Scroll to top

Serving local communities with weather warnings meets a clear user need

We’re living through a summer when increasingly extreme weather events are impacting communities across the globe. Correct and up-to-date information is critical for people affected. For local publishers, this is an opportunity to make a real difference to your readers. Service them with reliable extreme weather updates, and they will likely start to come back to your site to stay on top of the situation and plan for what will come next.

There are good examples of services in this space. The New York Times recently launched the Your Places: Extreme Weather newsletter in the US. This free product lets readers pick four locations on which they’d like to stay updated. I now receive a daily newsletter whenever there’s a risk of any of four types of extreme weather – excessive rain, tornadoes, high winds, or hail – in Charlotte, Charleston, Chicago or San Francisco. The email provides a run-down of the risks of extreme weather over the next three days and links to the Extreme Weather Maps page on nytimes.com, which is constantly updated. The New York Times’ reader promise is to “monitor the possibility of extreme weather in places that are important to you”. It’s an impressive and well-designed product.

The Your Places: Extreme Weather newsletter product development process was written up in NiemanLab, with the framing that it’s part of the New York Times’ push into personalised newsletters as well as experimentation with weather data. Clearly it is also there to help fill the subscription funnel – users of course need to register to receive the email and manage their locations. In the NiemanLab piece, John Keefe, Editor of Weather Data, notes that the team saw an opportunity to “provide more geographic diversity” in the service: “Weather alerts have long been the purview of local news organizations. But a lot of readers across the country are in situations where…the local news that’s available to them is dwindling.”

This is undoubtedly true. At the same time, there are many local news organizations in the US that are working hard to fulfil all the information needs of local communities, including better coverage of weather events. One of them is McClatchy, who last year asked us at United Robots to develop an automated content service based on hurricane data from the National Weather Service (also the source for the NYT product). The idea was simple: with automated texts about extreme weather, the local newsrooms would be able to service readers with fast and reliable updates, allowing reporters to focus on covering developments on the ground. 

We built the Hurricane bot ahead of the arrival of Hurricane Ian, the deadliest hurricane to strike Florida since 1935. We provided McClatchy titles like Miami Herald with constantly updated automated articles, which meant reporters did not need to monitor the NWS to write updates, but instead could focus on the quality journalism. Says Cynthia DuBose, VP Audience Growth and Content Monetization at McClatchy: “Our journalists reported evacuations, city preparedness, other efforts. That’s journalism that cannot be replaced by a robot.”

For local media to be sustainable, it needs to be relevant to local readers. It’s about creating reasons for readers to come back to your site, regularly, knowing they will find the information they really need. Information like reliable and accurate updates on extreme weather events. 

The beauty of national weather data, like that from the US National Weather Service, is that it is of high quality, constantly updated and free to access. That means it’s perfectly suited for automated text generation. Local publishers United Robots works with are keen to take advantage of it to build out further indispensable services to their readers. We are now working on a broader Weather Warnings product, with automated updates in a range of extreme weather categories. The aim is for publishers to be able to easily update their local communities on all things weather, including through push notifications. And hopefully create a stronger relationship with their readers, by providing a valuable service when it’s most needed.

Previous
Thriving with AI – some thoughts for publishers
Next
ChatGPT a year on: time to talk about journalistic AI