’s apps to apparently crash, and it didn’t throw your development team so far under the bus that your app crashes on sensors that people don’t even have paired, or are using.

And, if they wanted to provide non-PR speak statements, they could have shown just how few people were actually being impacted – by going into specifics like saying “Only 0.x% of of all Strava app uploads utilize sensor connectivity”, thus helping the 99% (or even the 1%) understand why Strava’s resources could be better spent elsewhere. [Note: This paragraph added to help clarify why I think their PR statement was such a blunder.]

But that gets us to the next part of this notice:

“So to significantly improve the stability of the app, we’ll no longer support direct Bluetooth and ANT+ device pairing as of October 28th, 2019.”

Here, let me explain what they just said from a developer standpoint:

“So, because we can’t be bothered to figure out how to fix it, we’re just gonna kill it entirely later this month.”


Does it actually matter?


But does this actually matter to people? Well, that’s a very valid question.

Strava seems to imply it impacts “millions of athletes”. For reference, as of last month, they had 46 million users. Of course, Strava doesn’t disclose how many of those are actually paying members. It’s largely an unknown, but it’s clearly nowhere near the majority. In fact, paying membership is largely estimated to be in the low single-digits – between 1-6% based on multiple sets of data.

Still, I have no reason to doubt Strava’s statements here. Mostly because most companies would never actually write in a public statement that they crash millions of athletes. It’s not really something you’d normally want to disclose. But remember, that’s crashes – not actual usage of sensors. And again, only Strava knows that.

Take for example myself. Aside from last night’s final hurrah, the last time I used the Strava app with sensors was a few months ago this past summer. Given I routinely take 3-5 GPS watches or bike computers out with 3 power meters and 2-4 heart rate straps, I’m not entirely the target market for this feature. Yet at the same time, sometimes I am. Sometimes I just wanna record something quickly – like being on a hotel spin bike and having a heart rate strap with me.

2019-10-27 22.11.14 2019-10-27 22.11.42

Which is ironically, the exact casual fitness type segment that Strava says they want to court – their new CFO estimates that addressable market to be about 700 million people in total. If one looks at Strava’s latest feature release last month – it’s all about having users assign a perceived effort to an activity. This allows them to start figuring out how hard a workout is without relying on sensor data. In turn, they could develop products or services based on that, perhaps enticing you to become a paying member.

Still – users overwhelmingly upload runs and rides. Last year as part of Strava’s annual report, they noted that some 756 million rides and runs were uploaded (284m runs, 472m rides). Strava’s CEO James Quarles notes in an interview with Outside Magazine that Strava’s fastest-growing sport is running. Perhaps because they started charging bike GPS computer makers to enable features like Strava Live Segments on their devices, causing some companies to ditch plans there. Those users would be less likely then to pay for Strava Summit, which is required for functions like Live Segments.


As for the overall number of uploads per year, they haven’t specified exactly the quantity. However they did say on average “15 million per week” – which puts it at 780m activities (of which, 756m are rides/runs). So roughly 97% are run/ride. The remainder you need to piece-meal from the annual report. For example they disclose that men uploaded 20 million swims (the 3rd largest activity for men on Strava), while the 3rd largest activity type for women was walking, with 9 million walks. In total that gets you to 785m activities – more than the 780m activities Strava said occurs in a year.

And that’s before we even talk about the rest of the 32 sports profiles that people use, and undoubtedly there are tens of millions more within those categories. I guess math isn’t entirely Strava’s strong suit here.


(Note: The 36 million athletes, like all the upload data, is from their 2018 report. They add roughly 1 million users per month, bringing us to the 46 million users they publicized last month.)

But here’s the thing: I could spit numbers at you all day long. What’s far harder to do is the intangibles. Nearly any public comment or tweet I make about Strava – even positive ones – is met with upset users about lack of new features for month after month, year after year support. Or bugs that haven’t been fixed for years. In fact, the September update I mentioned above was the first time in *years* Strava has pushed out a new feature (excluding the one last January they got caught duplicating).

And feature removals like this always results in people claiming they’re cancelling or have cancelled their memberships as a result. Whether or not they do is frankly beside the point.

Strava has a perception problem, and things like this sensor connectivity debacle only add to that. If Strava wanted to remove this feature the ‘right way’, they should have highlighted (in proper PR speak) that instead of fixing their app and spending untold development hours dealing with clunky sensor connectivity across thousands of phone and sensor combinations, that they were going to invest that time into new features and new functions for their users. But they didn’t say that. They missed that opportunity, yet again, to communicate how they were going to make paying members’ lives better. Instead, they just took away features.

My Last Workout:

For my last Strava sensor workout, I decided I needed to do something special. Should I go for a ride, or perhaps do a run and draw a big ‘S’, or maybe just a big minus symbol? Maybe I should follow Strava’s market thinking around those 700 million people and do a yoga session or something, recording it with the app and a heart rate strap.

And then it hit me: I needed to take out the trash!

So I grabbed the Wahoo TICKR strap that I had sitting in my backpack (just in case, always practice safe strap behavior). I cracked open the Strava app on my iPhone and ensured it found the strap, which it did immediately via Bluetooth Smart. No crashes for me here.

2019-10-27 22.16.27 2019-10-27 22.16.30

I suppose in many ways, our household is just like Strava, I needed to get rid of the waste. Offload the things that might attract bugs or cause failures if left unattended.  Off I went to the garbage bin down on the street.


Of course, this wasn’t a long workout. It only lasted 2 minutes and 4 seconds.


But my Bluetooth Smart heart rate strap soldiered on and captured my heart rate without any issues. Smooth and consistent operate it (or I) was.


But it turns out I’d unlock myself a Strava badge of sorts on this workout. Look at that picture of the trash bin again up above.  Sitting off to the side of the garbage bit was a bicycle pump (perfectly functional). No kidding, welcome to Amsterdam! Dutch valve and all.


See on Sunday nights, people stick all sorts of random things out near the trash bins for other people to pick up. Everything from furniture to toys to bike parts. Especially bike parts. If not picked up by Monday morning, the city hauls it away. A surprising amount of stuff gets picked up by others in the wee hours of the morning.

As for the pump – sure, there’s no gauge on it. Nor any data metrics. And thus, as I do the work it inflates to some unknown value – but if I want to know anything about my ride, I’m gonna have to turn to using other apps or devices.

But hey, it looks like my Strava membership’s good for something after all: I got myself a free bike pump.

Thanks for reading.


Latest posts