JUMP TO:

Garmin Acquires Firstbeat Analytics: A Quick Analysis of the Winners and Losers

It was about that time of year for Garmin’s fitness or outdoor business to acquire someone else. It seems to be roughly a yearly tradition, for example last year was Tacx. The year prior, the remains of a bankrupt solar wearables business. This year’s entrant? Firstbeat.

And, if there were ever a smart business acquisition to be made – this is probably it.

Who’s Firstbeat?

You’re undoubtedly familiar with Firstbeat’s work, even if you’ve never heard the name. They power the vast majority of Garmin’s fitness and outdoor training-focused algorithms. For example, on the Garmin Fenix 6 there are some 18 different Firstbeat licensed algorithms being used to cover all sorts of features from training load to VO2Max and respiration rate. And that doesn’t even include another 2-3 modules licensed for sleep two weeks ago.

Here, just look at this list:

image

Even Garmin’s lower-end wearables license features here too – five features for the lower-end Vivosmart 4.

However, over the past few years, especially the last two years, Garmin has more and more heavily leveraged Firstbeat’s features. Be it on the wearables front or the cycling devices side, these were undoubtedly becoming a significant competitive differentiator for Garmin. Not to mention, also becoming a cost incurred for each product in fees.

But more challenging for Garmin would be that Firstbeat was also giving competitors a chance. Sure, some companies leveraging Firstbeat weren’t likely to be mainstream competitors to Garmin. But others were. For example, Suunto has long leveraged Firstbeat analytics algorithms. And more recently we’ve seen Casio do so with their latest solar-powered GPS watch that starts to encroach on Garmin’s Instinct and Fenix series lineup. Others, like Huami’s Amazfit or Huawei’s wearables, while mostly not competing in North America, were leveraging these features to become more serious competitors in Asia.

image

So, one could see how a move to simply buy up the company, for what was probably a relatively cheap price, would make sense. Not only does it eliminate the licensing cost incurred (which means higher-end features could likely more readily be fit into the budgetary constraints of lower-cost devices), but more importantly for Garmin: It could cut off access to metrics for their competitors.

And it’s been clear lately that wearable companies are more and more heavily leaning on Firstbeat to do the heavy lifting of getting these metrics correct. The bar has been raised across the wearables industry that simply using basic napkin guessing math for certain physiological metrics isn’t a substitute for actually measuring it. Which isn’t to say Firstbeat is perfect. I often find quirks in their algorithms.

But really, about the only other company operating in this space in developing their own algorithms at scale here is Polar. That could be good news for Polar in that they’ve developed all of this IP in-house, and won’t potentially lose access to it down the road.

IMG_1659

Now, the big question is: Will Garmin cut-off the licensing of algorithms to all these 3rd party companies? Or, will they instead see it as a profit center?

Well, in Garmin’s press release on the announcement today they stated: “Firstbeat Technologies will continue operating its Wellness and Professional Sports businesses.”, as well as noting that “Firstbeat Analytics is headquartered in Jyvaskyla, Finland, and will remain operating in this facility. Financial terms of the acquisition will not be released.”

Keep in mind that Firstbeat technically has a few business lines. First, they license algorithms/modules to companies (that’s the biggie), second, they have corporate wellness programs that support companies trying to get employees healthier, and third, they have a professional sports business focused on teams.

(Update heads up – Garmin has clarified they only bought the licensing piece. I’ve updated the post accordingly.)

Neither the corporate wellness nor professional sports sites were bought/acquired by Garmin, only the consumer product licensing portion. That’s notable because neither of those two components contribute hugely to the algorithm side. They have historically worked closely, especially the pro sports site, on validating algorithms however. So it’ll be interesting to see how that relationship continues going forward.

But let’s get back to the core here: What about licensing to their competitors?

So, I asked simply: Will Garmin continue to license the technology to others?

And, Stephanie Schultz (Garmin’s Fitness PR lead) confirmed back that “Yes, Firstbeat will continue licensing to 3rd party companies.”

The question is – how long will they do that before they decide it’s worth more to keep things from being licensed? Or, maybe they decide there’s enough money in licensing to smaller wearables that aren’t going to compete with Garmin?

Going forward:

Ultimately, acquisitions tend to fall into two basic questions:

A) Is this good for the two companies involved?
B) Is this good for consumers?

In the case of the first, the answer is obvious. It’s undoubtedly good for Garmin. And it’s a move I’ve long assumed they’d make. It’s also probably good for Firstbeat, otherwise they wouldn’t have taken up the offer.

The next question though – is it good for consumers? Well, probably not long-term.

My guess is that it’ll be good for Garmin consumers generally. That’s because I suspect that Garmin requests were probably driving a lot of the R&D at Firstbeat anyway (directly or indirectly). However, there may be cases now where money isn’t spent on R&D for something because it’s not interesting to Garmin.

The large piece though is that regardless of whether or not Garmin says they’ll continue licensing to 3rd party companies, it just doesn’t make any business sense to. Garmin undoubtedly is losing wearables sales to 3rd party competitors because of these features being present in those watches. That’s absolutely a fact. The licensing fee cost of pennies or dollars per feature isn’t anywhere near the revenue compared to the profit of selling a single wearable.

