Sweating it out. The technical journey of the Gx Sweat Patch
Technology rarely acts alone these days.
Your car talks to your smartphone, your banking app asks distant servers for information and verification, and whatever movies you’re watching or music you’re listening to, it’s just the last link in a global chain of storage, compression, and streaming technology. Offloading technically-demanding tasks to distributed systems is so common, and works so well, that it’s easy to forget that until recently, most devices had to do everything themselves.
So anytime a complex, expensive piece of hardware gets replaced by something light and cheap, that leverages the technology around it, it’s worth paying attention. The March 2021 release of Gatorade’s Gx Sweat Patch is a good example: a disposable patch that essentially replicates a professional sports testing lab.
In many ways, Gatorade is one of the world’s first sports technology companies. Back in 1965, years before Nike was experimenting with waffle irons to make a better running shoe, football coaches at the University of Florida were mixing glucose and water with salts and other electrolytes to keep their players from dehydrating in the subtropical sun. Sports drinks are now a multibillion dollar industry, but for Gatorade (now part of PepsiCo), off-the-shelf beverages are just one of several ways they support athlete performance.
Working with Smart Design in 2014, Gatorade’s Sport Science Institute (GSSI) developed a personalized hydration platform for players on the Brazilian national soccer team. Standout innovations included smart bottles that track hydration rates, and “smart pods” that allow players to customize what they’re drinking based on sweat analysis. It was that last element—sweat testing—that became the focus of a more recent design effort, to give everyday athletes the kinds of metrics and personalization that thousands of pros already rely on.
When a pro-level innovation becomes a candidate for large-scale consumer roll-out, new considerations jump to the forefront. Cost is an obvious one: spending that’s justified for world-class athletes is often a non-starter the rest of us. So are accessibility and ease of use. Professional sports teams have doctors and trainers on staff who can perform sweat tests and make sure weight, training time, and hydration are meticulously tracked. Without specialists like that, a technology has to work flawlessly every time with little to no expert attention. It’s not an unusual problem—many ubiquitous technologies, from email and GPS to movie editing on your smartphone, started as professional tools that got democratized—but the bridge between pro and consumer is no easy thing to build.
It’s not the device. It’s what’s around it.
In order to give someone personalized recommendations on what and how much to drink for optimum performance, you need to gather information about their body, what it’s doing and the environment it’s in, then combine the resulting data through an algorithm developed with Gatorade’s sports scientists. The obvious solution is to create a device that does all this, but the result will almost certainly be expensive, heavy, and hard to use.
Instead, GSSI started by asking what could be done off the body. Crunching the data was an obvious one: there’s no reason the algorithm needs to run locally, as long as it gets the information it needs. An athlete’s personal information can reside elsewhere too, in a profile that includes things like age, gender, weight. They can quickly log the day’s activity on a smartphone. The day’s weather—which matters a lot when figuring out hydration—can be gathered automatically.
So what’s left? You still need to monitor the athlete’s sweat. The chemical information contained in each droplet is crucial to understanding how a human body is faring under exertion, and no statistic or third-party data can take its place. So this became GSSI’s primary focus: how to analyze an athlete’s sweat, inexpensively and unobtrusively, and pass the resulting data to the algorithm so it could work its magic.
In practical terms, this meant something that could take sweat samples, ideally with no effort from the athlete, then talk to the cloud about what it found. The lighter the better; the simpler the better. A wearable device was one candidate, but what about something thin and flexible, like a patch or Band-Aid? A company called Epicore Biosystems was doing fascinating things with “skin-like wearable microfluidic solutions that are capable of non-invasively measuring sweat biomarkers, skin health, and physiology, in-real time.” GSSI partnered with them on the project.
They also needed a partner to turn the concept into something people would actually use. When GSSI approached Smart Design, they knew that the solution was going to be a patch, and they knew what it needed to do. John Anderson, the Director of Technology at Smart, quickly saw that the real design challenge lay in building up the ecosystem of technology that the patch would connect with. But there’s a lot of infrastructure to backfill when you’re doing something new—things that established technologies can take for granted. There was no database of sweat patch readings to train the algorithm on, for example, and no best practices for applying one or taking readings from it.
Case Study
Gx Sweat Patch
This is the pilot speaking
In many ways, Epicore’s microfluidic patches offered a perfect solution: thin, light, mass-producible, and able to detect a wide variety of compounds. It wasn’t feasible to embed processors or transmitters in them though, so the team had to look at other ways of displaying and communicating data.
Working with Smart and GSSI, Epicore came up with an initial prototype: a flexible, disposable patch that adheres to the forearm and absorbs sweat through tiny holes in the back. The resulting reactions produce data, which is conveyed visually, rather than digitally. This shows up as two visible collection channels, one that turns shades of orange to show sweat rate, and another using shades of purple to show electrolyte composition.
