Weekendesk shares a sneak peek how AB Tasty’s Roll Outs solution allow them achieve faster releases with less risk
Travel website Weekendesk has a simple mission: help hard-working weekend warriors easily find the best deals for their trips. They were also one of the first early adopters of our feature management platform for product teams. Their tech and product teams got a sneak peek of the tool, and got to play around before the official release. Here’s what they thought:
To start off – what got you interested in AB Tasty?
Generally speaking, what are the advantages of using a feature management and product experimentation platform?
Julien Martin, VP of Engineering: A/B testing is essential for any business. But in our case, I was very particular about the kind of tools I wanted to use. With our new front-end application, I decided to block any ‘classic’ (client-side) A/B testing solution – I didn’t want a layer of JavaScript/CSS tag on there, to ensure maximum performance.
We thought about this quite a bit, and after trying our own (pretty limited) in-house system, AB Tasty told us about their Feature Experimentation and Roll Outs functionality. It was the solution to all of our issues – no added JS/CSS layer, and no development outside our repository, which will let us maintain the same high level of quality and performance in any situation.
Alexandre Duarte and Alessandro Pintaudi, Co-Heads of Product: The platform lets us be more autonomous and also mitigates risk thanks to progressive rollouts. We can test new features on specific segments of our user base, and also better integrate user feedback in real-time, set up kill switches, and especially, run experiments.
What were your first impressions of the platform?
Maxime Biloé, Lead Front-End Developer, and Sonia Alliche, Front-End Developer: We think the interface was really clear, everything is very logical and easy to understand, even with the variety of options available in the platform. We were pleasantly surprised at how easy and quick it was to set up a test, from the interface itself to the integration with our application.
Alexandre and Alessandro: The dashboard is easy to use, especially since there are pre-defined use cases – this is a real time-saver. The interface is also very intuitive and easy to get started with the first time around. This is true in terms of configuring the tests, retrieving data and analyzing the reports. Finally, the support we received from AB Tasty allowed us to deploy the tool very quickly, and to launch our first tests in only a few days.
What challenges were you able to tackle thanks to our platform?
Julien: We started with a really simple test, just to get used to the platform. From a technical perspective, the idea was to put the performance of the tool to the test. Just one extra https request to create a better user experience and better processes – this is just perfect.
Alexandre and Alessandro: We want to be quicker and more efficient in our releases, without increasing any risks associated with putting new features live. The platform lets us do this thanks to the different functionalities available in the platform (progressive rollout, A/B testing, feature flagging…), as well as advanced experimentation to optimize our product.
After playing around with the tool, the major takeaway for us is that we need to systematize and speed up how we deploy A/B tests, so that we can better understand the issues our user base faces, and find solutions – all with less risk.
What are the next steps for Weekendesk and AB Tasty?
Julien: We’re going to integrate a new payment service provider (PSP) in our application. The platform lets us deploy a progressive rollout, so we can avoid any messy roll-backs and confirm our theories. We’re organized into squads, who will be working on continuous optimization on a variety of subjects, and therefore running a lot of tests at once.
Alexandre and Alessandro: We’re going to run a series of A/B test campaigns at the same time during the last stages of the purchase funnel. One of our goals here is to simplify the decision to buy something, but also to uncover the factors that lead up to it.