As such, it’s bad for consumers to have less competition in the space – even if it might be good for Garmin consumers specifically. And heck – now there’s zero excuse for people’s uploads to Garmin Connect from Zwift or TrainingPeaks not to count in their Garmin device’s Training Load (which is powered by Firstbeat). So, that might be good for us. Right?

But, I guess we’ll have to see where it takes us from here.

With that – thanks for reading!

Post a Comment

Your email address will not be published. Required fields are marked.
If you would like a profile picture, simply register at Gravatar, which works here on DCR and across the web.

You can click here to Subscribe without commenting

Add a picture

*

99 Comments

  1. Occamsrazor

    The argument previously as to why some new FirstBeat analytic features couldn’t be added to older models was because it would cost Garmin to do so due to licensing…. so now that argument has gone with the acquisition, what can we expect?

    • inSyt

      Yes the Forerunner 645 should receive body battery? And the sleep algorithms on Garmin Connect should be replaced with Firstbeat’s algorithms?

    • ACtavian

      Honestly, regulators in Finland and the EU need to take a look at this deal. Garmin is cornering an entire billion dollar market.
      And for all of you who own Garmin devices and are hoping for backported features for your 1 year old device: You are delusional.
      You matter to Garmin at one point in time:when you make a purchase!

    • inSyt

      Right now we have Amazfit, WearOS, Garmin, Polar, Suunto, Apple, Samsung and Firstbeat all good at something, but all with their drawbacks. Maybe some consolidation is a good thing?

    • dikamilo

      Garmin Sleep algorithms are already replaced by FirstBeat on beta software for Fenix 6 series.

    • John

      @ ACtavian My two year-old Edge 1030 got a bunch of new features just this Spring, mostly matching features in the newer Edge 830/530.

    • ekutter

      There’s way more cost to adding a feature to an older product than just the licensing. It’s not like you can take a module and just plug it in. There is significantly development and testing resources involved in any changes like this. And those are resources that get taken away from newer devices. I doubt it would make any business sense for Garmin to add these to any product where a new one is already out or soon to be released.

    • flokon

      Agreed. While Garmin might put more of FirstBeat’s features in lower tier models, they won’t do so retroactively but going forward.

    • The deal is presumably big enough for it to be under EU rather than member state antitrust jurisdiction. It doesn’t appear to me that there any significant market overlaps between the two companies’ products that would give rise to any antitrust issues. This looks to me like a supplier – FB – being acquired by a customer – Garmin, a classic case of vertical integration.

    • Samu

      If the bring sleep to the 245 i buy it tommorow. But i guess they wont und will. We have to wait for VA5/Venu2 and so on. Unfortunately garmin does not relay care about entry level. Venus lag with the most watchfaces is not repaired since month. Battreie drainer are there since month/weeks.
      They own forum is full of that

    • JimC

      Up to a point it does make business sense to add new funcionality to older models – I’m far more likely to buy a product if I think it’s going to be supported and updated for a reasonable length of time. (This is quite a major selling point for iPhones vs. Android, for example). Of course if I’m happy with the product I’m less likely to update, so that’s something the company needs to weigh up.

    • Andrew

      Do we think Firstbeat’s revenue will be high enough?

      I would however argue there are obvious overlaps. Garmin has historically developed and still self-supplies various fitness metrics, body monitoring etc (e.g. sleep being an obvious example). Given that basically all Garmin’s devices will have at least some Garmin in-house solutions they will prima facie have a share of supply of something over 25% – Firstbeat will also be active as a supplier of these type of metrics so there does appear to be a prima facie horizontal overlap.

      The theories of harm would therefore be:

      1. Lost innovation and slower development of these health traffic metrics. The number of companies capable of doing this credibly going from three (Polar, Garmin and Firstbeat) to two (Polar and Garmin/Firstbeat).

      2. Vertical foreclosure that Garmin could refuse to supply and/or supply on unreasonable terms some or all modules to competitors, notably Suunto. This would then have a downstream impact on Suunto’s ability to compete therefore strengthening Garmin’s already very strong position in the sports technology devices market (effectively limited Suunto’s ability to compete).

      I’m not saying that either of these will be enough, to cause trouble but if I were wanting to cause trouble on this as Suunto, or I was a regulator that wanted to probe this, that would be my approach and I’d be pretty confident I could get commitments out of Garmin on licensing and equal treatment to other device manufacturers (because quite simply by refusing to give such commitments Garmin would in effect be confirming the theory that they intended to foreclose the likes of Suunto).

    • Dan G

      inSyt, I would *love* the Forerunner 645 to get Body Battery (and have made a “feature submission” to Garmin saying that—others should do that too).

      Fingers crossed.

    • Dan G

      Suggestion, not submission.

  2. Enrique M

    “now there’s zero excuse for people’s uploads to Garmin Connect from Zwift or TrainingPeaks not to count in their Garmin device’s Training Load”

    +1 (well said!!!)

    • RSTL13

      Absolutely agree here! This is a potential big win if Garmin does this.

    • John D

      Totally agree. Hopefully the acquisition will fix Physio TrueUp oddities across devices, like changing it so you can have one training status based on all devices, instead of separate statues for separate devices.

