Trello: Connecting people and discussions

Desktop feature

About Trello

Trello is a collaboration tool that organizes your projects into boards. In one glance, Trello tells you what's being worked on, who's working on what, and where something is in a process.

Our mission

Our brief was to add a chat feature in order to increase user engagement by capturing conversations and resources shared around boards.

Conversations and resources shared around boards are being lost

Our approach

Looking at the competition we could see the IM/chat market was quite crowded and the two main players were Hipchat and Slack.

The key takeaway was that these tools were popular because they were easy to use and offer integration with project management tools like Jira, Basecamp and Trello.

Analysis of the direct and indirect competition for a Trello chat

User research: user surveys

We wanted to know which tools our users were using to communicate and also recruit Trello users to interview. So we set up a survey targeting project managers and people that work in teams.

We got a lot of quantitative data from the survey which we analysed to get a better understanding of what people were currently doing when they had to communicate and exchange files while on a project.

First insight was that people were using Skype as their primary tool of conversation. Second insight was that from our Trello users, part was using Skype and part was using Slack to communicate.

These insights were key for us to formulate our discussion guide and to know which areas to explore on the 1:1 interviews.

User research: 1:1 interviews and contextual enquiry

We had the great opportunity to visit a team at the Ministry of Justice and observe how they use Trello. We then identified that we could split our users in two types: power users and normal users.

Power users would use meta data tagging to collate and synthesis project progress and normal users would be the ones who have a good common of Trello and they use it in conjunction with a physical Kanban board.

Our visit to the Ministry of Justice where we talked to Trello users and watched how they use Trello in their daily work

The key insight was that users were happy using chat tools outside of Trello because they were using those tools (email, Slack, Hipchat) to communicate not only around the Trello board or a particular project but to do all their daily communications.

Although Trello is integrated with Slack, users pointed out that the lack of control of Trello notifications in Slack was really poor and this would generate a lot of spam.

One curious insight we got is that some people on top of using Trello at work, would also use Trello for personal projects. This opened up our minds to a different type of user that we hadn’t considered before.


Affinity Mapping to synthetise our data

Synthesising our user research, we started having a clear idea of our users key behaviours, needs and pain points.

We translated this into three different personas: Tony - the generous planner, Sarah - the open leader and Phil - self starter.

Each of the personas would define a different type of user, Tony was more on the personal side, while Sarah and Phil would use it for work. Tony was an expert and power user while Phil and Sarah were normal users.

Our main personas: Tony, Sarah and Phil

Sarah was our main persona as she epitomised the behaviours of most users we spoken to. Sarah is a team lead in a game design company. She coordinates a lot of different projects and has to speak with a lot of different people: from developers to stakeholders.

She uses different tools to chat with people so her two main pain points are that conversations can get off topic really quickly depending of the tool and that was sometimes hard to retrieve past conversations and files.


Analysing the current structure of Trello we could see that chats only happen at a card level. We designed a proposal to change that, with the idea that our feature would bring conversations up a level and enable users to have chats at a board/global level.

Design studio and user flows

First sketches from our design studios with a user

With the idea of conversations at a board level, we ran a design studio with one of our users where we generated 10 ideas in 10 minutes. From this session, we started exploring the idea of the conversations living at a global level, much like the boards. We designed for three flows, each of them essential to how the user would interact with the new feature:

We designed for three flows, each of them essential to how the user would interact with the new feature:

Prototyping and user testing

We started by introducing a side panel and a global button for conversations, which would show and allow you to create conversations on a board and card level. We did two iterations, starting from low fidelity up to high-fidelity.

One of the main feedback we got from user research was users didn’t see the point on having a separate function for conversations at a card level. So we designed for conversations only at a board level, keeping the possibility of them being categorized by topics, if they wish, by using tags.

You can interact with our full clickable prototype


The key insight from our user testing was that people loved Trello because of it’s simplicity and they were happy using the existing integrations with communication tools.

Users felt that adding another chat tool would distract them from the main reason the use Trello and for quick responses they would rather stick to user other means of communication.

“Speaking face-to-face it’s quicker and solves any misunderstandings. It’s a lot easier to pick of the phone or go to someone’s desk instead of sending them a message/email”

Given the feedback, our recommendation is that a chat feature would not be the most suitable to increase user engagement. We went back to our research and we identified two possible areas to explore:

Access to Resources: Our research showed us that a lot of users struggle to find resources, finding it too time consuming and fiddly. Our proposed solution would be to design a few different ways (colour code/pictures) to identify/tag resources in order for them to stand out in searches.

Personal Management: Another key finding was that most users find the lack of personalisation for notifications in Trello a major pain point. Currently, there is only three possibilities which is based in frequency not task. We believe that changing those to perhaps being based on task, boards or lists, would impact not only engagement but also issues around losing resources and valuable conversations already happening in Trello.