
The Buzzard: Showing all sides
The team behind The Buzzard news website is training a machine-learning algorithm to include all sides of a debate and provide users with a balanced, holistic overview.
Every publisher and journalist knows the value of charts and wants more of them. Done well, charts throw light on difficult topics – they make stories easy to read by breaking up oceans of text. The reality is different. Busy newsrooms don’t always have the tools or skills to create great charts. And in a world of daily, if not hourly, deadlines, they often don’t have the time.
Frames is a new business established by Portugal’s leading online newspaper with support from the DNI Fund. Its mission is not only to make great charts simple to incorporate – but to increase page views and generate revenue at the same time.
“Together with some financial support from parent company Observador, support from the DNI Fund has enabled five people to work full time on the Frames concept since 2016,” says Leo Xavier, the CTO at Observador. "It’s funding that’s already delivering benefits. Before we started Frames, only a small fraction of articles on observador.pt used charts. Today, that figure is closer to 50%. As well as improving reader understanding of complex issues, the charts are generating 5.6% more page views. Even more impressively, they’ve created a new revenue stream for Observador. Companies such as Vodafone, Deloitte and SportTV are now sponsoring charts relevant to their businesses – and this is already contributing gross annual revenue of around €100,000 for our publisher. We now have the revenue stream to support Frames’ continued operation as a standalone business.”
This is how Frames works...
This is how Frames works...
A dedicated team in the Frames newsroom continually monitors current events and decides what charts could best help readers make sense of them. If there’s an earthquake in Mexico, for example, the team might create a chart showing the magnitude of previous earthquakes. Meanwhile, a journalist at the publisher writes up the earthquake story exactly as usual. Software on the content management system analyses the story and, as the journalist is typing, if there’s a good match the software automatically integrates a relevant chart – in this case, one about earthquake magnitude – into the layout. That’s all there is to it. There’s no extra work for the journalist. And he or she can delete or edit the chart if they don’t feel it adds to the story.
We think that Frames has applications right across the industry,” Leo adds. “So we’re now in the process of using Observador as a case study and talking to publishers in other countries, explaining how Frames can make a real difference to their newsrooms and their readers – and make money at the same time.”
The team behind The Buzzard news website is training a machine-learning algorithm to include all sides of a debate and provide users with a balanced, holistic overview.
An in-house training platform that enables journalists to seize the opportunities offered by new media.
QuoteBot is a robotic solution that frees up journalists’ time by automatically producing personalised articles and structured data.
Google serves cookies to analyse traffic to this site. Information about your use of our site is shared with Google for that purpose.
Find out more.