Crystal Ball on Your Wrist? Why Wearables Are a Bigger Deal Than You Think

“Hey, it’s been a while.Want to grab a coffee?”

I text my friend to catch up. Grabbing a coffee and sitting down at a cafe has become the default activity to have a conversation with people, even if one doesn’t enjoy coffee. This time, however my friend responds,

“How about a walk? I gotta hit my steps for the day”

Before the age of wearables, this comment would’ve been nonsense.

I got my first wearable in 2012, as part of a corporate wellness initiative. It was a simple device that only captured steps and offered no connection to any other device. Admittedly, when I first got the device I was motivated to hit my daily step goal of 10 000 steps. Early on, I was excited to take the stairs or to go for a run and then relish in the huge boost in step count. The dopamine hit of reaching 10 000 steps felt great.

But almost as fast as they came in fashion, wearables seemed to fizzle out. Big players dropped off. Nike abandoned their wearable, Nike FuelBand, after 2 years. Jawbone, who pivoted from Bluetooth headsets to health oriented wearables, shut down. I eventually got bored of step tracking and stopped using wearables altogether.

Even the Apple Watch, the dominant wearable with over 35% market share, arguably had a slow start since its introduction in 2015.

NIKE__FUELBAND_BLACK_ICE_hd_1600

The Nike FuelBand launched in 2012. It was discontinued in 2014. Photo: Nike

It’s 2020 now and I think wearables have finally come of age. We’re at an inflection point. Wearables finally have the ability to live up to the “quantified self” concept and can massively influence personal well being and healthcare as a whole.

I’ll cover what’s different today, what that means, and why it’s such a big deal.

Wearables finally have the ability to live up to the “quantified self” concept and massively influence personal well being and healthcare as a whole.

The Promise of Wearables

The promise of wearables has always been to improve one’s health. Wearables, in theory, would track some useful data that would drive some meaningful behavior change that leads to better health.

For example, a wearable might track my distance walked per day and present the information in a progress chart. I could then review the chart and it would perhaps motivate me to go for an extra walk. Over time, this would lead to behavior change from scrolling twitter at my desk to taking daily walks at lunch.

We thus get a virtuous cycle of: Collect data, analyze and present data, and trigger behavior change. The new behavior is tracked, analyzed, and triggers further behavior change.

Paper.Sketches - wearables

Wearables promise to improve our wellness through data

The theory makes sense, but wearables have failed to live up to this promise and haven’t driven change across a general population.

I believe we’re at a point where data collection and data analysis are improving to a level where new and much more effective behavior change can be enacted at scale.

Data Collection – Better and More

A wearable gathers data. But for data to be useful, users have to be able to collect data and also motivated to do so. Users have to not only collect enough data, but to also provide the data for analysis.

Collecting rich data is easy

Tracking data such as heart rate has long been available before the modern wearable existed. But dedicated heart rate monitors were clunky and expensive, and typically only used by niche segments who were highly motivated to track their heart rate, such as elite athletes.

Wearables of today have made data collection incredibly accessible. The magic of the modern wearable is twofold:

  1. Wearables now can collect an incredible range and accuracy of data, including motion, heart rate, location, direction, sound. And that’s not including the data collected via a smartphone, which a wearable usually pairs with.
  2. Wearables can collect this data in a manner that is natural and unobtrusive.

The most popular wearables are currently worn on the wrist and feel just like a watch or bracelet. The similarity mean wearables are not only easy to wear, but also socially acceptable (perhaps even fashionable..?). Battery life has been made long and charging made simple. What results is a device with extremely low user friction to gather large amounts of data over time.

Perhaps most significantly, the price of wearables have dropped to a level accessible to a general audience. A Fitbit tracker starts at around $50, and in 2019 Apple introduced the lowest pricing ever for the Apple Watch, at $199. These price points move the accessibility of wearables from niche audiences to mainstream users.

Users are motivated to collect and provide data

Just because a wearable is capable and unobtrusive doesn’t mean users will use the device, and over the extended period of time necessary for sufficient amounts of data.

How do we get people to track their data consistently and provide it for analysis?