      Plus finally allow me to record a Zwift workout and write it to Strava (for map, elevation, segments, etc.), and concurrently to Garmin Connect (same data, now with Firstbeat analytics applied to it). Because the current is crazy: Record the same workout via Zwift and via the Garmin device(s). Zwift for the aforementioned reasons, Garmin Connect for the analytics. And because I automatically send from Garmin Connect to Strava, I then have to delete the duplicate activity in Strava

    • Nick Wall

      Same workflow here. Fixing this would be a big UX win

    • Tommy

      More to the point, I should be able to leave my fenix behind while riding with my edge 830 and still earn intensity minutes

    • Dan G

      AIUI you don’t see FirstBeat metrics without using a Garmin to record the session because the FirstBeat stuff uses HRV data collected while you’re working out. Zwift, TR etc don’t and can’t get HRV data. I don’t see that changing, nor the problem of recording workouts through a Garmin while an app runs your workout.

  3. Just a couple of points from a (former) M&A guy:

    1. If they do plan to terminate the third-party licenses, they will have to comply with the terms of those licenses. They will have notice periods to comply with etc. They will also almost certainly have “change of control” provisions that would allow the licensees to terminate on a change of control (ie acquisition) of FirstBeat, but clearly the licensees would be shooting themselves in the foot if they did so without an alternative ready to go.

    2. For tax reasons, Garmin will probably continue to license from and pay fees to FirstBeat, but that will just be “circular money” – a bit like transfer-pricing between subsidiaries. It can be an internal accounting nightmare, but is usually a necessity from a cross-border tax point view.

    3. This sounds like it would be a fun deal to work on – due diligence and integration planning in Finland? Sign me up! I did a big deal there about 30 years ago – lovely people, great food!

  4. Chris

    Looks like that Firstbeat page listing the devices (link titled “But others were”) has already been taken down! I can’t see it on their site anywhere now. I guess that’s a sign of things to come.

  5. Dave

    Garmin can now better integrate Firstbeat into their future devices, so lets all hold out for Fenix 7.

  6. Joris

    Hi Ray, have you ever heard of the Belgian start-up Trenara (a running app/virtual coach)? There algorithms seems to be pretty spot on. They use Strava data to predict finish times and adjust the personalized training plan. They got an integration with Garmin as well. Definitely worth checking out 😉

    • Nico

      Do you happen to be the “goede vriend Joris” mentioned on the website? If so, maybe worth mentioning “they” is actually “we”?

      Never heard of it, will have a look…

      On topic: I’m afraid you are right Ray: probably a good move for Garmin, potentially also for Firstbeat but in the end not so much for us (saying this with a big pile of Garmin gear lying around in the house).

  7. inSyt

    Firstbeat Analytics has made big changes to their website. Devices are no longer listed on the website. This is not a good sign for 3rd party companies.

  8. Daniel

    What about licensing Firstbeat technology of last year to competitors?
    That could actually make the most sense for Garmin.

    • usr

      This. Driving away third party licensees would be a short term win, but competitors would either start building in-house expertise or fuel the growth of a competitor who will likely step in faster than firstbeat contracts are winding down.

      If they can keep third parties aboard however, Garmin will forever remain “first amongst equals”. “Garmin algorithms are so good that even their competitors are licensing from them” would be a huge sales driver. If I was Garmin I’d actually consider lowering fees for third parties.

      ANT+ remained open after Garmin bought Dynastream, so licencing to competitors isn’t foreign to Garmin.

    • ANT+ is a good initial example, but these days much of the feedback I get from Garmin’s competitors is that ANT+ mostly only moves when Garmin asks them to. Whether or not that’s true, it’s probably debatble, but it’s a sentiment I’ve heard more and more frequently in the last year or so.

  9. Matthew

    If only Suunto would work with polar To deliver great analysis. Or if polar would open up Flow for other devices to import data to. As you say, could be great for Polar if they play it right.

    • inSyt

      Polar is probably smiling right now hey. Just wish they could add support for custom watch faces as this will open them up to a much larger market. Suunto’s best bet would be to find a way to get a battery saving screen that plays nice with WearOS.

    • Yeah, I think at this point (well, even before today), I really think Suunto’s best play at this point is to mostly give up on the Suunto 9-type devces. They just aren’t competing in any way/shape/for with Garmin or Polar (or barely even COROS anymore).

      However, I think focusing on the Suunto 7 and in particular Wear OS actually makes a ton of sense longer term. It is risky though to bet your entire company on another competitors operating system. But honestly, I’m not sure Suunto has a choice anymore.

    • Brandon

      I actually think Suuntos moves with SuuntoPlus make more sense for them long term then android wear. Linking right into Strava or training peaks on a watch is something fairly unique and given the amount of people who subscribed to one or the other that’s a big potential customer base for them. Now garmin has the live relative effort field for Strava but it appears to work for about 20% or less of people.

      On top of that the Strava effort system and the Training Peaks loading algorithms seem to tell a better overall picture of training effort then firstbeat/garmins training effect.

      Training effect only looks at what peak effort you hit but the total training load for the session is calculated in the background using some unknown formula that goes off position of the moon, sunrise time and some other voodoo apparently being it’s not disclosed (they say epoc factors in and Im sure training time does too but without knowing your Epoc for the session you can’t begin to know what your training load is during the session, and if you can’t do that then what’s the point in using it.)

