Accuracy is a top priority among the users of wearables, and a recent Valencell survey shows that a lack of insights from the data collected is a common reason that people stop using their devices, and cost is still a barrier that prevents many from buying one in the first place.

The lack of accuracy is a major concern among consumers, and active signal characterization needs to be incorporated in order to improve accuracy for some data points, said Steven LeBoeuf, Ph.D., Valencell’s co-founder and president. Valencell develops performance biometric data sensor technology.

The survey was conducted because, “we were interested in what wearables people had. The reasons for dropping them, and what types they had. This is important for us to design wearables people don’t drop, and what will make people change their minds on buying a wearable device. One of the things that definitely turned up is that people are very serious about their wearables being accurate,” LeBoeuf said.

SEE: Wearables have a dirty little secret: 50% of users lose interest

Nearly half of the respondents own, or have owned, a wearable device, and of those who do not own a device, 58% said they’d only consider buying it if they trusted the accuracy, and 30% have not purchased one due to the cost. The online survey of 706 U.S. consumers ages 18-65 was conducted in collaboration with MEMS & Sensors Industry Group in June.

Interestingly, only 27% of those who use, or have used, wearables say that they like the accuracy of the data. And this was across a variety of devices. Of those surveyed who own or have owned a wearable, the chosen form factor for 52% was a wristband, 36% earbuds and 32% a smart watch.

According to the survey, 35% of wearable owners feel step counting is the most useful function, 18% find heart rate monitoring most useful and 12% prefer the notifications.

As far as conditions they’d like to monitor beyond what they are currently measuring, here is what respondents said:

  • 55% would like to monitor stress
  • 48% would like to monitor hydration
  • 46% would like to monitor blood pressure
  • 38% would like to monitor sunlight/UV exposure
  • 35% would like to monitor key vitamin and supplement levels

“I was really surprised at things people wanted to measure on top of heart rate. It wasn’t clear to me the types of things that were metrics that needed to be focused on. Basically it gave us insight into what kind of metrics people wanted to see in their new devices,” LeBoeuf said.

Of those surveyed, 37% who have owned a wearable have stopped using it. The reasons include it being too much of a hassle to continually recharge (40%), not accurate enough (29%), and uncomfortable to wear (26%).

Another surprise was that only 48% of respondents want a wearable for constant day and night use, while 32% want one they can wear all day, and 20% want a device they wear only while training or exercising. “I thought maybe 70% would want day and night. But it broke out into a more interesting dispersion,” he said.

“I’m guessing based on the survey is that some people don’t want to sleep with anything. Comfort was a big issue. Some people don’t want to sleep with something on. A day-worn wearable is all they really care about. They sleep well already and they’re not worried about tracking their sleep. Some people all they care about is monitoring their exercise. They don’t really care about a daily life tracker to guide them and their health per se,” LeBoeuf said.

“It bodes well for companies interested in integrating technology in jewelry. People who wear jewelry products don’t like to wear them while they’re sleeping but they do like to wear them all day long,” he said.

Steve Caldwell, co-founder and CEO of Strap, which does wearables and mHealth analytics, said that not having enough insights from the data is a key reason people drop the use of wearables. “I think step one is surrounding the data with more insightful experiences whether that’s in true health care or in consumer gamification. To get there there has to be a layer that enables folks to connect their data in a streamlined way. The development layer for software developers to integrate this data quickly is key.”

Randy Bergstedt, senior marketing and business executive for smart wearables at Epson said that he found the study to be right on the mark in terms of hardware and how people demand accuracy. He said the only thing left out that he considers valuable is the motivational aspect of what the data means to a consumer or user of the hardware, such as the online community around a device.

“It’s different for every user. Some users just love the data and want to look at the data. For others it’s about the relationship and the community and the context that they’re in,” Bergstedt said.

How to improve accuracy

To improve accuracy, active signal characterization is key. Most wearables use optical heart rate monitoring, but an issue with this method of data collection is the presence of motion artifacts and how not all motion and environmental noise are created equally. Active signal characterization would help with accuracy because it’s similar to active noise cancellation in that it actively identifies and characterizes different types of raw signal data from the biometric sensors found on many wearable devices, LeBoeuf said.

He explained, “The active characterization of the signal data is imperative because there are many different types of motion noise, which must be processed differently in order to properly filter the optical blood flow signal. The motion information collected by the sensors can also be used to facilitate biometric assessments based on both heart rate and motion information.”

The 3 big takeaways for TechRepublic readers

  1. Only 27% of current or past wearable users like the accuracy of their device.
  2. Only 48% of users want a wearable for day and night use.
  3. Active signal characterization can improve accuracy of wearables. Wearables vendors should start including it as a feature.

See also