Does a real-time dashboard turn journalists into click chasers?

Something has gone wrong when a newsroom's success is judged solely by which stories get the most clicks. A substantial investigation can end up looking rather poor next to a hastily written story designed to provoke outrage. On the screen, at least.

The criticism of real-time analytics in journalism is understandable. Clickbait, traffic targets, the constant pursuit of the next spike: these are legitimate concerns. When the numbers are always in view, people can gradually start shaping their work around them.

At NowMetrix, we build a real-time analytics tool for newsrooms ourselves. That gives us a particular responsibility to engage with this criticism.

The effect numbers can have on people is well documented. In a 2015 study for the Tow Center, researcher Caitlin Petre described how audience metrics and rankings can produce feelings of competition, validation, pressure and self-doubt, among other responses. A dashboard influences what people pay attention to and how they judge their work.

There is another way of experiencing those numbers that belongs in this discussion, too. Over time, a newsroom can develop a feel for its own website: how the day is unfolding, which stories are reaching people, and when something deserves a closer look.

At first, the main thing you notice might be a single number: how many people are on the website right now. Gradually, the changes acquire context. You become familiar with the pattern of an ordinary morning. You see what happens after a newsletter goes out. You notice when an older article starts attracting readers again, or when a story holds people's attention for longer than expected.

That sense of the website comes from combining observation with editorial experience. You know what has been published, which stories have moved on the homepage and which events are shaping the day. The numbers add to that knowledge.

They can also challenge your assumptions. A subject you thought was niche might attract more attention than expected. A carefully written explainer might still be finding readers long after the initial news story has faded from view. Observations like these can be a reason to give a subject more space.

A simple example shows how useful this can be. A few months ago, a newsroom published a guide to passengers' rights when trains are cancelled. During a major disruption, that article suddenly starts receiving a lot of visits again.

The team now has some very practical questions to consider. Is the information still accurate? Have the compensation rules changed? Do the links still work? Does the article need a note about the current disruption?

If the newsroom notices the renewed attention, it can check the guide while people are using it. There is editorial value in that work beyond the number of pageviews.

Low readership can be a useful observation, too. Suppose a newsroom publishes an important story about funding for local schools. It receives very few visits.

That alone tells us little about whether people are interested in the subject. Perhaps the article is buried near the bottom of the homepage. Perhaps the headline is so abstract that readers cannot see what the decision means for their own families. Perhaps the story needs more time.

Before deciding to publish fewer stories like it, the newsroom should consider those possibilities. A more precise headline or a clear explanation of the consequences can help an important subject find its audience. The headline must, of course, deliver on its promise.

The limits of the numbers still apply. On its own, a pageview tells us that a page has been viewed. It does not establish agreement, trust or journalistic quality. And people who never reach the website remain largely invisible from this perspective.

A rising curve does not always reveal its cause, either. A more prominent position, an external link or a news event could all play a part. Learning from the numbers requires context and other observations. The Reuters Institute describes this very aim: analytics should support editorial judgement with relevant information.

A newsroom therefore needs a clear idea of what it uses live data for. A list of the most-read articles is a poor way to rank the best journalism. Subjects, intended audiences, placement and publication times differ too much for that.

The freedom to keep pursuing a story regardless of its current reach matters just as much. Wrongdoing deserves investigation even if, at first, it affects only a few people. Careful analysis deserves space even when a lighter story is attracting far more attention.

In everyday work, that also means being able to look away from the screen. Nobody needs to follow every change in visitor numbers while working on an investigation. Live data is useful when there is a concrete decision to make. Understanding whether a story has an impact over several weeks takes a longer view and other kinds of feedback anyway.

For us, this sets a clear expectation for NowMetrix: newsrooms should be able to understand quickly what is happening on their websites. They should find useful signals that help them make their journalism easier to discover, more up to date and more useful to readers.

Having a good feel for a website includes being able to put the numbers in perspective and, sometimes, deliberately set them aside. A story may stay on the homepage because the newsroom considers it important. Even when another one is getting more clicks.