    • But SuuntoPlus is nothing but marketing. At least today.

      Strava’s Relative effort was introduced on Garmin wearables so many years ago that they’ve already stopped doing it because nobody used it. And the TrainingPeaks algorithms have been on Garmin devices for I’m guessing 7 years now?

      I appreciate Suunto finally added the TP metrics, but man, far too little too late. And of course nothing has happened in months at Suunto (at least with any output to the public), because they’ve put piles of employees on furlough.

      I had really hoped we’d see a quick succession of Suunto 7 updates to remedy the relatively easy problems it had. But instead we’ve seen a single update in 6 months (to the date since the Suunto 7 started shipping), which literally just re-arranged a few fields.

      It’s super frustrating because they were *THIS CLOSE* with the Suunto 7, and then botched it.

    • Funny update to my previous comment. I just looked – Suunto actually just released an update today to the Suunto 7. Seems mostly minor, but will need to dig into it.

      connect.garmin.com/modern/activities

    • Brandon

      I find Strava live effort incredibly useful as it helps me know in the moment how much effort I’ve accumulated. What’s the point of having any kind of stress or recovery data if you don’t also have a system in place to show how much workout effort you’ve given at any point in your workout.

      That’s where I think polar and garmin miss the boat because it can say Im recovered or i should do a light workout today but unless you follow polars daily planning what does light or heavy training load mean. If know my recovery score is 40 for the night (taking that from my Biostrap) then I can use my average Strava relative effort score to factor in ok I should probably only shoot for this many points in training effort today. Isn’t that the point of having all this biometrics? So that it can be actionable intelligence? If not then it’s all just for show (not just SunntoPlus)

      Also I should have said that my overall point with Sunnto not being harmed by losing any firstbeat rights in the future is that Sunnto doesn’t leverage data from firstbeat very well at all. The stress data isn’t saved anywhere on the app so no way to look at long term trends and the sleep data shows you the bare minimum information (it doesn’t as you know even have a timeline as you know to see if your night time awake times are even correct).

      I will completely agree though that Suunto’s software update efforts are pathetic. I really thought with them shutting down movescount the Suunto app would get a lot of updates but it’s still the featureless piece of garbage is been for sometime. It’s weird too because their watches capture all this data but then the app doesn’t do anything with it. They’ve got your training effect your epoc and all this heart rate data but can’t do a simply training load feature (both short and long term). It’s just such a waste of resources to me. Really feels like in the training smart watch arena it’s down to Garmin and Polar and both of them just waiting for Apple to say ok that’s it we are taken over the health wearable market.

      Really it might be between Apple and garmin as polar gives you no reason to wear their watches outside of training or sleeping. The rest of the time it’s just an expensive step counter as neither captures nor displays any other useable data.

    • Scott

      Qualcomm’s announcement today regarding the new Snapdragon 4100 soc mentioned a new wearable from Suunto… 3,5,7 and 9 are taken? A 7+ on the way?

    • Nice find. Just dug in and found the quote (attached). Though, I can’t see how it’d be before 2021 given it implies they don’t have the chip yet. Maybe a CES 2021 announcement?

    • This time attached for realz.

    • Eli

      Maybe just a quick update of the platform and not fully take advantage of it? The new hardware is faster and uses less power then the old so easy to do almost all the dev work without access to the new chips

    • Yeah, it’l be interesting to see. My only concern is power savings are noted as “20%” compared to previous chips, which isn’t really want we wanted to see. But, that’s including the huge uptick in performance. And depending on how you read it, also including an always-on display (this part is muddy).

      So, we’ll see. Ultimately, the Suunto 7 (or whatever) needs to be able to clear roughly the 24hr marker in 24×7 mode with 1hr workout and an always-on screen. I don’t need it to go like 30 hours GPS-on mode. Plus, it needs to do all the other things people expect from a Suunto wearable (which, is merely software/platform features that largely have no impact on battery life).

    • Eli

      20% less power under load. If the load is low it will be able to sleep more meaning it should have a much longer battery life

  10. JR

    Cutting off access to competitors is risky from an antitrust perspective, even in the the U.S. (see the recent draft vertical merger guidelines). The risk is compounded because of Garmin’s dominance. A wearable company that is struggling to survive might easily conclude that its most valuable remaining asset is a treble damages claim against the industry behemoth. I can say that when I was doing defense side antitrust work, I would have strongly advised Garmin to continue licensing on fair, reasonable, and non-discriminatory terms. Garmin doesn’t need to cut off its competitors in order to win; it’s already winning.

    • inSyt

      You make very good points. Looks like the best play for Garmin would be to continue licensing to third parties but to keep adjusting licensing fees upwards. This will gain Garmin a price advantage, forcing third parties to use their own inferior algorithms or to give up a larger share of their profits to Firstbeat/Garmin.

