Bettina D'ávila/ Sr Product Designer

Dashboard for CXO campaigns management

AB Tasty is an all-inclusive B2B/SaaS tool for website and conversion rate optimization, personalization, user engagement, and A/B and MVT testing. Marketing, product and tech teams use AB Tasty’s AI-driven experimentation, personalization and product optimization platform to rapidly drive ROI and continuously deliver outstanding user experiences.


This project was the first one we built together as a team, so we knew that we had challenges ahead. The dashboard on AB Tasty is one of they key pages on the platform where users see, manage, edit and create new content. It provides a global overview, with access to the most important data, functions and controls. In AB Tasty's case, the campaigns dashboard is also the homepage in the platform, for newbies and power users. And because none of the previous squads had any ownership on this project, the Dashboard has generated a huge design debt throughout the past years.

Why we did this project

The previous Dashboard screen, the object of this project.


The design process started around two months before preparing the backlog for the 6-week cycle with the research phase. Inspired by the Double Diamond Diagram framework and Design Thinking methodology, I focused on the Discovery phase first, trying to collect as many data and feedback as possible before framing the problem.

UK Design Council's Double Diamong Diagram framework

At that point the team has already gathered a lot of existing feedback from clients from previous projects and from different team. Therefore, I conducted new interviews and new usability testing with same clients, plus with other clients that that also fit in the scope, in addition to Customer Success Managers and stakeholders.

Snapshots from the usability testing sessions that I conducted with Loom. Although I was side-by-side with the interviewees, Loom supported me in recording their screens and their faces for later analysis.

After detecting the main pain points and prioritizing the most critical ones, we decided on the scope for out MVP. We knew the project was big so therefore we used the MoSCoW method to cut it into different versions, so we were sure we could deliver something valuable to the end clients at the end of the development cycle.

We planned a development cycle of 6 weeks, using sprints of 2 weeks (3 sprints in total). But the entire project took 16 weeks instead, due to some bumps on the way (COVID-19, less working hours, change of people on the team).


I started the ideation process for our MVP + future versions. In terms of solution, I decided it was better to design not only the MVP but the full scope to make sure that the MVP would be scalable. This approach affects not only the UI and UX decisions but also the development process. We wanted to make sure that all decisions we made as a team would be futuristic: making the MVP scalable to embrace improvements and new features we know will come.


I started creating different versions of mockups that were iterated and tested internally, using tools such as Sketch, Zeplin, InVision, Adobe XD and Loom (to record the interviews). I used Sketch to build the mockups, Zeplin to show the specifications to the development team, InVision to create the click-dummy prototype for the usability interviews and Adobe XD to generate animated mockups to explain interactions and transitions to developers.

Sketch prototypes
Prototypes in progress in Sketch
Animation created with Adobe XD to show the transition between tabs

A new round of usability testing sessions took place after the project was validated by the C-level board. I wanted to make sure that the final solution for the MVP was correctly solving the problems we framed before starting the development cycle.

Main pain points & solutions found

State vs. action

  • Lack of consistency and clarity on campaigns status: not clear when a test should be played or paused.
  • The solution found was to visually separate the static elements from buttons, which require action. We also cleared up the dashboard view and hid secondary actions on hover, improving usability and reducing clicks.

    Status visibility

  • According to many clients, it was impossible to know which campaigns are in QA mode. This caused distress and waste of time before launching a campaign.
  • A simple change in the UI solved the problem. The QA mode status is now quickly detectable amongst the other status buttons thanks to its outlines.

    Same functionality, different UI

  • Filter components look more like CTAs than filters.
  • All the filtering actions were gathered with the same UI and same behaviour. Things that look the same should behave the same - for better usability.

    Folders management

  • Folders functionality is confusing. The folder components appear in 3 different places on screen, with 3 different UI and behaviours. The user engagement with this feature is very poor.
  • Currently the folder functionality behaves like a filter. So I removed all the previous components and created a single new one: a dropdown list with dedicated actionable items on hover including a new "create" CTA at the bottom.


    We planned a development cycle of 6 weeks, using sprints of 2 weeks (3 sprints in total). But the entire project took 16 weeks instead, due to some bumps on the way (COVID-19, less working hours, change of people on the team).

    With this project, while addressing many user pain points accumulated over the past months, we were able to update the UI, increment small but smart UX improvements (without disrupting the current knowledge that users have from the tool, avoiding a big learning curve) and also launch new features.


    Challenges & Learnings

    This project was the first one we built together as a team and, at the beginning of the project, we didn't have a dedicated Product Manager in our squad, which "forced" me to grasp some of the PM's responsibilities side by side with the tech lead throughout the project. All in all, I am very happy to have done this very complex project with a balanced team and a very structured process. It was the first time I took part on backlog prioritisation and Sprint planning rituals, which means I was able to contribute with the Design perspective to the process of project management. For instance, we added Design review as mandatory step in the workflow for developers and we added the research process and its deliverables in the Sprint planning as well, delivering design value and context to the project at the end of every sprint.


    Product designer: Bettina D'avila
    Tech lead: Manhulay Siriphol
    Front-end developers: Jihan El-medini, Yanis Tam, Romain Giordano, Florian Eysseric
    Product Manager (arrived at the end of the project): Christelle Mampuya

    Latest projects

    Dashboard for CXO campaigns management

    B2B Web Platform

    AB Tasty is an all-inclusive B2B/SaaS tool for website and customer experience optimization.

    Mobile app for CXO campaigns management

    B2C Mobile App

    B2B mobile application for end users to follow and monitor their campaigns on-the-go.

    Helping users to create a new password

    B2B Web Platform

    Changing our password is often necessary and, sometimes, mandatory. But never a fun task.