Reset password:

Strategic insights
Identifying Your Real Readers, Not Just Traffic

When it comes to web analytics, we have no idea how many people actually read an article, nor how many of our visitors consider themselves to be readers.



Written by on August 16, 2011

Shared By Plus Subscriber

Matthew Neale

This is Baekdal Plus content. It is shared with you for free by a member. Please reshare it.

Here is a simple question. How many readers do you have? Real readers? Not unique visitors or other misleading statistics.

Several newspapers have studied how many of their readers who are willing to pay for news. the result was that about 5% say they would. But when you look at what actually happens online we are not reaching 5%:

  • 0.68% has subscribed to New York Times
  • 1.9% has subscribed to the Financial Times
  • GigaOm Pro is around 0.4%
  • The Times is around 0.5%.
  • ...and I am seeing the same with Baekdal Plus.

That is nowhere near 5%.

Part of the problem is probably the way the surveys were made. The newspapers would call people and ask, "are you a reader of New York Times?" and if people said yes they would ask, "would pay for online news?"

They are only asking the people who consider themselves to be readers. If you just picked a copy of the New York Times while waiting for the train, you don't consider yourself a reader. A reader is someone who reads the paper on a regular basis.

The 5% is probably right. The question is just ...5% of what?

When it comes to web analytics, we have no idea how many people actually read an article--nor how many of our visitors consider themselves to be readers.

Real readers are a combination of several things. It is a person who sees several articles, read the articles (as opposed to just clicking on them), comes back often and is not referred from another site.

Let's analyze these one by one.

How many actually reads an article?

For the past two weeks I have been analyzing how many people reads an article.

I measured if the article was longer than the screen itself. I measured when people started scrolling. And I measured when they reached the end of the article and how long it took them to get there.

For most articles the result looked rather disappointing. The view-to-read rate (how many times an article is being read out of the total number of views) was just 11% for most articles.

But after running the test for a couple of days I started to see an interesting pattern emerge. Every time a page increased in popularity it didn't just increase the traffic, it also increased the view-to-read rate.

The view-to-read rate for popular pages was around 41%.

Even more interesting was the social effect. The the view-to-read rate skyrocketed whenever an article was talked about on Twitter. I didn't measure this specifically (I failed to anticipate this), but it was clear that people were much more likely to read an article if it had been recommended by someone they know.

This was a significant find that I will study further in the future. I knew that social was important, but I didn't expect to see such a drastic difference.

Looking at the overall traffic (all articles) the average is that about 28% of all page views results in the article being read.

Another interesting element is the behavior in between just viewing an article and reading it. 68% started to scroll down the page but never reached the end. 43% scrolled all the way down to the end, but so fast they could not possibly have read the article.

There can be several reasons for this. One could hope that people were just glancing at the length of the article only to bookmark it or add it to "read later." But most people probably only looked at how long it was and decided they didn't have time to read it. Unfortunately, there is no way to measure it.

How many articles do people see?

The next thing we have to look at is the number of articles people read. The question is, "how many articles do people have to read in order to be considered a reader?"

This, of course, varies greatly from one site to another. For this site, I might say that people should at least read four out of the 14 articles I publish every month, before I can categorize them as readers.

Unfortunately I know of no analytics tool that can actually measure this. They are all measuring something else.

How many times do people come back?

Next question is how often do people come back to your site. A person who came to your site 3 weeks ago but never returned isn't really a reader. You need to measure each person's loyalty to your articles.

You will have a graph like this one. It is showing how often people come back to this site.

There are several problems with this graph. It is just measuring visits. It doesn't tell you what people do on your site. A person coming to the site five times but never reads an article is worthless.

I can come up with vague assumptions. E.g. If people have to read four articles to be considered, they probably also have to visit the site at least that many times. Based on that I can disregard everyone who have just visited this site 1-3 times. The result is that only 11.68% of my visits are potential readers, for you the numbers might be different.

Referrers don't count

Finally, you also need to distinguish between people who specifically wanted to read your articles, vs. people who just happen to click on a link.

E.g. Here is a link to an article over at Forbes called "Building a Digital Audience: How Forbes Staffers and Contributors Do It."

If you click on that link and the read the article, you are not really a reader of Forbes. You are a reader of this site, and the only reason why you read the other article was because I told you to. That is an important distinction to make.

You might turn into a reader over time, but the traffic you have from other sources are not (yet) readers.

On this site 13% come via search, 3% via other sites, 73% via social networks and only 11% come directly to the site.

It is the same for readers who visit you via news aggregators like AOLs new "Editions." If one the "sources" in their app were to go behind a paywall, people would never know. Even though they might have been reading many articles from it before, they are not real readers. They are readers of AOL Editions--that is where their loyalty is.

Note: The only exception is when people follow you. But there is no way to track if a click is from a follower or not.

The real reader formula

In order to determine how many readers you have, you need to look at these four factors. How many people actually read an article? How many people read more than a certain number of articles? How many people come back more than a certain amount of times? And how many people come back because of you and not because they were referred by another person or site?

The number of readers are the number of people sharing these four characteristics. Add to that the number of followers you might have, who are likely to read your article via other formats (RSS readers, Flipboard etc.)

5% of these are then likely subscribe to your paywall or paygate.

The rest of your visitors are interesting too. You need to:

  • Focus on increasing the view-to-read rate, which can be done in many ways.
  • Focus on getting people to read more articles and come back often, which you can only do by providing relevant value for the individual (forget about page views--it increases clicks, not reads.)
  • And most importantly, you have to make sure people remember you because of you. The most effective way to do that is to get people to follow you.

This also shows that turning people into readers is a lengthy process. A big problem with the closed paywalls is that they cut the nonreader. You have nothing to work with.

It takes a very long time to turn a visitor into a reader. It takes even longer to turn a reader into a subscriber. first you have to build up your brand in people's minds (by creating value.) Nobody subscribe just because they see a "Subscribe Now" button.

Note: I am still trying to measure how many real readers I have. I know how many people read an article (28%), but I do not yet know how often they do it.

The bottom line is that it is very easy to get traffic on the web. The New York Times have 33 million visitors per month and that sound really impressive. But the fact is that most of their traffic is not worth anything. The analytics we have today is generally misleading.

We need to develop drastically different forms of analytics in the future. One that can measure real readers, people to readers rate, view to read rate, and readers to subscribers rate.

We also need tools that can look the other way. What are people doing with out content? How is it being shared, and why are people sharing it? Especially considering that social sharing is a fantastic way to get more real readers.

Shared By Plus Subscriber

Matthew Neale

This is Baekdal Plus content. It is shared with you for free by a member. Please reshare it.

Share on

Thomas Baekdal

Thomas Baekdal

Founder of Baekdal, author, writer, strategic consultant, and new media advocate.


Check out my book: THE SHIFT - from print to digital and beyond? Free for Baekdal Plus subscribers, $8.79 on Amazon.

There is always more...