Skip to content

How to measure reader engagement

February 16, 2011

This is the second part of a series on reader engagement, and why it’s important for news websites. Read the first post here. If you need a primer on web analytics, click here.

Ok, so reader engagement is something we should care about. But how do we care about it? How can we measure or even quantify how readers are using a website?

Well, you can’t, at least not exactly and succintly. There’s no single metric that will tell you sum this up without disguising more sins than you’d enjoy on a night out with Silvio Berlusconi.

But there are ways you can measure the different bits that make up the murky notion of “engagement”, and come up with a pretty good picture of what’s going on. Here are the major ones:

1. Time spent on site – Some people use average page duration (APD) or average session duration (ASD) for this. There’s nothing super wrong with that, but it’s a pretty rough metric and won’t tell you much. For example, if you get a lot of referred casual traffic to a story one day, your APD and ASD could drop, even though your regular readers spent more time on the site that day.

I prefer to use readers segments. This breaks down your audience into blocks – eg readers who spend less than 1 minute on site, readers who spend 1-5 minutes, and readers who spend more than 5 minutes. It gives a much clearer picture of what your readers are actually doing, and is less prone to wild fluctuations.

2. Page views per session/user – It’s obviously good if you can make a reader load more pages (more ads, but also they find your content more interesting). I prefer PVs per session, since it’s showing real user behaviour during a visit, but both are useful.

3. Return frequency (aka visits per visitor) – If people are coming back to your site more often, that means they like your content and are more loyal to your brand. Plus the more this metric goes up, the more regular readers you tend to have, and that’s usually what you’re after since they look at more pages, are more likely to subscribe or pay  (vs casual readers), and are a more stable audience to sell to advertisers.

4. Video and gallery completion rates – If lots of people aren’t getting to the end of your videos and galleries, you’re putting work into them that isn’t paying off. Don’t push these as low as possible, since that’s short-changing the readers who do want longer content. Just pick a point of balance between the two that you (and your advertisers) are happy with.

5. Comments – Readers who care about your content are more likely to take the time to post comments on your stories. It also shows you’re putting up stories that your readers care about, so it’s a metric worth tracking.

6. Facebook/Twitter engagement – You can look at followers (try to strip out the fake accounts if you can) but you should also look at how many people liked and shared your content. What’s the reach of your accounts vs the spread of your content through other people’s? How many new followers did you pick up last week?

If you’re interested in better Twitter analytics to measure stuff like this, try Klout.

7. Bounce rate – the ratio of visits with only a single page view to total visits (so the percent that pop in for one article then leave). This is a good indication of how sticky your site is, and how well you’re promoting your other good content. It’s also useful to look specifically at your homepage and other landing pages. High bounce rates here mean readers aren’t finding much to click on.

8. Page depth -This shows you how far readers are getting on your site. Eg how many readers just read one page, how many read two, etc.

9. Return frequency (visits per unique browser) – This is a good indication of how habitual your content is. If your readers keep coming back for more over a given day/week/month, you’re doing a good job of giving them fresh, interesting content. If this metric consistently falls, you’re doing something very wrong.

10. Submissions – If you accept photos, videos, stories or news tips from the public, tracking the submission numbers is a great way to see how people are interacting with the journalism on your site. The more they interact, the more connected they are to your brand.

11. Sign ups – Do you have an email newsletter, or extra features on your site that require a free subscription? If so keep a close eye on how quickly these numbers grow.

12. Compound metrics – You can make up an equation which includes the metrics you’re most concerned about, to easily track engagement in one number. Be careful with these though because:

  • Tracking one number will hide the details of which components are doing well, and which aren’t
  • The weighting you apply to each term in the equation makes a huge difference to the final result.

Here’s quite a complicated example used by

Courtesy of Nieman Journalism Lab

More on this metric at Nieman Lab.

Like with any compound metric, they’re useful as an overview of what’s going on. But don’t expect to be able to read just the headline and understand what the story is really about.

So there’s a dozen ways for measuring engagement. But hold on, you might say – these all look good for a news site to see reader behaviour, but what’s in it for the advertisers? Just because someone spends a lot of time reading my content, signing up, posting comments, etc, doesn’t mean they even see the ads right next to it.

That’s true. I can put up a bunch of long articles and people will spend a lot of time reading those pages, but they’ll still scroll down off the banner ad in the same time as for a much shorter article.

And that’s a big problem if you rely solely on one metric or category of them. And that’s why, of course, you need to look at lots of metrics to find out what’s really going on.

Any metric can be gamed to make a site look more attractive to advertisers (and let’s face it, a lot are). If someone’s bragging about how their site is the stickiest of them all, take a closer look and see what’s really being reported. You’d be surprised how nebulous the metrics become once you start poking at them.

Up next: How to make your site sticky (er…by raising reader engagement)


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: