Cycling sportswear brand Rapha says its MACH marathon is proving a winner

The move to a MACH (microservices based, API-first, cloud-native SaaS and headless) infrastructure is providing leading sports clothing brand, Rapha, a range of benefits. These include a 20% improvement to Largest Contentful Paint, a metric that indicates the speed at which content is loaded on a website, and an overall 40% boost in page load times.

MACH has also helped the cycling brand achieve an 80% improvement in user search success, thanks to better image-load and rendering metrics.

Plans are also being drawn up to do what would have been much harder before - adding in functionality from supplier New Store to create more unified, omni-channel customer outreach.

John Kilpatrick, Rapha’s Principal Engineer, also says his team’s new MACH-based architecture has slashed the time that his creative and merchandizing colleagues need to spend working with product and customer images and videos.

Even better, his new best of breed back environment (which is hosted on Amazon Web Services) can automatically deliver images in their most optimized destination format.

It is also allowing use of more lightweight digital asset file formats like WebP instead of JPEG, to optimize site and mobile user experience even further.

He says:

We saw an opportunity to speed up the ingest process to allow our merchandizing and brand teams to deliver value much faster. Speed is incredibly important in our business, especially for merchandizing, which must work ahead to pull together the right content. Some processes that used to take up to 24 hours are now effectively instantaneous.

Choosing the best

As detailed in diginomica, MACH is all about promoting a best-of-breed approach to ecommerce and content management.

The industry body promoting this, the MACH Alliance, claims that users that have moved to MACH from monolithic, first-generation ecommerce stacks report an increased ability to respond to market changes and an ability to build and implement new functionality more quickly, at reduced cost.

So far, Kilpatrick broadly agrees. He says:

For us, MACH’s mostly about choosing who we consider to be the best in class. All the things the monolith used to do, we'll one by one take off the monolith and switch them to the best in class, so that over time we will eventually get to the best in class experience we can find.

Kilpatrick says that embracing MACH was driven by how the previous system, though adequate, was chosen by non-technology specialists.

He says:

I transitioned from freelance to full time here four years ago strictly on the condition that we replaced the monolithic stuff we were having to work with: I literally said, ‘I’ll become permanent for you if we get rid of this, because it's awful.’

‘Awful’ in this case means that the firm’s older integrated platform did a lot of things, but in his words “not very well” - and even worse, it was still a new system when he onboarded.

He adds:

I think the users of the old website were happy enough, but from a product and a technology perspective, we weren't happy because it was too hard to release onto, so things got outdated very quickly. 

The development cycle became all about keeping running; there were just too many pain points, and we couldn't deliver the experiences we needed.

Guided by Kilpatrick’s pressure, changing this up meant a move away from an incumbent SAP ecommerce solution, as well as a range of other software including Amplience for image handling, Algolia for search and Auth0 for authentication and authorization.

Instead, Rapha has a new MACH stack that includes the Commerce Layer commerce platform at the back end, Contentful’s headless CMS, DAM (digital asset management) and image manipulation software from Cloudinary.

For now, partly for pragmatic and contractual reasons, this new composable architecture is running in parallel with Kilpatrick’s legacy stack.

However, Rapha is planning to move more and more business processes to MACH components in the next few months.

He says:

Right now, there are areas where it's new and old working together. We're still at the start of our MACH journey, which makes it even more interesting.

Early MACH benefits

With a mission to ‘inspire the world to live life by bike, because it's never just a ride,’ Rapha (as it’s usually known) is a UK-headquartered supplier of high-end cycling apparel—though Kilpatrick (himself a keen biker) prefers the term ‘functional clothing’ to fashion.

Founded in the UK in 2004, Rapha’s named after a famous 1950s French cycling team whose main sponsor was French apéritif brand Saint-Raphaël. It is owned by the grandsons of Walmart founder Sam Walton, who purchased a controlling stake in the brand via a 2017 $260m private equity buyout.

Rapha sells a range of garments for cyclists, from jerseys to tights, as well as accessories like shades and water bottles.

Its two main sales channels to do this are online - which is dominant - and a global network of what it calls “Clubhouses” - local community centers for cycling fans which also act as physical stores at 21 “cycling cities” around the world, including Austin, Melbourne, and Berlin.

The importance of online but also community engagement means that both great brand imagery, but also support for UGC (User Generated Content), is seen as critical for the brand.

Here, the good news is that image handling is getting better - with Kilpatrick citing more agile and slicker management of assets, and his creative team being able to deliver images 90% faster and videos 50% faster than under pre-MACH workflows.

Summing up his MACH experience, for Kilpatrick, there’s now a third choice between going for one e-commerce platform or building it all yourself in-house: clipping together what you want, when you want it, via standards and Application Programming Interfaces (APIs).

Two years into doing that, he concludes:

So far, MACH at Rapha has been mostly about efficiencies across the board, not just for customers, but a lot of teams inside the business. And we’re very happy about that.

But we also have the runway to be more experimental if we want to be, while always guaranteeing a great user experience.