    • Claude

      I agree, and also I think the intent of this not to cut off other competitors from using what they’ve used before, but rather giving them all (Garmin first of course) against the real behemoth waiting around the corner to acquire Firstbeat-like functionality: Apple.
      Imagine the apple watch came with firstbeat technology, the whole market for Garmin, Suunto, Polar etc wearables would shrink overnight to mere niche products, and disappear!
      This is a lot more about protection of potential market share than it is about cutting off existing competition

    • Andrew

      I agree with JR (I’m an EU anti-trust guy) but the “continuing to push license fees upwards” will also raise concerns with predatory pricing or vertical foreclosure type concerns – and again, not very FRAND.

    • Eugene

      The whole market would not shrink overnight, there are still too many huge issues with Apple and their ecosystem that many won’t buy Apple devices.

    • Eugene

      This type of acquisition happens so much now a days we can predict exactly how the future will be.
      1. New features and improvements will come to Garmin devices first and then roll out to other licenses later.
      2. License fees will increase for third parties for new and renewals
      3. New features won’t roll out to other devices since that could cut into sales of new devices.

      These are typical for any industry.

    • Ken

      ” the whole market for Garmin, Suunto, Polar etc wearables would shrink overnight to mere niche products, and disappear!”

      Hardly.
      Apple watch requires iphone which eliminates any Android users & the battery life is awful. A mid to back of the pack marathoner couldn’t get thru their race with one, let alone a triathlete, ultra athlete, or even most people doing a metric century charity bike ride. Garmin’s battery life runs circles around Apple’s watch & doesn’t need to be charged every night.

    • Pavel Vishnyakov

      Well, to be fair, battery life on Apple Watch is not THAT awful. You can easily get 7-8 hours of data recording without any effort and more if you sacrifice some of its functions.

    • Jr

      I agree, Eugene, but I think there’s a real question whether the new stuff will matter all that much. There’s only so much that we can usefully do with heart rate variability data. Doubtless there will continue to be new First Beat metrics developed, but will they drive sales outside of the Fenix series (which seems to be bought mostly byprice insensitive feature geeks)?

    • Eli

      I don’t see how FRAND would matter. None of the FB stuff is part of any open standards

    • JR

      Eli, FRAND isn’t a concept limited to open standards. It’s also relevant to antitrust/competition law. It can be a required remedy to approve a merger that would otherwise have anticompetitive effects. Even outside of the merger context, a refusal to license on FRAND terms can essentially constitute “tying,” which is illegal under certain conditions.

    • Christian Köhler

      How could that affect Polar?
      They have their own algorithms, they do not use FirstBeat at all.

  11. Charlie

    Right now all advanced analytics is done on-device, because that’s what the licence is tied to. This deal allows Garmin to use the same algorithms more freely, e.g. in Garmin Connect, and show Training Load there. To optimize revenue, Garmin could try to de-link the algorithms from the watches themselves and offer them as a premium subsription service (similar to a Strava Subscription).

    • usr

      Subscription sounds nice from the perspective of less perfectly good hardware getting unnecessarily replaced to gain access to a software update, but the addressable market for a subscription is much smaller. Everybody who uses fitness algorithms buys hardware, but only a small subset of those would be willing to pay for a subscription. Most people are far more willing to hand over their money when they get some physical token in return.

  12. Andrew

    Hi

    I think there are things to say about this for two reasons:

    1. Part of me thinks this really doesn’t matter. I have recently found FirstBeat data to be valueless, in particular both its race prediction and lactate threshold data which appear to ignore data that proves them demonstrably bollocks. Two examples – 8:50am FirstBeat predicts 5k race time of 19:20 (not a lot of spadework for an IM triathlete in December) – 9am run a 17:46 5k at park run. FirstBeat predicts 5k race time of 19:20. Second and similarly, I recent ran a solo lockdown half in 1:24:20 (bang on 4min/k pace) and FirstBeat data on my 945 still has my lactate threshold at 4:11. Given that the FirstBeat data is a bit crap and not really something you can put too much stock in its hard to argue this is even good for Garmin consumers, its probably just a bit meh.

    2. On the acquisition itself, I think this is interesting – I don’t think Garmin can come out and say that they’ll be turning off FirstBeat for competitors (either current or future features) as I think this would simply see those competitors stirring up trouble with anti-trust authorities about consumer detriment in the next generation of devices if Garmin refuses to supply much of this functionality to competitors therefore as you say limiting consumer welfare. I think if the Suunto’s of this world wanted to kick up a fuss this could be interesting (and I’m sure they’d love to see what Garmin’s internal documents say about such a potential strategy).

    • Aaron

      Are the algorithms public, open source and can go through peer evaluation? If yes, there should be something to explain the bollocks you see. But that doesn’t make much sense since copyright doesn’t cover algorithms, yes? And folks could simply copy them instead of licensing. So they must be trade secret sauce; industry “magic”, which is fine I suppose since these are consumer level metrics.

    • Tim Grose

      I found those metrics can take a while to “learn” you and also they are heavily dependent on having a realistic max HR set and I bet many users just leave it at the default 220-Age. When I initially got my 945 I saw things like this that but now they look pretty good. In fact I did a marathon in Feb just before lockdown and my time and the prediciton were only a minute different and I only got down to around the time I ran when I started my taper and my fitness “improved”. Can they be better – of course and especially if you run a 17:46 and it still says you could only do 19:20.

  13. jww

    Fascinating. Great timely analysis here.