There must be some value or incentive for the user to collect and provide the data. The app ecosystem is a fantastic way to individually motivate users to collect and share data. With a range of apps, users can find apps that fit their specific needs best and provide the value their looking for that motivates to collect and provide data.

For example, I’ve been interested lately in fitness tracking. Even within fitness apps, there are endless options tailored to specific use cases and audiences. I’m currently using Zones, which gives me the ability to track my heart rate zones in real time as I workout. For me, this has led to not only more effective workouts but also to more workouts.

IMG_3053

Zones, an app on Apple Watch, gives me real time feedback into my heart rate zone.

For others, step or stand tracking and nudges, or a community aspect, can motivate users to collect and provide data continuously. Incentives could even go further and apps can provide rewards or discounts for activities which collect and provide data.

Data Analysis – Next Level Information

Data is nothing without analysis. A massive pool of data for users is meaningless unless useful information is provided from the data.

Looking Backwards

Wearables have been great at providing a review of the data collected and insights based on the data. The main characteristic of the analysis to date has been backwards looking – it enables users to understand their prior behavior. An example would be when Fitbit sends me the “Penguin March” award badge, for walking the equivalent distance of the March of the Penguins (70 miles).

Beyond cute badges, there’s also more value backwards looking analysis can provide. It’s possible to understand if treatment plans or new habits are providing desired outcomes. A wearable will help me understand if my sleep quality has increased if I cut back on junk phone time before bed, or cut out coffee in the evening.

Screenshot 2020-02-23 at 6.08.27 PM

Badges provide addictive dopamine hit. Fitbit also encourages me to share my results.

Looking Forwards

Where things get really interesting are the forwards looking analyses that are on the horizon. Machine learning, where algorithmic results are not based on a set of rules, but based on patterns learned from a large sample of training data, will enable a new level of prediction previously not possible.

What does it mean when I say machine learning enables prediction? For example, machine learning is what enables Alexa to respond with “Toronto,” when asked for the largest city in Canada. Alexa was not programmed to respond with the answer “Toronto.” Instead, Alexa learned over time to predict that when provided the input “what’s the largest city in Canada,” the response people sought is “Toronto.”

In the same way as Alexa predicted a response to “Toronto,” machine learning could predict health results given a set of health input.

For machine learning to work, massive amounts of data are needed to train and improve the algorithm. Wearables have made data, the costly yet imperative component for machine learning to work, cheap.

Machine Learning and Wearables Data

With wearables, machine learning algorithms now have an amount and range of input data never before available: granular data on activity, location, sleep, heart rate, and more. Additional data such as age, height, weight, blood pressure, can be input or captured fairly easily to complement data that an Apple Watch, for example, could not capture. All of this data is captured at scale as more and more people adopt wearables into their daily lives.

Now imagine this input data is complemented with resulting health data. Health records on conditions, illnesses, or health specialist visits is combined with the ongoing input data to generate predictions. A machine learning algorithm could take a person’s data and use that to predict undetected or future conditions.

Machine learning moves us from data insights to data predictions.

For example, a wearable could predict the onset of a panic attack and nudge the user to take a few deep breaths. Or a wearable could suggest users to check in with their doctor, because it predicts the user could have an undetected heart condition.

The predictions become more powerful because they’re also personalized. Rather than making predictions based on generic population data, machine learning algorithms can compare personalized data alongside massive amounts of data from other people.

All of this leads to more powerful behavior change.

Behavior Change – Be Better

Behavior change is hard. Behavior change is where users have to act and actually change their behavior towards better health.

Sometimes, we just might not have known we needed to act. Wearables can provide the data to inform us when or how we should act. For example, my Apple Watch tells me heart rate zone while I’m exercising. This data has informs me if I need to push myself harder or ease off on the intensity of my workout.

Screenshot 2020-02-23 at 6.27.57 PM

The feedback provided enables me to set the right workout intensity

Other times, and probably more commonly, we know we should act but the follow through becomes difficult. We’ve all been told countless times of the benefits of a healthier lifestyle, whether it’s one that is less sedentary or has less stress.

