Home furniture shop

Tell us a little about yourself and the project

Hi, my name is Artem and I’m a CTO of roooms.ru !

I’m not a professional developer - my main expertise is in data warehouse development and product analytics. roooms.ru is a small family business. My wife has lots of experience in everything related to interior design. She grew from a junior designer to deputy CEO of one of the biggest Russian interiors companies. At one point she decided to start her own business, and that is the start of the roooms.ru story.

The first version of the shop was made by freelance developers 3 years ago. The start was rather successful, there was plenty of traffic. After first release this website hadn’t any significant improvement, so each month traffic was slightly decreasing.

At one point my wife decided that she wanted to relaunch and grow her business. And I agreed to help her, and in collaboration with a former frontend-developer colleague, roooms.ru became for me my pet project.

Tell us about the project. What were the challenges you faced?

At first, we estimated a target quantity of SKUs that we can sell - it is about 1 million different item types with various attributes. So the platform should be able to deal with rather big catalog. This catalog should have different taxonomies - by product type, by brand, by room of usage etc.

Next, our shop should be able to extended with some specific features for designers - e.g. wishlists for different projects/roooms or 3D modelling. And this extension should be rather easy.

Moreover, we decided that using modern technologies, approaches (like headless architecture) and open-source will be great advantage in choosing a platform.

And one of the most important factor is the price. Our business is rather small, and we don’t want to pay much money if we don’t yet have a large amount of orders coming in.

What factors made you decide to use Vendure?

Firstly I researched many OSS systems: ReactionCommerce (now Mailchimp Open Commerce), Saleor, Spree, and some other small projects. I also evaluated more established solutions such as WooCommerce, Magento, Bitrix, Shopify, BigCommerce and many others.

Every solution had some disadvantages. Some have old tech stack, some are not nice to work with, some solutions have commission on transactions and can become very expensive to run.

After 4 weeks of testing different approaches (none of the systems were particularly developers-friendly) I decided that we should use ReactionCommerce.

We started development. We implemented the catalog listing and founded a lot of bugs. We started researching solutions to these bugs in the community and realized that some of them had existed for a couple of years already. I was ready to change solution to Saleor or Spree but eventually our frontend developer found Vendure.

At first we were sceptical - the project was in beta and was very young. But after deployment in 5 minutes and a couple of days of evaluation I realized that all of our requirements were covered. Moreover, I didn’t find any bugs in the basic workflows.

So we decided to give it a try!

Can you tell us about the process of building the project?

We built the storefront with Next.js, which our frontend developer was already familiar with. It took for us about 5 months to create the storefront and fill a small part of our catalog (we work 15-20 hours each week). Not bad for pet project!

The process of building was rather easy - Vendure has great docs and if something can’t be found there, there is a Slack channel where the Vendure maintainers are very responsive with help.

While implementing the frontend, we found 2-3 bugs. Despite being relatively new to Node.js development, I was able to fix them by myself. I was very proud.