Becky Taylor • 7 minutes
18 months of hard work recently culminated in the launch of our new content editor.
Our vision is to truly support teams to work together to produce effective content. The editor has been completely rebuilt from the ground up and alongside a sleek new design it now supports live collaboration! I sat down (OK, we’re a remote team, we were on Zoom like always) with our Product Director and Product Designer to find out what went into the the biggest experience update we’ve made to GatherContent since we launched in 2012.
Left: Angus, Product Director, Right: Ben, Product Designer
For an agile team that usually focuses on iterations, how did you reach the decision to completely rebuild the content editor?
Angus: Primarily it came from customer feedback. Over time we noticed a theme of people having lots of challenges when producing content in our product. It was hundreds of points accrued over time. We could (and tried) to fixed them individually, lots of little things, but this theme of feedback wasn’t going away. With being a Content Operations Platform, creating content is one of the most fundamental jobs that we are hired to help people do. So the editor – where people write, collaborate, and approve content is the core of our product, the most important part, it needed to be better.
Ben: It went deeper than user experience and interaction issues. There were technical restrictions that we had with the way that we built it in the first place – technical debt that was making it hard to fix what we had. From the way that we stored content, the text editor that we used, the way the front-end was built. It all felt broken to some degree.
Angus: So we made the pretty massive decision to completely rebuild the content editor. We continued working in an agile way but we had to accept the fact that this was not going to get released in a month. This was going to be a big project. At first we thought it would be three months. Then it went to nine months. Then we just kept adding three-month periods. It was frustrating, especially when you’re getting feedback from the business and customers about the old editor and not being able to show them all the new stuff we knew was coming. But we couldn’t continue with the old editor – we had hit a wall essentially. We did as much improvement to the editor as we could, but we had a technical wall where we couldn’t do what we needed to do, such as enabling live collaboration and really improving performance, without rebuilding it.
What research did you do to scope this project out?
Angus: We started with customer research. Every single bit of feedback that comes in gets put through an analysis process, where we interrogate it, tag it, categorise it, look for themes and extract user stories. We look at the volume of requests but also customer behaviour data using Heap. We combined this with interviewing lots of customers on a regular basis. We conducted at least 50 customer interviews trying to understand what they think of GatherContent and any issues that they have. We asked qualitative questions, like, “What do you use the editor for?” and “How do you feel about it?”. We also ran a survey which had about 350 responses and gave us a good view on what the the editor needed to do. We also worked very closely with the content strategy community and other content experts, to make sure we were building something that would be useful in the long-term.
How about from a design perspective – how do you get started?
Ben: I worked alongside Angus with the customer research, really trying to understand how customers needed to be able to use our content editor. Then of course experimented with other products that had a writing experience that didn’t feel like you were just filling out a form on a page. We tested the writing experience in Dropbox Paper, Medium and even more traditional editors like Microsoft Word and Google Docs. Lots of our customers use those tools to write content day to day, and we definitely wanted to bridge that gap between modern functionality and expected familiarity. Then as early as possible, we got started by paper prototyping. Basically, we looked at our old item editor, what was working and what wasn’t, what kind of information we definitely needed to include. And then broke it all down into components. For instance, the workflow drop-down that you see was drawn on little sheets of paper, and we moved it around and until we landed on the information hierarchy that made the most sense. We brought those decisions into the browser and started testing it with users, getting feedback as quickly as we could before we got too far down the rabbit hole.
Tell me about the process. How did you make sure you felt progress over an extended project period?
Angus: This was a pretty hardcore journey for us as a team and a business, to take on such a huge part of work, and not be able to release it until 18 months after we started on it. It was hard to to manage expectations of the business, and hard to keep telling people we’re behind. We felt after that first three months that we were behind psychologically, even though it was only our own goal. It took awhile to become comfortable with the fact we were laying down the foundation of a better editor in the future, something seriously meaningful for our customers and that truly solved the problem we know we needed to solve – far more effectively than if we had continued to tack on functionality to the previous editor. Testing with users throughout the process validated this and helped shape our decisions and getting that feedback continued to add fuel to our engine.
Ben: Obviously we’re a remote team but we’ve begun having meetings in person more, which I think helped morale. Then in each sprint planning session we’d celebrate the progress from the previous sprint – this is part of our regular process anyway. We also have a Wins channel on Slack so we could share updates with the whole business too which was great to get more people excited for all the little victories throughout the project.
Angus: Retrospectives (another part of our regular process) were also really important. We kept trying to improve our processes by talking to each other about how we’re feeling. I think our whole team had ups and downs at certain points. So being open and transparent, chatting it out in the retrospectives really helps.
Were there any conflicts between how you wanted the editor to look versus what it could do/be? How did you resolve this?
Ben: There was a lot of healthy conversations around where the item editor stood and where we wanted it to stand. We were trying to pull it apart and put it back together in a more logical way. It was something everybody had opinions on. Every detail was considered and debated! Even things like having the labels to the left of fields became a hot topic for about a month! Once we launched the editor internally and the team were using it day to day, I think everyone felt onboard – the quicker you bring people in, the more involved people feel in the decisions and then everyone could take ownership of it.
Angus: There was also a lot of internal opinion about something like this, because it had existed for so long before. So everyone in the company had an opinion about it, and to some degree we all had anxiety about it because it’s such a huge change. It’s easy to manage change in a week. It’s much less scary. But when you’re making such a big commitment it feels so much more significant, and it was. With so many views to interrogate, this is why we had to consistently come back to the data and the research, and what customers are really saying. The volume of testing was critical to give us real data to make decisions from. Eventually we got 250 customers to use it on their real accounts, allowing us to get both qualitative and quantitative data.
How does the editor fit into the whole value offering of GatherContent, has it had any impact on any of the other features?
Angus: Well, we always said that, as a product, we’ve got five major jobs to be done – we help people organise, structure, produce, manage and migrate content. The new editor has hugely improved our ability to help people structure and produce content. I now feel confident in saying we’ve built the best solution we can for this massive problem. We’re the best version of ourselves! The new editor has been available to customers in beta for long enough now that we have enough data to look at people’s interactions and see that they’re collaborating in the way that we hoped that they would. They’re commenting more and content is getting approved faster. We’d like to think that the quality of content that’s getting produced is better because of the way we present guidelines and structure. It seems like customers are thinking about content in a slightly different way now. There’s probably a shift in the industry too, but I hope that the new editor makes it a lot easier to execute on content design and facilitates content operations at scale. The new editor has everything that old editor needed, and in some ways that’s actually a lot less UI.
What are you most proud of with this release?
Angus: We really wanted to be a product with an opinion and clear intent. We didn’t want to be seen as a product that’s pulled in lots of different directions, or tries to be all things to all people. So we wanted this work to have an opinion. That content should be something that’s respected. That it has to exist within a workflow. That it should be structured. And that content can be effective. That’s what the new content editor says to me. I’m also really excited about what else this development will now allow us to do in other areas of the product now that so many technical barriers have been removed.
Ben: The impact that this project has had on the design direction moving forward is huge for me. The decisions that we’ve made within the editor and the design values that we’ve established, are already having such a positive impact on all other areas of design within the company. I really wanted this change to make GatherContent a product that you want to write in, that you don’t mind spending 8 hours of your day producing content in, because many people do use this interface all day every day. I really feel that we got there.
Angus: Now, we can move on to the other three jobs we help customers do. The next of which is organising content. We’ve got lots of exciting features coming up to help organise assets and supporting more rich content. There’ll also be more improvements to the editor, of course, now that we’re back to working in a more fully agile way, and releasing improvements to the product every day.