--- title: Decathlon - Case study url: https://blog.octoperf.com/decathlon---case-study/ date: '2020-06-09' authors: - gbetaillouloux categories: - Case-Study tags: - JMeter - REST - API - Automation - User Experience description: Learn more about how Decathlon tested its new RFID platform with OctoPerf. --- # Decathlon - Case study [Decathlon](https://www.decathlon.com/) **is a network of innovative retail chains and brands providing enjoyment for all sports people.** At Decathlon, 85,000 of co-workers live a common Purpose on a daily basis: `To sustainably make the pleasure and benefits of sport accessible to the many`. Decathlon presently operates in 58 countries in Retail with more than 1654 sport hypermarkets, and in 26 countries with Production. At Decathlon, **innovation** is at the heart of all activities: from research to retail, including conception, design, production and logistics. Signed sport teams channel all their energy into developing technical, good-looking and simple products, always at the lowest possible prices. These products are aimed at all sports enthusiasts, from beginners to experts, and are sold at Decathlon.
Nicolas Zangari
Nicolas Zangari, the digital platform manager at Decathlon was actively looking for better, simpler load testing solutions than legacy tools.
Nicolas had previously heard about OctoPerf during a french testing event called JFTL.
He was looking out for a tool for his project since the RFID API was being redesigned and would soon require testing.

## RFID API The RFID API collects data from RFID Antennas and is being used to track each article every step of the way: - Production - Storage - Transportation to the stores Once the product is in the store it can then be used for: - [Stock management](https://en.wikipedia.org/wiki/Stock_management) - [Electronic article surveillance](https://en.wikipedia.org/wiki/Electronic_article_surveillance) - [Cash registers](https://en.wikipedia.org/wiki/Cash_register) ![Electronic article surveillance](https://blog.octoperf.com/img/blog/decathlon-case-study/electronic-article-surveillance.jpg "Electronic article surveillance") This allows a very easy and efficient tracking of each item but it also means that **this application is critical to Decathlon**. ## Getting started The initial situation was as follows: - The JAVA developpers had little to no experience with load and performance testing, - The updated API was about to be deployed worldwide. **The challenge is to find a tool with a short learning curve while still being complete enough to conduct proper tests** Having heard of OctoPerf from past experiences, Nicolas was eager to give it a try. He quickly found out that OctoPerf was very easy to use and would be a great fit.
The free account allowed us to get started right away with the test scenarios preparation and bottleneck analysis.
Nicolas Zangari
## Using OctoPerf ### Test around the globe As the API was about to go worldwide, one of the key features is **to be able to test from a lot of locations**. And OctoPerf makes it very easy to [split the load accross various cloud locations](https://api.octoperf.com/doc/runtime/edit-scenario/edit-user-profile/location/). Using [OctoPerf SaaS platform](https://api.octoperf.com/ui/) load agents are started on the fly based on the test you configured and you don't need to focus on the test infrastructure at all. ### Model complex loads Another important topic is the ability to **emulate different load profiles** coming to the API. - Large one-off calls through stock management systems, - Small and regular calls through cash registers. This was tackeld by creating different [virtual users](https://api.octoperf.com/doc/design/edit-virtual-user/) and [user profiles](https://api.octoperf.com/doc/runtime/edit-scenario/) in OctoPerf: ![Cloud locations](https://blog.octoperf.com/img/blog/decathlon-case-study/cloud-locations.png "Cloud locations") Nicolas also appreciated the possibility to **create your own custom load policy** for each of these User Profiles and ramp up the load step by step: ![Userload](https://blog.octoperf.com/img/blog/decathlon-case-study/userload.png "Userload") This allowed the tests to slowly increase the load up to several thousands of requests per second. ### Support & UI Another key topic when you get started with a new tool is the **reactivity of the product support team**. Here as well Nicolas was satisfied with the experience since we spent some time helping him and his team get started. Later on they were able to reach us quickly over the live chat if anything happened. Another point that Nicolas appreciated is OctoPerf's UI. It was a very nice thing to have since it really made communication with other teams easier. **Easy to read visualisations are critical when you want to highlight bottlenecks** and quickly fix them: ![Analysis](https://blog.octoperf.com/img/blog/decathlon-case-study/analysis.png "Analysis") ### Flexible licencing model Working on scaling the performance also require a [flexible licence policy](https://octoperf.com/pricing/). Otherwise you can quickly get stuck because you didn't plan for a large enough licence. Our reactivity combined to the possibility to define/update your license quickly brings a lot of agility.
The flexible, on demand pricing policy was very convenient in our context where each team can pick their own toolsets.
OctoPerf's no-commitment approach was really appreciated.
- Nicolas Zangari
## Conclusion In the end the combination of the short learning curve, the intuitive UI and the fast support allowed Nicolas and his team to quickly get confident with load and performance testing. **Thanks to their hard work and OctoPerf, they were able to meet their performance goals**.
We will be using OctoPerf on all our new projects in the future.
- Nicolas Zangari
As a final note, **I'd like to extend my thanks to Nicolas for his nice feedback**, from all the team at OctoPerf. I know we will be in touch soon for his next tests and **I hope to see you as well on OctoPerf soon!**