    May quibble slightly with the point on future R&D. Garmin’s resources are effectively unlimited. The strongest non Apple balance sheet in consumer tech. So one could also make the case Firstbeat the subsidiary is now free to fund any moonshot pet project on the whiteboard.

    On the overall strategy, I can picture a future with Garmin continuing to license the basics like VO2, training load, recovery time, to competitors, while reserving the new and innovative features for its highest end devices.

  14. TheStansMonster

    Only planned structured workouts sync from TrainingPeaks to Garmin Connect – are you thinking that Connect should be able to leverage that for a planned Training Load in Connect/on the device?

  15. Are there any other companies out there specifically similar to Firstbeat? In other words, not just another app or virtual coach, but a company whose primary mission is to produce training metrics that are intended to be licensed? A Secondbeat, if you will. Valencell comes to mind. Anyone else?

  16. pakibhai

    I’d be happier if the crappy Garmin implementation of FirstBeat derived metrics went from device specific to account specific. Right now they can’t even keep training load and/or training advice in sync between say a Fenix 5 plus and an edge 530 owned by the same person. In GarminConnect under Training Status there is a pulldown to see training load/advice by device and they conflict with each other. And yes I know about Physio TrueUp, look at the forums they are full of complaints that it doesn’t do what it promises to.

    • Dan G

      Yes I do hope they unify the inexplicably different per-device Training Status displays in Connect Mobile. It should just be one chart.

  17. Just an updated clarification here from Garmin…

    While they did acquire Firstbeat Analytics, as part of this whole scene, Firstbeat broke off the Corporate Wellness & Pro Sports divisions into a company called ‘Firstbeat Technologies’. Garmin only acquired the wearables licensing bit.

  18. Poul

    When I read the press release from Firstbeat Technologies it is clear that Garmin “only” acquires the Analytics part of the company. The corporate and athlete focused parts of Firstbeat continues to be an independent company.

    It is unclear how mich of the r&d will follow Analytics and how much will remain with Firstbeat.

    So the changes to the Firstbeat website only reflects the changed focus of the (remaining) company

    • They split the website into two, but have removed sections related to the modules licensed to 3rd party competitors. Or, they at least made them hard to find easily.

      As for the press release wording, I’d argue it’s a bit ambiguous. Most M&A announcements would have made it a bit more clear stating specifically which portion was not acquired. While this sorta says that in the opening, it does so only if you knew the new names of the companies they invented today to then break off today. Otherwise it reads as though those divisions are simply going to keep on operating (as opposed to being shut down, which is common in M&A). Just my two cents…

  19. John Poisson

    Value in Machine Learning (AI) algos are in the data. More data, better algos so Garmin (First Beat) would benefit from expansive data across multiple devices to fine tune the predictive capability of their algos. The real value creation opportunity is to create a truly connected physiological ecosystem across multiple devices to include blood sugar (Dexcom) or HRV (Oura,…). Like you mention above I have found Garmin’s workload woefully inadequate (I track via WKO5 and compare to what Strava predicts). Without full connectivity across multiple devices / user interfaces this capability will continue to be less than optimal.
    A platform that could connect this ultimately creates more value for all users and leads to even more accurate algos. When it comes to competition, with this step, Garmin appears to be setting itself up to compete more with Apple than just fitness device companies.

  20. Andy Coughman

    Ugh. I have such a love/hate relationship with Garmin. Like many of you, I’ve got loads of Garmin stuff, from the Running Dynamics pod up to Satellite (InReach) Emergency devices and several watches and radar lights. I’m definitely a ‘Garmin’ guy.

    But Garmin is so… cheap (product differentiation ‘making’ us buy more expensive devices for marginal gains [features]). So slow to move. Always “so close” to nailing something and getting it perfect, but not quite. Now, I’m worried there won’t be any further development of the Firstbeat stuff.

    Sure, Garmin can say the Firstbeat employees keep their jobs in Finland and we all live happily ever after. They can SAY they’ll still license Firstbeat to other brands… But do we REALLY think they’re going to do that?

    Garmin is a very solid company, and I think “we” only look at their fitness/wearables side. I don’t have experience with their marine side, but lots of experience with their aviation products, and they are SOLID.

    On the wearables side, Garmin is “doing great” but there are too many pain points for customers. But never quite enough to push many people to Polar or something else.

    Sleep tracking. My wife’s old FitBit is now not used since she got a Vivoactive 4, but it had better sleep tracking than my FR 945. It’s possible to get better sleep tracking with FAR more insight by using your phone and the right app (SleepAsAndroid, etc.) than it is with any Garmin watch.

    I paid a lot of money for my FR945, but it does not accurately know when I fall asleep! We can speculate about Garmin competing with Apple in the health industry (like tracking blood sugar, etc., as John Poisson mentions) but Garmin needs to step up their game big time if they want to compete like hat.

    Like “John Poisson” says, the value is in the data. Google bought FitBit, right? And that’s like 27 million users? Apple obviously has a huge user base. If Garmin wants to step up to that level, they better be ready to play like the major league players.

    And when we still wonder if our $500+ FR and Fenix watches will get “better sleep tracking,” well, that just doesn’t sound like a “major league” player problem.

