Only say “over” or “under,” or “more than” or “nearly,” with round numbers. Otherwise you just sound silly.

There are lots of correct ways to describe this data. The Journal didn’t choose any of them.

There are lots of wrong ways to add visual interest to charts. McKinsey shows us the right way.

James Clear has sold 15 million copies of a book that apparently contains nothing but made-up charts and graphs.

The best way to explain an abstract number? Encourage your readers to imagine something concrete and familiar.

We think square pie charts are amazing. McKinsey & Company agrees.

You can’t just present two facts that sound logical together and call it causation.

When someone says “up to,” you can be pretty sure they’re the ones who are up to something.

Yes, data stories must be correct and reliable. But they have to be meaningful as well.

One day soon you’ll be glad you did.