Bloomberg News editor in chief John Micklethwait sent the following to the editorial staff on Wednesday:
To All in Editorial & Research,
We are still publishing too many enterprise stories that are either mediocre or too long, or both. Before you embark on a piece of enterprise, ask yourself whether it’s really going to be worth the investment, And if an idea doesn’t turn out as envisaged, be prepared to spike it and move on. A week or two spend on source-building, without any story published, is often a much more valuable use of our reporters’ time.
The key person to think about is the reader. They are busy people — who normally read only one screen or a story and SELDOM read more than two screens. (A screen typically is around 300 words, though a lot depends on the illustrations.) That does not mean no long stories. As I have said before, a long story on a complicated topic can save readers time if it replaces the need to read a lot of short ones. We need to do pieces that join the dots. And everybody will always read something long of it is fascinating. Some of our most-read stories are from Businessweek. But is has to be good enterprise. Some of our enterprise stories at the moment should really have been Blasts — and a few should just have been spiked.
In general our readers either want to have a quick piece of information or something that justifies the longer read. A good rule of thumb is that stories should either be shorter than 400 words or longer than 900. You either have a simple news story or a single observation (in which case go short) or a yarn or a chance to explain something very complicated (in which case go long). But taking 700 words or 800 words to make a single point will just annoy readers.
In order to better measure the volume and readership of enterprise stories, we’re going to ask editors to code these stories to NI FEA. So yes to good enterprise. But otherwise keep it short. Think of the reader.