    But in the end, I, along with many of you guys and gals, will look to Garmin for our next watch, our next bike head unit, and pay that premium because it’s “Garmin”

    Like I said: love/hate relationship.

    /rant

    • Neil Jones

      Yeah, my thoughts too. My worry is that FB will be allowed to stagnate, wither and die. When FB was independent, it was a feature that Garmin needed to include to remain competitive with other manufacturers who also used FB. However, assuming Garmin cuts off that access, unless competitors manage to maintain the pressure with alternative algorithms as credible as FBs, then there’s no incentive for Garmin to keep FB going and my worry is as well as no further dedicated analytics development they’ll just revert to their own in-house analytics (I mean independent from the FB acquisition) as “sufficient to tick the box” which, to be polite, don’t have anywhere near the depth or (subjective) credibility of the FB ones from what I see.

  21. Les Rice

    thank you

  22. Marklemcd

    “ And it’s been clear lately that wearable companies are more and more heavily leaning on Firstbeat to do the heavy lifting of getting these metrics correct.”

    They should be outsourcing to someone else given how “well” firstbeat stuff works.

  23. Nathan M.

    So I guess this is great news for those who have older Garmin devices who want some of the first beat features the newer products have. Right? Right? I mean it’s a nice thought. I’m not sure what excuse would be left for Garmin at this point. Except money, profits, and and more money.

    • John Kissane

      I’ve one of those older devices (645) and much as I’d like to see it get some love from Garmin would be surprised if it ever got another update of any sort. The 655 would probably be with us now if it hadn’t been for the global pandemic.

    • Dan G

      I think the 645 has been “replaced” by the 245. That there won’t be a successor to the 645 is my guess.

      So the range is really 45/245/945 now, and 55/255/955 etc in future.

  24. Leo

    Firstbeat functions are only interesting for people who want more data, without questioning the quality of that data.

    I’ve seen way to many cases where the firstbeat data made no sense at all. Recoverytime and training status all seems to be focussed on short interval stuff and cannot handle long runs.vo2max increases to unbelievable values with short intervals, and return to more normal values with long runs (although you end up with an unproductive status)

    Too bad the majority of the buyers just want more and more data.
    If you want a device with great battery live (which is an actual useful function) you have to buy a device with all de nonsense features and is therefore very, very expensive.

    just my 2 cents.

    • JR

      Leo, I agree that all of this stuff should be taken with a grain of salt, ESPECIALLY if you don’t use a chest strap. Let’s be honest: almost nobody uses a chest strap.

      I suppose FB could be very useful to someone who is totally new to training, but with any experience I think most people progress to a stage where they know better than the watch what they need to do, how they’re recovering, etc.

    • Neil Jones

      That’s an interesting assumption about the chest strap – I’d always assumed the opposite (for the DCR target audience or higher-end Garmin users), though my own assumption wasn’t based on anything. It would be interesting to know how chest strap/oHR numbers compared.

    • GLT

      I have no idea how many enthusiasts & athletes use a chest HRM versus wrist or forearm OHR, but I do know I will personally use chest strap solutions during exercise until they are no longer available. Pre-carpal tunnel & chest straps have never bothered me so my preference is probably opposite of many people.

      Training Effect has been valuable for me for years now, but it has just been the last year or so that I’ve felt continuous HR measurements had a reliable value for me. Body Battery is now a stat that I consider worth review even if it is imperfect. Blood oxygenation & respiration rate feel generally useful, but that may just be the current pandemic making that data seem more important than it actually is.

      It would be great to have all of the OHR & running dynamics data measured constantly from a rechargeable chest strap, but I doubt that will be happening.

    • Leo

      @JR

      I use a cheststrap. OHR doesn’t work for me. It’s more likely to pick up my cadence than my heart rate. Especially when it’s cold.
      My current forerunner 935 does this, my previous tomtom did it. I guess my veins are hard to detect.

      But I don’t mind running with a chest strap. And for 24/7 rest heart rate OHR works fine (although Garmins algorithm is a but strange. I see lower heart rates than the reported resting heart rate. But then again, Garmin is notorious bad at writing software)

      FB very useful for someone who is totally new? I strongly disagree. Someone who is totally new has no reference to judge the errors in FB advices. And since FB has e strong prejudice for intensive interval training, it is now wise for a newcomer to blindly follow FB data.

  25. Tim Grose

    Am a little unclear here by “Update heads up – Garmin has clarified they only bought the licensing piece…”

    So what have they not bought and how does that work in practice? How can a company make any money or indeed maintain any independence if some 3rd party (now Garmin) decides how to sell what they produce and presume takes that cash. Indeed who determines what their work is.

    • They bought the consumer wearables algorithm aspects, but not the pro team portion of the company or the corporate wellness divisions.

      The part of this that’s likely going to be semi-challenging for Garmin’s competitors is that they’ll ultimately have to disclose/tell Garmin unit sales volumes, because that’s how these are licensed. Note that some of that already exists today with ANT+, but that’s in theory a bit more firewalled off.

