The Evolution of the Umso Editor

Around 8 years ago I decided to start my own company. Having previously worked as a freelance web designer and consultant I decided that the website builder market could use some fresh ideas. I gave myself one hundred days to publicly release an initial version, which I built from "scratch". The next day, Umso (then called Landen) had its first customer.

The first Umso Editor: V1

The editor I built back then is now known as the v1 editor. As you can imagine, the thing I built in 100 days wasn't perfect. As Umso gained traction and found its first customers, I kept chipping away at the editor to make improvements and add new features. Since the product now had some paying customers I had to be more careful about breaking or taking away things. With every new feature or fix, the codebase grew increasingly more complex until a few years in, it was near impossible to work with. 

At this point, Umso had grown to a team of 4 and we had ambitious plans for the product, but the old tech was holding us back. So we decided to build an entirely new editor with everything we've learned over the years. On the outside it would be mostly the same but the internal data structures, code and architecture were now optimized for what the product needed and would need in the future. This new version would become known as the v2 editor.

The Modern Rebuild: V2

To effectively build the v2 editor we had to let go of the convoluted data structures that had grown in the v1 editor. That meant a clean break. Sites created in the v1 editor would not be editable in the v2 editor because their underlying data was completely different. This was the right decision. However, we made a critical mistake right after: we didn't immediately build a migration tool.

My incorrect assumption at the time was that if there was demand for a migration tool then customers would let us know and we'd build it. Because migrating the site data would be a very complex task and we wanted to focus our efforts on improving the product, we did not prioritize such a tool. 

As it turned out, existing customers were not even aware of the new editor because they'd only get to use it when creating a new site. If you had an existing site, nothing changed. This was intentional since we didn't want to disrupt anyone's existing workflow. However, it led to many customers not realizing for years that there was a new option.

In short, existing customers were unaware of the new editor and therefore never asked for a migration tool and we didn't prioritize it ourselves because we wanted to work on more exciting things. This led to us not building such a tool until this year. I want to apologize to our loyal customers and users who've felt let down by how we've handled the entire situation, including our communication around it. 

A new Editing Approach: V3

While the underlying technology of the v2 editor was solid, we now ran into some issues with UX approaches, mostly around text formatting. To solve this issue we've experimented with editing the preview directly instead of managing all text from the editor sidebar. 

This approach felt very good for text and led us to the v3 editor which entirely bases its editing UI on the preview and no longer has a dedicated sidebar for content editing. Because we didn't want to disrupt the workflows of existing customers we've decided to launch this editor as a separate version as well.

This time too we repeated some of the mistakes we've made after building the v2 editor and didn't immediately provide a migration tool or clearly communicate the changes and our plans. Ultimately we should've focused on incremental improvements to the v2 editor instead of making another breaking change. 

Reflecting on the Changes

I've wanted to take the time to acknowledge the mistakes we've made but ultimately not dwell on them. It matters to us that you can trust our product and our team. I hope this background and transparency will help with some of the confusion and frustration that may have resulted from our earlier mistakes.

Earlier this year we've built a migration tool for the v1 editor, improved the migration tool for the v2 editor and added several tools to support the migration process, such as the ability to copy analytics data and a tool for transferring domains between sites. We've added labels and documentation to explain the differences between editors and how to migrate. We're offering free concierge migration services for paying customers.

In addition to providing a positive experience for users of the older editor, we're also changing our approach for editor updates in the future.

Will there be a V4?

In short: no. We will not release any more updates that require you to "switch" editors. Instead we are making incremental improvements to the V3 editor. Updates won't be stopping. On the contrary: we can now focus our energy on more productive incremental updates instead of big jumps. 

We've already released a number of great improvements to the v3 editor and got a lot of updates planned. As the web continuously evolves, so do we. To stay up to date with our feature releases you can subscribe to our changelog here.