My friend and former colleague Simon Rogers, editor of the Guardian’s Data blog, has posted a defence of the increasing use of data visualisation. I agree to a point, but I also think it’s really important to remember that visualisations are not the only product of data analysis. They can help readers see patterns in complex sets of data, but I also think that sometimes we’re missing other opportunities with data analysis by focusing on data visualisation. Sometimes, the result isn’t a visualisation but a key insight that underpins a story. I often worry about the problem of seeing a world as full of nails when you think all you have is a hammer. Sometimes, visualisations are just not the right end product of data journalism.

I’ve heard statisticians grumble about information being seen as simply beautiful instead of being, well, informational. Good data visualisations hit a middle way being being beautiful and simplifying complex concepts. I’ve heard designers grumble about data visualisations that aren’t beautiful, and they rail away against the lack of aesthetic of some of the publicly available tools. Sometimes people are using the wrong chart or visualisation to visualise their data. When it comes to charts, I often show this simple chart during training, which really breaks down what types of charts or visualisations are appropriate for what kind of data you’re working with.

I am always in favour of the democratisation of tools, but when it comes to digital story-telling, editors need to remember all of the techniques available and have a clear way of deciding which technique is appropriate.