Gymshark turns to Google Cloud for infrastructure revamp and generative AI trials

The CEO of direct-to-consumer fitness apparel startup Gymshark has turned to Google Cloud to help standardise the “mish-mash” of IT underpinning its operations, and is drawing on the hyperscaler’s artificial intelligence (AI) expertise to explore how generative AI (GenAI) could benefit its business.

The e-commerce company, founded in 2012, has recently branched out into bricks and mortar retail with the opening of a flagship Gymshark store on London’s Regent Street to provide its customers with a mix of in-store experiences, personalised products and community-based events.

During the first day of the Google Cloud Next London conference, the public cloud giant’s managing director for the UK and Ireland, Helen Kelisky, talked about how the tech giant’s cloud and AI offerings were being enthusiastically adopted by the startup community and digital natives such as Gymshark.

The company’s founder and CEO Ben Francis joined Kelisky onstage at the conference where he detailed the work Gymshark is doing with Google Cloud to tidy up its IT infrastructure and make better use of its business data to help enhance the experience for its online and in-store customers.

“First and foremost, we need to move onto the Google Cloud Platforn,” he said. “We’ve been a fast-growing brand, and we are still very much a startup, so much of our infrastructure has been mish-mashed and sort of sticky-taped together.”

And doing so will lay the infrastructure foundations for Gymshark to improve the quality of the data it accrues from its customers during the course of their dealings with the brand, and – in turn – provide them with a more personalised user experience.

“We’ve got this really cool thing where every single customer that’s ever ordered from Gymshark, [as] an e-commerce business, we have that information and we have that data, but it’s not sophisticated or detailed enough. And we need to really get our arms around that and improve the quality of our data,” said Francis.

As an example, he talked through the work the company is doing to combine the transactional data it has from people who purchase goods from the company with the data it collects through its fitness app.

“We know that a lot of customers will use that app three or five days a week [during workouts] and some of them will be into weightlifting, some into CrossFit, [but it gives us] different, more contextual data that we can overlay onto the transactional data that we have for our customers.”

From here, the company can set about making product recommendations, for example, to its customers based on their past purchases and the type of exercise they like to do.

Gymshark is also at what Francis calls an “exploratory phase” with its use of GenAI, with the company using the keynote to showcase a potential use case whereby its fitness tracking app could use a person’s workout history to make recommendations about what type of exercise they should do the next day.

“We have to take our time to explore, so with generative AI we haven’t launched anything, but we are in a purely exploratory phase,” he said.

“With all [new tech] things, what we try to do is be as exploratory as we possibly can,” he said. “When Gymshark first started, there wasn’t many businesses that were purely e-commerce and there weren’t many people that were heavily invested in social media and influencer marketing…but those things are really, really pivotal and important in our business.”