  26. Lev Bronshtein

    Interesting and insightful. As machine learning is becoming more accessible and easier to implement I suspect there may be a few others. I recently came across atlas wearables while doing research into excersize detection and reporting. So far they do not even have a product it seems beyound a paid beta, but they are marketing both the watch and the algorithms. link to atlaswearables.com

    • Wow, it’s been a long time since I heard the name Atlas Wearables. I figured they’d gone under. Used to always see them at CES each year.

    • Brandon

      Oh yeah! I thought they too had gone under. I nearly pulled the trigger on one of their original wearables. Nice to see they embraced a more traditional form factor going forward.

      Their atlas 2 wristband had a face
      Only a mother could love lol. I wanted to pull the trigger on buying one a couple of times but never did.

  27. Marek

    Would be great if Garmin bought also some company to increase precision of their footpod (Stryd). Or how about getting Concept2? There is a lot of physio data to be extracted from those ergometers. They are very precise at measuring the power output. Firstbeat even used the rower to calibrate some of their algos.
    Best,
    Marek

    • Tim Grose

      I always thought that Garmin thinks foot pods are “old tech” especially since a large part of using them is that they can work better than GPS for pace/distance when probably 99% of runners (including me TBH) think what they get from GPS is good enough. Mind nearly all Garmins now have OHR and yet I still use a chest strap…

  28. Martin

    Sorry, I can’t follow why less competition is good for the customers of the market leader.
    It’s good for the company itself (if it works out), but customers and users will always benefit from an active competition!

    • GLT

      In a normal & productive business climate competition is certainly useful. In the situation the world has entered into in recent months, it is probably okay for everything that is adding value to be absorbed into a business that will weather the storm. It can all be spun off willingly or through mandate when the centralization-decentralization cycle flip flops again.

      Better Garmin than another major player IMHO.

    • Leo

      Better Garmin than someone else? I don’t see that. Garmin in notoriously bad at writing software. Really bad. And they don’t care. Really don’t.

      So if they use their own quality of software development on firstbeat.. well, it would become just random data generators.

      And is firstbeat in a position they would not survive the crisis as you imply? No way! Every major watch manufacturer, except apple, licences firstbeat. People buy sport/fitness watches like crazy.

      It just that some rich people just wanna be more rich.

    • GLT

      Will certainly agree that things would be better if Garmin chose to be the world leader in hardware/firmware/software quality. I think consumers have a better chance of seeing Firstbeat tech in multiple products with Garmin than they would if Alphabet, Amazon, or Apple made the transaction.

      Since this isn’t a classic hostile take over, the Firstbeat leadership would have looked at their own forecasts for their future transactions and found this change to be their best choice. There may be non-financial and special circumstances behind the scene that amplified their decision, but they certainly aren’t accepting the transaction with expectation of becoming a weaker company.

      In the US all for-profit companies have the imperative to increase profits continuously & endlessly, so long as they stay within the law. Whether this qualifies as greed depends on things like hiring practices & social responsibility efforts. It would be good for all of the highly profitable companies mentioned to have more visible social investment.

  29. Dan G

    Let me preface this by saying I really like the Firstbeat metrics. They’ve just never been explained properly (you have to spend time reading on the Firstbeat website and time pondering to grasp them).

    Training Load guides you to what is sustainable for you personally.
    Recovery Advisor let’s you know when you’ll be fully recovered (ready for hard intervals).
    Training Effect is nice, but you need both Aerobic and Anaerobic for them to be worthwhile.
    Load Focus/Training Load Balance uses TE to measure what you’re doing — and where you’re lacking.

    It’s only with the Suggested Workouts on the new Edge 1030 Plus that I think the true potential is being realised: the device now tells you how to train to your full, personal potential, i.e. what to do each session to get the best Load Focus and Training Effect.

    If Garmin pulled this together and put it on every device, they’d have a killer app.

    (I only use OHR and, following Ray’s advice on how to wear a wearable, get much better data than I ever did with a chest strap. I’ve also never had the problems others seem to have with using a watch for running and a bike computer for cycling and using Physio TrueUp to sync them.)

    If we lose this because of this buy-out, it will be a real shame. I hope whoever sold up asked for enough.

    • Manos

      Hello,

      Where can I find Ray’s advice on how to wear a wearable? I’m having some issues with thr OHR sensor of the fenix 6x pro.

      Thanks!

    • GLT

      There are bits & pieces scattered across many of Ray’s articles and subsequent comment threads. His detailed reviews often have a section specifically on OHR performance so consider re-reading that section of the F6X review if you haven’t.

      Also, use the site search function to find his “Thoughts on the wearables studies” article for some of his insights into the topic.

      In general the sensor should not be pointed at the wrist bone. The device should be snug enough to avoid light leaking in. If the device is bouncing or sliding around during exercise it is not likely to get an accurate reading. Some skin tones & types are problematic for some sensors.

    • Manos

      Hello again,

      Thanks a lot for the advice I did read it once again (obviously I’ve read it and watched the video before buying the watch 😃) and found the articles you’re talking about regarding the Microsoft and Mio trackers and the study, a very interesting read.

      I’m experimenting with the beta software for the watch and today during my run I wore the watch inside my wrist, with the sensor facing my veins and I believe that I got better results, when I was feeling out of breath I hit 160-170 when I reviewed the HR chart.

      The next goal would be to increase the anaerobic load hehe.

      Thanks again,