One of the biggest brands in athletic wear is charging into wearable tech with a pack of products and a pair of tech industry partnerships. Under Armour is leveraging artificial intelligence to give its offering an edge over competing products from Nike and Fitbit. 

Retailing for $400, HealthBox is a trio consisting of a Fitbit-like band, a digital scale and a heart-rate monitor. And UA isn’t starting from zero in its effort to tap demand for wearables.

Over the last few years, the company has snagged three massive online fitness communities: Endomondo, MapMyFitness and MyFitnessPal. They now control the largest online wellness-focused ecosystem, at 165 million users. And it’s what HealthBox can do with all that data that makes this a compelling package.

Under Armour partnered with HTC to develop the hardware and a smartphone app called UA Record, which ties all the products together. The UA fitness tracker is cleanly designed, made of a rubber-like material with a LED display. The heart rate monitor is constructed of durable band and the monitor glows when it detects a heartbeat. The scale measures weight and BMI. All the data from the three devices talk to each other and transfer data via Bluetooth to UA Record.

Cognitive Coaching

Under Armour’s partnership with IBM and the “cognitive coaching” potential of Watson differentiates HealthBox from the competition. By feeding nutrition, training, and sleep information into Watson, it’s “able to understand data in large volumes, make recommendations, and continuously learn,” says Chris Glodé, Under Armour’s VP digital, connected fitness. “The more data UA Record inputs, the smarter Watson becomes.”

UA has experimented with Internet of Things in the past. They built a sensor-laden compression T-shirt back in 2011 for the NFL Combine, where college stars worked out for prospective pro teams. The shirt provided raw data on acceleration, speed and heart rate for scouts to pour over. HealthBox and the partnership with IBM means that the Record app will be able to send the data to Watson to disambiguate.

Watson’s Cognitive Coaching uses a comparative model, grouping users based on criteria like age, gender and activity level in order to provide training and recovery recommendations.

And the experience will get richer over time.

“As you record more data and as more data is recorded across the community, the smarter the insights will become,”  Glodé says.