Turning complex data into a tool journalists can trust

Immersion with Founders

Since MindBugs Discovery was a highly technical project, my first step was to fully understand what the tool was, how it worked, and what the founders aimed to achieve. I held several meetings with the developers (who were also the founders) to map the technical logic behind the backend, the data sources, and the intended user outcomes. This gave me a clear picture of both the goals and the constraints of the product.

Reviewing Existing Design & Research

Next, I reviewed the existing interface and the interviews the team had already conducted with journalists. By clustering the insights, I identified recurring pain points such as:

• Navigation problems – users struggled to find their way through the platform.
• Unclear meaning of data – visualisations weren't self-explanatory.
• Identical visual representations – charts all looked the same, so journalists couldn't tell features apart or understand the different functions of the tool.
• Fear of interaction – because of this lack of clarity, journalists felt hesitant to click or explore, worried they might 'break' something.

This groundwork helped me define the priorities for the redesign: making the interface clearer, simplifying flows, and ensuring that complex data could be presented in a way that journalists could confidently use.

Working Within Constraints

Given the very strict timeline, the founders wanted the redesign implemented quickly. We decided on using the Metronic design theme as a foundation. This allowed us to:

• Save development time by relying on pre-built, consistent components.
• Ensure a professional, responsive baseline without custom overhead.
• Focus design efforts on information hierarchy, clarity, and interaction patterns rather than reinventing the UI from scratch.

This decision meant we could move fast while still delivering a clear, functional, and trustworthy interface for journalists.

Dashboard Redesign

• Rebuilt the information hierarchy: critical functions surfaced at the top, related actions grouped logically.
• Simplified navigation into a few clear entry points.
• Added visual supports (maps, charts, and colour cues) to make data more approachable.

3D Graph Rework

The core of the tool was a 3D graph showing how disinformation narratives connect. Narratives could be linked at different levels — by country, by keywords, or as part of a larger overarching narrative.

To make this complexity usable for journalists, I redesigned the graph around two key improvements:

• Filtering options – allowed users to view connections at different levels (country, keyword, or parent narrative) and switch perspectives without losing the overall structure.
• Node cards – clicking a node opened a card with further explanation (when available) and options to expand navigation, letting users dive deeper into the narrative while still keeping context.

This made the graph both explorable and understandable, turning what was previously overwhelming into a tool journalists could use with confidence.

Ongoing Journalist Feedback

We ran interviews with the same group of journalists who had been part of the earlier research. This continuity allowed us to directly compare how they experienced the old vs. new design. Their feedback highlighted how the redesign improved clarity and reduced hesitation when exploring data.

We also collaborated with a journalist advisor who acted as a continuous sounding board during the design process. Their practical insights helped validate design decisions and ensure the tool aligned with real investigative workflows.

Real-World Testing at Conferences

The project was tested in real-world conditions at two major international events:

• PI World Journalists Conference — Here we tested the tool with a diverse group of journalists from different specialisations (investigative, political, regional correspondents, data journalists). Watching them in action gave us invaluable insight into how different research approaches shaped the way they used the platform.
• French press conference — The tool was also showcased at a major French event attended by political leaders, where it gained international visibility. Fun fact: French President Emmanuel Macron briefly interacted with the interface and commented: 'Nice design.'

These testing rounds not only refined usability but also demonstrated the credibility and appeal of the tool to both practitioners and stakeholders at the highest level.

Prediction & Alerts (Planned)

As part of the design process, I also developed a future improvement plan for predictive features. The concept was to integrate alerts that could warn journalists when new disinformation narratives emerged or when existing ones gained unusual traction. This would have transformed the tool from a retrospective analysis platform into a proactive early-warning system.

Although this feature wasn't implemented during the project due to scope and time constraints, it demonstrated the potential roadmap for evolving MindBugs Discovery into a more powerful investigative companion.

Final Reflections

The redesign turned a highly technical backend into a tool journalists could actually use in their day-to-day work. Iteration with real journalists, close collaboration with developers, and strategic use of existing design systems all made it possible to deliver a usable product under tight deadlines.

Most importantly, the project proved that complex data can be transformed into clear, actionable insights — and that design can play a decisive role in enabling journalists to confront the global challenge of disinformation.