Personalized means more relevant

Personalized insights and predictions carry significantly more weight than generalized recommendations. I’m much more likely to act on an analysis of my data than rather a generalization of a wider group. How am I doing against myself? (I wonder if this is also the reason personality quizzes are so popular.. seems people love personalized insights about themselves)

Personalized insights enable doctors to make better assessment for a unique situation. Technology will provide the prediction ,and doctors will take the prediction and use with their expertise to make better personalized judgments.

Continuous tracking offers nudge opportunities

With a wide range of data collected, and continuous collection of the data, feedback can be provided more regularly to users.

Wearables have been nudging users since the days of simple step counters, but the power of the wearables now enable new opportunities for nudges.

With apps there are endless opportunities for gamification of the range of data collected. Badges, awards, leader boards, and so on provide an addictive dopamine hit to users that incentivize behavior change. Other incentives more monetary related, such as insurance discounts or prizes, can create win/win situations for users and companies alike.

The benefits from the behavior change that wearables enable are huge.

Societal Benefits

In 2019, Fitbit announced a collaboration with the Singaporean government, integrating wearables into a national health program and providing participants with free wearables if they commit to the program. It’s an interesting signal of growth for wearables from individual wellness to societal wellness.

In 2019, Fitbit announced a collaboration with the Singaporean government. It’s an signal of growth from individual wellness to societal wellness.

Governments have always been keenly motivated towards wellness. A healthier population means less health related expenses and better productivity. Individual wellness is valuable, but even more so is information at an aggregate level. Wearables give governments the ability to understand the population at a deeper level than ever.

Health care programs can be better evaluated for efficacy, and more specific programs can be developed for various population segments. People in cities versus rural areas and of different ages have different habits and lifestyles, and programs can be tailored with personalized information alongside relevant demographic data.

The predictive nature of the data analysis enables people to catch issues before they become larger and costlier conditions. Detecting and changing behaviors that lead to type 2 diabetes is much more cost effective than treating type 2 diabetes.

Wearables have been incorporated into corporate health programs for some time. I believe we’ll see more adoption at a larger scale, just like with Singapore.

Opportunities

Low cost, powerful, and accessible devices, paired with powerful data analysis in an area that could affect billions of people, leads to opportunity and likely interesting new business outcomes. Here are a few areas I think have opportunity for new businesses:

Opportunity for analysis and presentation of data for specific audiences. Wearables and smartphones together can collect a wider range and amount of data than ever before. The app ecosystem enables endless opportunities to track, analyze, and present data that in a manner useful to specific audiences and use cases. As wearable adoption grows, the opportunity for specialization and scale of businesses that provide this service will grow. The predictions provided for the uses cases have value. The audience doesn’t have to be the end user, and could be teachers, coaches, doctors, or other professionals.

Opportunities to lower costs for health care payors. A business that can levereage wearables in a way that provides monetary value to payors (eg. lowers costs), has opportunity. For example, an app that enacts successful change with users, such as a safer driving record or better BMI, that directly lowers costs for an insurance company offers value to the company.

In a world where “sitting is the new smoking” and technology has turned many of us into smartphone zombies, I’m excited about the potential for large scale improvements in wellness.

Technology has eliminated many ailments and threats that were once widespread. Yet behavior based conditions such as depression or obesity are on the rise. Can we reach a place where nobody dies from heart disease?

Can we reach a place where nobody dies from heart disease?

Either way, it won’t just be step targets we strive for anymore. Social and daily behavior as we know it could be completely different. And hopefully we’ll continue towards a healthier world.

And as long as I still have a way to catch up with friends, I look forward to it to what’s coming.

Thoughts or opinions? Let me know! 

Subscribe to keep up with new posts and check out more at www.timothykung.com

2 responses to “Crystal Ball on Your Wrist? Why Wearables Are a Bigger Deal Than You Think

  1. Have been wearing my FitBit for awhile and it was a huge help curing chronicle insomnia I used to suffer from.
    Specifically liked the health applications of variables described in the article and near future predictions about how this tech can change the healthcare system. If interested to look into long term future applications of the ideas described in the article check out Home Deus by Yuval Noah Harari.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s