Pageviews have been much maligned as a measure of web traffic for the last few years. They’re flimsy, prone to being distorted by artificial inflation, and have generally fallen out of favour as a universal metric. Instead, web analytics for content is fracturing, with news and content businesses starting to promote their own individual ways of measuring what matters, while other businesses make big money by producing proprietary, comparative metrics that let competitors analyse each other and help agencies decide where to spend money.
Upworthy is the latest business to come out with its own way to measure itself: attention minutes, a composite metric that takes in whether and how its users are engaged with what it’s producing.
We built attention minutes to look at a wide range of signals — everything from video player signals about whether a video is currently playing, to a user’s mouse movements, to which browser tab is currently open — to determine whether the user is still engaged. The result is a fine-grained and unforgiving metric that tells us whether people are really engaged with our content or whether they’ve moved on to the next thing.
Upworthy’s making what looks like a sensible move here, focussing on the things that make them money and with a clear understanding of how they’re going to act on their new number. The most important questions about any metric for content online are as follows:
- How does it make money?
- How can I act on it?
If any metric doesn’t have clear answers to those things, it’s probably not something you should be asking your newsroom to try and affect. There’s no point in having targets based on numbers you can’t do anything about, especially opaque ones made up of multiple factors that are difficult to tease out. (This is why anyone even thinking about using the phrase “Klout for content” should desist immediately and repent.)
Upworthy, as with Medium, is finding a metric that works for its business, which doesn’t rely on banner ads and therefore doesn’t make money from pageviews. From the looks of things it’s found a solid metric that allows editorial staff to make decisions about what they promote; closing the circle like that is absolutely crucial.
But one big metric for everything is an unachievable goal for most content businesses, especially if you want more than just an inscrutable number: you need complexity, granularity, and ways to actually affect the number. That granularity comes from segmentation, not from bundling more numbers together, but it isn’t at all important for everyone to see every metric that matters – as long as those who can change things see the ones they need to change. How do reporters, writers, editors, people who are making and working directly with content, make a dent in a metric like “engaged time on page” in real time?
What matters for the business is different to what matters for newsrooms on a day-to-day basis. Of course it’s important to be aware of your views per visit, your engaged time on page, your DAU to MAU ratio, your signed-in user rate – whatever metric you’re using as a proxy for loyalty or engagement – and to understand how to make differences to it over time. But most people in most newsrooms will struggle to articulate what those metrics are, and will struggle even harder to take actions to increase them.
Which is where pageviews come in. Pageviews have a lot of advantages over other, more complicated metrics. They’re relatively easy to measure. In a business where display ad revenue matters, or which revolves around driving traffic to content in some other way, they have a clear value attached. And they’re immediately and obviously affected by activities people do in newsrooms: editing, commissioning, subbing, promoting.
Everyone in a newsroom can get behind the idea of maximising the number of potential pageviews for a piece of work – the number of potential reads. They’re an equaliser: a click from a front is the same value as a click from social media or search or from a different article, so anyone with skin in the game in any of those areas can make a difference. Everyone can get invested in that number, because they feel like they can control it, even if only a little. If you’re trying to get people to act on data in real time – rather than just building a dashboard that looks pretty on a big screen – that usability can’t be an afterthought. It has to be baked in.
Honesty matters too. If you’re inflating your pageviews with galleries or autorefresh, you’re creating perverse incentives for your editorial staff, who are going to take the easy routes to boost that number. If you’re gaming the numbers as a business, your newsroom will game them too. That leads to big problems – but those problems are with the dishonesty, not the metric itself. It also makes it unsuitable for external comparisons, unless you’re sure all your competitors are no more or less honest than you – but in a world where every competitive metric is plagued with accuracy, comparability and gaming issues, that’s not a crippling defect when it comes to internal use.
Pageviews aren’t a bad metric, if they’re used well. New metrics like attention minutes or engaged time on page aren’t either, as long as they’re being used sensibly. If the number that everyone understands and everyone can act on is pageviews, you should be using it.
Thanks for this Mary. It’s long struck me that pageviews have long been disingenuously maligned by journalists working in newsrooms undergoing a transformation to digital.
Disingenuously because those sanctimoniously sneering at pageviews typically come from print newspaper backgrounds where circulation numbers were (and in many case still are) everything.
Our printed newspaper is doing so well – look at its circulation numbers … so many people are reading our articles! Our digitally delivered news is doing so well – look at … well, don’t look at the number of people that read our articles, that’s a meaningless metric!
Every metric has both the potential to inform decisions and, yes, be misused. But those in the news business that dismiss the value of pageviews without postulating more meaningful metrics that support their business objectives don’t come across as particularly credible.
However much it hurts, flat or declining pageviews should prompt the question, “why is our traffic – our ability to connect with readers – flailing or failing?” To respond instead with “pageviews don’t matter” is a not particularly elegant way of sticking one’s head in the sand.
Totally agree. There’s also the problem of conflating external and internal usefulness – pageviews aren’t a particularly sophisticated metric for external comparison, especially when they can be gamed so easily, but that certainly doesn’t stop them being useful internally.