Six months have passed since my last article and it’s time to tell you what we have been working on. I finished my last article with this intriguing sentence: “And what about people building their website without CMS? Well, we have something coming for you guys, more on that later.”

We have worked hard this semester to extend Weglot capabilities to not only WordPress and Shopify websites but to everyone: Weglot is now able to make any website multilingual. Any CMS, any framework, any technology. We have a universal version.

Why the universal version?

Until now, we have been very active developing Weglot in 2 CMS, WordPress and Shopify. Our growth has been (and still is) carried by this 2 CMS. We have a product people love, great feedback and new ideas to keep improving it.

Moreover, when we tried to expand last year, it fails. Remember, we launched an integration with PHP frameworks like Symfony and Laravel but that didn’t work and we quickly abandoned it to refocus on WordPress and Shopify.

So why taking risk to build a universal version ? For three reasons.

The first reason is philosophical. When Augustin and I founded Weglot, the mission was clear: make all websites multilingual. The fact that we ended up having most of our clients on WordPress or Shopify was never our end goal, but a way of starting and getting initial traction. Since the beginning, Weglot has always had the same purpose: Create the most simple yet powerful software to make any website multilingual.

The second reason is to improve our product due to synergies. Each CMS and communities being different, it’s hard to make one unique tool for everyone. But making this effort forces us to make a simple and complete product. For instance, at the beginning on WordPress, we didn’t translate emails. When we started developing for Shopify, we had to think about translating emails as it’s an important feature for merchants. Once we developed it, we were able to add it in WordPress. Having a large audience is the best way to build an inherently simple product.

Last but not least, it prevents having a too risky dependency on a CMS. Imaging, if we have 90% of our customers from WordPress, what happens if WordPress launches a multilingual feature natively? Probably hard times.
Having customers split into independent CMS or technologies is the best way to prevent this risk.

Building a universal version

Building a version that would work with all websites hasn’t been easy.

Historically, on the first month of Weglot in 2015 Weglot was just a Javascript snippet to copy/paste in your HTML page. It worked with all websites. So in a way, we had a universal version in the beginning!

Unfortunately, we quickly understood that it was not working with SEO as we didn’t translate server side. To make a clean server-side integration, we abandoned the Javascript and build specific integrations: A WordPress plugin and a Shopify app. It worked great but required a unique development each time.

At this point, we figured that to be able to serve all websites, we would launch a new integration for each technology. We would build a Drupal module, a Laravel package, a Symfony bundle, etc … So in April 2018, we tried: we released several new products to make Weglot work with Laravel, Symfony and Drupal. But it didn’t work as it did for WordPress and Shopify. It was too developer oriented, lacked documentation and was in conflict with the native way these frameworks worked. We decided to abandon these integrations 3 months after we released them.

For a few months, we didn’t do anything about other CMS and refocus on WordPress and Shopify. But we kept thinking of new ideas to integrate will websites. With the help of Mehdi, a freelancer who has been working at Weglot for a while, we designed a new technology, based on subdomains and reverse proxy. Instead of integrating a plugin or an app, the user will just have to enter a DNS record to activate subdomains and Weglot will do the rest. It’s server side (so optimized for SEO) and easy to integrate. As we were happy with the way it looks, we start testing it manually with customers on the last trimester of 2018 and had great results.

In January, we wanted to test at a larger scale but without publicly releasing it. We launched connect.weglot.com as a sub-product and advertise on other CMS like Squarespace or Wix and other technologies. The feedback was really great, the product was ready!

So finally, in April 2019, we integrated this sub-product into Weglot. Now, when signing up you can choose the technology upon which you built your website and use Weglot accordingly. It doesn’t change anything for WordPress or Shopify. After all, why change something that is working already. But it opens Weglot to a lot more users.

WordPress & Shopify

This semester we haven’t only work on Weglot’s universal version. Even if the WordPress plugin and Shopify app have reached some maturity now, we kept improving it. On WordPress for instance, we improved button placement and AJAX compatibility. We also added a private/public mode that is more flexible. You can now easily control what language you publish and which one are just private for work-in-progress.

Global speed and code quality have also been improved even more.

We have also been most active than ever in the communities. We have announced a partnership with Caldera Forms on WordPress and have been sponsoring many events. WordCamps in Bordeaux, London, Paris, Caen and soon Berlin.

The team kept growing also and we are now 15 people. We have recruited 2 developers, Nicolas and Mehdi. An office manager, Meryl, and an intern to reinforce our customer support team, Arthur.

This semester I’ve learned several lessons that I’m sharing here :

• Keep your ADN: Even if you move away from your mission temporarily, always keep your company purpose in mind.

• Do experiments, do not fear to fail, even after the beginning of the company.

• Separate experiments from your main product and do it without marketing. This way you have nothing to lose if the experiment fails.

We have a lot ahead now and our new objective, in terms of MRR (Monthly Recurring Revenue) is to reach 220,000€ in the next 6 months.