So far, so good—but integrating the patch into a consumer-friendly product and ecosystem, without sacrificing accuracy, would be a lot more complicated. So they did what any team of designers with a prototype on their hands would, they started running pilots with target users.
“Pilot tests with people keep us grounded,” says Anderson. “They never go the way you want, and you have to stay focused, but you always come out of them with insights that are invaluable.” Would the patch give a reliable reading in real world situations? How user-friendly could the process be? You can sit around for days debating the answers, or you can start trying things out.
The first pilot for the Gatorade Gx system, held at a high school for elite student athletes in Florida, focused on answering functionality questions and building a database of sweat readings. The patch’s color-coded channels could be read with a simple photograph and optical recognition software—in theory. But whether it would be accurate enough, under a wide range of lighting conditions, on a variety of skin tones, in a variety of use cases, was far from certain.
So the pilot was designed to take as many readings as possible, and photograph each patch not once, but twice: first with an iOS application that had a sole purpose of image capture and data collection, then again using a DSLR camera. The resulting database of hi-res photos, smartphone photos, algorithm outputs, and direct sweat testing created the foundation that the rest of the project would stand on.
Running a successful pilot also means leveraging whatever assets you already have. Smart had previously developed an app for professional athletes, able to take the data from sweat analysis in a lab and come back with hydration and fueling recommendations that were specific down to the milliliter. The pros loved it, but an app for consumers had to contend with different inputs, different use cases, and different expectations. This became the focus for the next pilot, conducted at a gym in Frisco, TX called Cowboy‘s Fit.
Fifty two daily participants, ranging from casual gym goers to serious triathletes, put a prototype app and patch through their paces and gave feedback. The design team could quickly act on this, updating the app and putting it back in the hands of the same users. Some elements from the pro-facing app turned out to work just fine: serious athletes often use training and tracking apps already, so adding Gx to the mix was relatively smooth.
The level of detail and explanation had to change, though. Those milliliter-specific recommendations that professional teams love are overkill for most amateurs, so the new Gx app did things a little differently, giving enough detail to be useful, but not so much it became inaccessible.
This technique of focusing different pilots on different aspects of the experience—first data collection, then user journeys, and finally complexity of taking a sweat test—is critical to their success. “If you try to learn too much, you end up with noise—an inch deep and a mile wide,” says Anderson. “So it’s crucial to make sure you know what you’re trying to learn, and run the pilot with that in mind.”
Armed with this knowledge, the team was ready to run a third pilot, releasing Gx into the wild. Gatorade gave away 300 updated sweat patches, along with instructions for submitting photos and profile information, to real users.
This was the first time the system was used unsupervised, and it worked beautifully. Applying the patch and getting the photo it turned out to be no issue at all—people thought it looked great, and snapping a photo made a nice mid- or post-workout ritual. The pilot also showed that the best readings came 30-45 minutes into an activity, which informed another round of refinement for the app and the patch.
The results are in
After showcasing the Gx patch and system at TED 2019, Gatorade started piloting it with professional sports teams. It was also chosen by SXSW as a finalist in their Wearable Tech category, although COVID kept the Smart team from presenting it in Austin. By March 2021, the Gatorade Gx Sweat Patch was on shelves at Dick’s Sporting Goods.
The final design works because, despite doing something extremely innovative, the Gx patch itself is fairly low-tech. But you can think of it as the tip of a tech iceberg: a simple, highly visible element that leverages a vast ecosystem of servers, algorithms, connectivity and careful design to do something extraordinary.
This is increasingly the way new technology comes into the world, not with a single super-advanced gadget, but with dozens of devices, systems, and data sets interacting in meticulously designed ways, to produce something magical. It’s a ridiculous amount of work behind the scenes, often by partners working in diverse fields (like sports medicine, interaction design, and microfluidics, for example).
But when you think about the kinds of experiences that create a sense of “magic”, they often have that iceberg-like quality: a simple device or interaction with a sophisticated benefit. That’s only possible through tech ecosystems, and those ecosystems get more complex everyday. After all, if a team of designers, technologists, and researchers can give a skin patch magical powers…what’s next?
About John Anderson, Technology Director
John’s background is in software development, hardware engineering, data science and product design. Recently, he has been the project director on the award-winning Gatorade Gx Sweat Patch that mixes software, IoT, data science and a sprinkle of biomedical engineering. Additionally, John oversees a product and engineering team that’s building an IoT platform using machine learning, sensing and computer vision. John is also a technical advisor at NYU Stern and speaks regularly on how designers and engineers can create experiences for people with accessibility needs.
About Jared Billig, Technical Product Manager