The Pros and Cons of Personalization

When The New York Times recently announced plans to make its sites and apps more personalized, readers posted nearly 300 overwhelmingly negative comments. Many were angry or afraid that algorithms would narrow horizons and keep people less informed.

Whatever some commentators would have you believe, it is unlikely that the negative comments reflect the views of a majority of New York Times subscribers. According to the Reuters Digital News Report, most of the 50,000 people in 26 countries, who took part in a YouGov survey about news consumption, prefer a personalized news service based on recommendations.

The United States is among the 26 countries surveyed.

“Algorithms are more popular than journalists, as long as the algorithms are based on people’s personal consumption,” concludes the Reuters report for 2016.

The truth is that we have always personalized the way we consume the news — which stories or pages we read — while editorial teams have always filtered and curated the news. Journalists decide what readers will and won’t see.

It is enough to contrast CNN and Fox News coverage of events to see how much filtering goes on in the newsroom. Traditional linear broadcast news is the antithesis in many ways of the personalized services now being developed.

The New York Times and other mainstream news organizations are starting to use data to improve the relevance of the service they deliver. On a very basic level, it could be reorganizing the news to give readers what they want, in the order they want it, while ensuring that they continue to receive all the really important news.

A recommendation engine will suggest that if you liked a particular story, you might also be interested in other similar content. It is geared to helping users find content, including audio, video, text or infographics they might otherwise have missed.

A good example is Yle’s NewsWatch app, which has already been around for several years. It offers users a highly customised and personalised experience.

The news is based on individual topic choices, reading history, what is trending and what Yle’s editors select as the day’s top stories. The Yle app allows users to choose from 140,000 topics: for example, users may be interested in Cristiano Ronaldo stories, but not in football.

The idea is that users crave content that is both relevant and personal. Given that there is no shortage of competition, the fear is that if news services do not offer users what they want, they will simply look elsewhere.

Personalization is at the heart of news and media strategy around the world. Even the BBC, with its famous remit to educate and inform the public, is committing resources to a very ambitious personalization strategy.

The BBC is expected to introduce mandatory sign-up for its on-demand service, iPlayer, in the next few months. It will identify news and sport users as personalization is also about segmenting audiences to serve them better.

Indeed, newsletters and vertical websites dedicated to sport, health and well-being, or fashion, is a facet of personalization strategies that is often overlooked. Never before have consumers had access to such a range and depth of quality, specialized content.

“The closer and more personal our relationship with our audiences, the more I’m certain they will choose the BBC,” the BBC’s director-general, Tony Hall, told staff earlier this year. Certainly, there is plenty of evidence to suggest that increased personalization will retain, engage and grow audiences.

Notifications based on personal preferences, for example, will help bring users back more regularly and the NYT hopes, drive subscription. Media companies will be looking chiefly at two categories of metrics to gauge the impact of their personalization strategies:

• Consumption, including views and downloads
• Engagement, including session duration, conversion and sharing

Consumers are likely to enjoy a better experience and to consume more, not less content. Those who don’t like it will be able to ignore the recommendations or opt out.