How we encoded editorial expertise into a content system that scales across five European markets
Five countries. Four languages. Zero content teams.
Canal+ is one of Europe's most recognizable media brands. As they expanded their streaming platform across the continent, from the Netherlands to Austria, Poland to France, they ran into a challenge that had nothing to do with technology and everything to do with people. Content needs to feel local. Viewers in Amsterdam don't want to read a translated press release from Paris. They want recommendations that sound like they were written by someone who actually watches Dutch TV, follows the Eredivisie, and knows what's trending locally.
France and Poland had editorial teams. They had writers who understood the voice, the culture, the audience. But the Netherlands, Austria, and other newer markets had nothing. No writers, no editorial leads, no local content operations at all. And the content needed to be there yesterday.
The editorial bottleneck
Hiring writers for every market wasn't a realistic option. Even if budget allowed it, the timeline didn't. Sports content, especially match previews and recaps, has a shelf life measured in hours. A preview written the day before a Champions League match is stale by kickoff. You need speed and accuracy at the same time, and across multiple languages, that's a genuine operational puzzle.
Then there's the accuracy problem. Sports content is full of facts that need to be right: player stats, team lineups, historical records, tournament standings. Getting a player's name wrong or referencing the wrong match result doesn't just look sloppy, it erodes trust with an audience that knows these details cold.
And every market needs its own editorial tone. What reads well in French sounds wrong in Dutch. What works for a Polish football audience doesn't land the same way in Austria. This wasn't just a translation problem. It was a localization problem, and it required understanding how good editors think about their audience.
The AI Content Factory
We built a content intelligence system that encodes the full editorial workflow, from topic research to published article. Not a writing tool that someone prompts and edits. A pipeline that understands what good content looks like in each market and produces it end to end.
The system handles topic research, identifying what's relevant right now for each market. It structures articles following editorial standards that we developed together with Canal+'s existing editorial teams in France and Poland. It localizes content for each language and culture, not translating but rewriting with the right tone, references, and context. It runs plagiarism checks. And it delivers finished content directly into the CMS, ready to publish.
Sports, films, series. Match previews that go live hours before kickoff. Film recommendations that reference local viewing habits. Series roundups timed to release schedules. All of it localized, all of it automated, all of it publishable without a human editor touching every piece.
“We didn't replace editorial judgment. We learned it from the teams that had it, then built a system that could apply it at scale.”
From Dutch pilot to Paris headquarters
It started in the Netherlands as a pilot, the market that needed content most urgently and had the fewest resources to produce it. Within weeks, the system was generating localized content that matched the quality and tone of what the French and Polish teams were producing manually. That got attention.
The project expanded to five countries and four languages. Paris headquarters was briefed on the approach and the results. What had started as a practical solution for one market became a new way of thinking about editorial operations at scale across the entire Canal+ streaming ecosystem.
The impact went beyond just content output. One team member who worked closely on the project was so inspired by the approach that he started his own company building on similar principles. That's not something we planned, but it says something about what happens when you show people a genuinely new way to think about their work. The intelligence layer we built didn't just solve a content problem. It changed how Canal+ thinks about scaling editorial expertise across markets, treating good editorial judgment as something that can be learned, encoded, and applied wherever it's needed.
