Modular - Tell Me Like I Am Five

A while ago, we noticed something strange: a lot of people weren't reading news stories anymore. Some of our colleagues blamed social media. Others said readers had lost interest in important things. We decided to look at ourselves first. Maybe it was time for journalists to try something new.

We went back to our best stories and realized they were often too long, too cryptic – and didn't always answer the questions people actually cared about.

So we had an idea. What if we cleaned and chopped those big stories into smaller, useful pieces – like puzzle blocks – and kept only the parts that helped people understand the world?

We called this modular journalism. Each paragraph answered a clear question we knew users asked, like "What happened?" or "Why is this important?" That way, readers could get just the pieces they needed.

We didn't have fancy robots yet, so we used our brains – and a lot of patience – to do it by hand. We also noticed that some story pieces weren't just unhelpful… they were tricky, pretending to be facts when they weren't.

That early experiment taught us how to build better stories – shaped by what readers need and care about, not just what we, as journalists, think they should know.

Later, when AI got smarter, we asked: could robots help us build these puzzle-stories too?

That's the next chapter.