Weather can be a big deal for companies, especially retailers.
With New York slated to be hit with a snowstorm, it’s a good time to talk about one of the exciting automated news projects we’ve been testing.
Bloomberg is combining databases for weather patterns with store locations to determine which retailers could be hit.
Our algorithms take into account the amount of business disrupted as well as how much it will be affected.
For example, we know 2 inches of snow in Florida has a bigger impact than 2 inches in Chicago.
Today we generated a story about Bab Inc., which has 32 of its 75 restaurants in the path of the storm. Because it’s a small company, automation helps ensure we have coverage.
News automation is often misunderstood as “machines” replacing reporters to translate numbers into stories, often on corporate earnings or sports scores.
At Bloomberg, we view it as an opportunity to overlay separate data sets with smart logic to generate stories that matter in real time.
In the case of weather we did analysis to show that significant snowfall has a short-term affect on stock prices.
We already generate twice as many automated stories as human-written content and are just getting going.