How Sport 2000 brings the MACH idea to life

The cooperative group wanted to reposition itself online from scratch and, within the shortest possible time, build a comprehensive online business system that could measure up to the best-practice benchmarks in the industry, offer customers real added value and reflect the expertise of the physical retailers across various platforms. At the same time, the group wanted to give specialist retailers an easy way to take part in the platform without much effort.

Given the integration of the physical retailers with their multitude of products, more or less advice-intensive, from various sports, the requirements were complex and would hardly have been possible with common standard shop systems, or only with elaborate custom development. That is why the cooperative group opted for a MACH stack that combines various best-of-breed solutions with one another.

In view of the project requirements, the choice fell on an approach based on MACH principles: Microservices-based, API-first, Cloud-native SaaS and Headless. This makes it possible to use best-of-breed approaches for every task in the online shop's process chain. On top of that, the tech stack is flexible and can be expanded and extended quickly, dependence on individual software vendors is low, maintenance costs are reduced and the quality of the tools used is at its highest.

SPORT 2000 e-commerce manager Tim Wiese was quickly won over by the idea. "I know how sluggish the monoliths are", he says. "On the other hand, I also knew that connecting various modules to one another leads to many interfaces and therefore possibly to problems as well." But the advantage of being able to act quickly and flexibly in future and to bring new ideas to the road fast ultimately won out.

One of the big challenges in the project was preparing the data on hundreds of thousands of products, provided by various manufacturers in the most diverse structures and qualities, so that it could be used in the shop. And already at this point the advantages of being able to combine various best-of-breed approaches in the cloud become clear. The first station the data passes through is the P2C platform Productsup. There the data is automatically sorted and cleaned using various rule values, so that in the end everything is available in the same structure and quality. The article data then flows into the cloud-based PIM Akeneo. There it is likewise automatically cleaned, categorized, merged, enriched with images and refined using various rules. From a certain level of refinement, the data goes automatically and API-based via AWS into the underlying shop software commercetools. "Through the entire process we can ensure that in commercetools we really have clean article data of the highest quality standards", says Shopmacher CTO Manuel Ludvigsen-Diekmann. "And everything runs fully automated, purely by rule sets."

The link between the article data and the prices and stock levels from the ERP system of the SPORT 2000 retailers is solved via a cloud-based tool developed by Shopmacher itself, which acts as the central interface. As soon as the sellable products are available in commercetools, they are automatically forwarded to the cloud-based search engine Algolia. "That is a very powerful tool that offers users very fast and very smart search options in the frontend", explains Ludvigsen-Diekmann. "Retailers can decide, for example, which attributes are searchable or how filters should be weighted. On top of that, Algolia can also learn and suggest products automatically."

All the data that arrives in Algolia can automatically be used in Frontastic too, which SPORT 2000 uses as a Frontend-as-a-Service solution to visualize the shop. To be able to generate dynamic content in addition to the static content orchestrated via Frontastic, Shopmacher also built in the headless CMS Hygraph, through which the SPORT 2000 team maintains the entire blog area, for example. Texts needed for the frontend flow via interfaces into Frontastic and into Algolia as well. This makes that content visible to the shop search too.

When users register in the shop or order products, the data takes the cloud route in reverse. It is taken into commercetools and then automatically distributed to downstream backend systems such as CRM or ERP via the Shopmacher data distribution tool. The payment service provider Adyen and the email marketing also work cloud-based. "Basically, with six to eight different tools we run a pretty big process loop", says Ludvigsen-Diekmann. "And each tool is not a jack of all trades, but has a single task that it handles particularly well, and so brings its benefits to bear. That is the best-of-breed idea, focused on one goal: to really radiate quality outward."

The order loop back:when a user orders, the data flow runs in reverse:

• commercetoolstakes in the user and order data.

• Data distribution tool (custom)distributes automatically to backend systems.

• CRM, ERP, Adyen, email marketing:all cloud-based.

The hope of being able to bring speed to the road in shop development with a best-of-breed approach has been fulfilled. In a project time of just six months, a stable and scalable framework for the new online presence was built. In the coming months, the task is to use this framework to make the cooperative group's USP visible on the web. The worry about interface problems, which gave project lead Tim Wiese a bit of a headache at the start, turned out to be largely unfounded. Instead, the foundation was laid to meet online competition at least at eye level. Because hardly anyone on the market currently has a more modern tech stack.