Why did Make Sense decide to focus on this project?
The idea came from a very practical pain point: journalists in small newsrooms often have to manually repurpose the same content into multiple formats (e.g., news articles, social media posts, short scripts). This process is time-consuming and often inconsistent — especially when using general AI tools like ChatGPT without editorial alignment.
Make Sense wanted to change that. Content Transformer is an AI-based tool that allows journalists to automatically rewrite content into up to seven different formats — all while preserving the editorial style, tone of voice, and clarity. The tool significantly reduces routine work and saves valuable time. What used to take 30–40 minutes per format can now be done in under 5. It’s like a virtual editorial assistant that already knows your newsroom’s style and audience needs — no need to rewrite prompts every time or explain “how we usually write.”
For Make Sense, it’s not just about automation — it’s about strengthening editorial resilience, ensuring consistency across platforms, and freeing up time for deeper reporting and storytelling.
How did applying design-thinking principles help?
The team applied design thinking mainly during the research and prototyping phases. User interviews with editors revealed that the real challenge was not the quality of AI output, but that each journalist used AI differently. This insight guided the creation of global prompts, ensuring a consistent editorial voice across all formats. Iterative testing and feedback allowed Make Sense to introduce each new feature — such as format templates or the “editing mode” — only after testing it with the team, refining functionality step by step to match newsroom workflows. Engaging users continuously ensured the tool addressed practical problems and built trust, as editors could see their feedback directly reflected in the evolving product.
Did Make Sense’s approach change engagement with its audience in any way?
The team has completed the MVP version of the tool and is currently testing it with journalists. To fully assess the impact on audience engagement, they need additional time and resources, which Make Sense plans to secure at the next stage of development.
What challenges did Make Sense encounter and how did it address them?
The main challenge was developing a digital product rather than a traditional media project, which required a new management approach, including detailed technical documentation, versioning features, and close coordination with developers. Budget limitations meant the team could not implement all planned features at once, so they prioritised core functionality — automating content generation while preserving editorial style. To address these challenges, they focused on building a clear MVP, scaled functionality gradually, and maintained close communication between the editorial and technical teams to ensure the tool met real newsroom needs.
How was the project received by Make Sense’s audience?
Testing showed that the tool creates quality text adaptations for social media. Journalists noted the clarity of the interface and the ability to quickly adapt materials in different styles defined by the editorial team. This simplifies the work of social media managers and allows for the maintaining of unified communication standards, even for staff without much experience.
What insights or lessons did Make Sense gain from this project?
The main insight was that clear editorial styles in the tool’s settings are critical for quality results. An
intuitive interface lowers the entry barrier for the team and accelerates implementation. Also, budget significantly more time for development due to risks inherent in such projects. Currently, only large companies provide reliable development timelines, but they come with very high costs. Working with freelance developers is challenging due to the lack of guarantees for timely project completion. For projects like this, you need a strong product manager to create specifications and manage the development process.
To scale the project, additional funding is needed to refine functionality and integrate with editorial workflows.
What does Make Sense plan on doing next?
The team plans to continue developing the Content Transformer, implementing remaining features such as customisable templates, additional prompts, Google login, user dashboards, text-formatting tools, and user roles. Their broader goal is to launch a cloud-based SaaS version for local media to make the tool scalable and accessible.
What advice would Make Sense give to other journalists or organisations considering a similar path?
Focus initially on one specific editorial task instead of trying to cover everything at once. We chose text adaptation to different formats — this allowed us to quickly create a working MVP that can be gradually developed. Build in the ability to configure prompts, styles, length, and other parameters — this ensures fine-tuning of tone of voice and control over media identity. Use AI not only in the product but also in the development process — this saves budget. Plan the architecture with the ability to integrate different LLMs and add new features, such as planning and publication control. Invest in interface simplicity: the tool must work for the entire team, regardless of technical level. Test with real users as early as possible and budget for support after launch.
Credit: Transitions, Journalismfund Europe's partner for implementing the Microgrants for Small Newsrooms programme, produced this case study.
