I was reading this article on paidcontent over the weekend, which points out the value of analytics to publishers, but only if they are using the right metrics. 

The key point was the danger in focusing on pageviews, as this doesn't necessarily help to build the kind of audience that publishers and their advertisers need. 

I would agree with that, and though pageviews are not insignificant, there are many more useful metrics for publishers to view. 

In this post, I'll attempt to answer the question by sharing some of the ways I use Google Analytics for this blog, while this post presents 10 shortcuts to Google Analytics reports and dashboards for publishers, bloggers and content marketers

Content metrics used by publishers

Our Content Marketing Survey Report found a focus on the page view among respondents. The most common metrics for publishers are unique visitors (88%), page views per visitor (76%) and page views (71%).

Only a third (33%) stated that ad clicks were used as a metric.

Q: Which metrics do you use to measure performance and engagement with your web property or web properties?


What can pageviews tell us?

The obvious thing is the popularity of various types of content, which can be very useful. In our case, it can help us to identify what our readers see as hot topics / areas of concern. 

For example, posts published on Pinterest a year or so ago (when the platform was new) were very popular, which told us that people wanted to know how to use it for marketing. 

This may prompt us to delve deeper into a subject on the blog, or perhaps to produce reports or incorporate this into training and events. In the case of Pinterest, we have written a gazillion blog posts and a best practice guide

It's also important for advertisers. We don't rely on advertising as much as some other publishers may do, as we have various revenue streams, but we do have ads and, in this case, pageviews do matter. 

I think the danger explored in the paidcontent article is the focus on pageviews at the expense of metrics which can actually tell you a lot more about your audience. 

Then there's the risk of putting pageviews ahead of decent content and going the Daily Mail route of producing tons of articles on Miley Cyrus:

Measuring engagement

Our aim is to keep people on site for as long as we can, browsing blog content, and hopefully exploring the rest of the site to find useful reports, training courses or events. 

To this end, I look at how effective different articles are for keeping the visitor on site and sending them elsewhere. To achieve this we try to give people ideas for where to go next, with internal links, mentions of reports at the foot of articles, as well as list of tags and most visited / most commented articles. 

To measure this, I pay close attention to metrics like bounce rates and average time on page. In the stats below, the post on blogger relations seems to have hit the sweet spot in terms of time on page. 

Another metric to look at is the number of pages per visit, to see whether people are browsing around after reading the first article. If you have a good proportion of visitors doing this, it suggests a segment of loyal and engaged readers. 

For this, I use a custom segment (which you can find in the previous link) that looks at visits where people view more than three pages.

The same can be done for less or more than this number, or it can be combined with dimensions such as traffic sources, or the location of the visitor. For example, I can use this to compare the level of engagement of UK and US visitors (the former are more engaged according to this). 

Monetising content

While the blog is there to inform and entertain, we do also need to build awareness of the other areas of the site/business. We want to promote awareness of our subscription packages, report content and events. 

So, we do produce posts which are designed to promote our report content, but they should also be useful in their own right so that, even if readers don't go on to download the full report, they can take something useful from the post. 

First and foremost the content has to be good. In a way, we're showing a bit of leg to encourage people to explore further...

I do look at how successful these sorts of posts are at promoting our paid content, so I measure how many visitors are referred to report pages with a handy advanced segment. 

In this shot, it shows which posts sent viewers to report pages, and I can delve deeper into this to see if visitors downloaded reports or bought subscriptions after viewing a blog post. 


Setting goals

We don't obsess over pageviews, but we do set targets for writers, and these do include number of impressions. These are achievable targets given the typical traffic we receive on the blog. 

This is more about encouraging writers to use analytics to see what works and what doesn't, and that their content is popular with readers. Also, a little bit of competition between writers doesn't hurt. 

For this one, I use a custom report created by Ben Barass, our resident analytics guru, which uses a custom variable so we can view stats by author.

This allows us to see pageviews by each author, which of their posts are most popular, which have the lowest bounce rates etc, while the page value is a measure of how many conversions followed posts by a given author. 

We can also look deeper into the conversions that come from authors' posts, and look to see what lessons can be learned.

For example, we can look at things like placement of links in posts, types of post that 'inspire' people to join up or download a report and use this to inform the way we write and present future articles. 

Measuring 'evergreen' content

Evergreen is my favourite type of content, and it's what we try to do as much as we can. News can be found in many different places online, and it's often basically the same article with subtle differences. 

This is why it's so hard for news sites to attract paying customers, and why they need extra hooks such as the Premier league goals app offered by The Sun, but that's a topic for another post. 

News articles tend to be popular for a day or so, then this tails off very quickly, with perhaps just a trickle of views thereafter. We know because we tried news and it didn't really work so well for us. 

Our aim is to produce content that is valuable for our readers, and that will keep people coming back again and again. This means the article is useful beyond the first few days after publishing, and is also likely to produce better results in terms of SEO and promotion of our site. 

If an article is consistently popular, it also tells us that we have done something right, and we can learn from this. 

We manage to do this with reasonable success, so if you look at the top posts for August, you can bet that a good percentage weren't published in that month. 50% in this case:

To take one example on that list, Chris Lake's post on scrolling websites, this shows the value of evergreen content. It was published in March 2012, but still brings in around 5,000 visitors every month.  

So, for publishers, it's worth identifying these posts, as they may well be bringing in the most engaged visitors, meaning they may have greater potential for monetisation.

In our case, we try to ensure that they have some links to relevant reports and events, as well as to other posts. 

In the case of a large news site, whose visitor numbers are far higher than ours, these posts could be used for targeting contextual ads, or links to ecommerce areas of the site. 

Social traffic

This is something that our Social Media Manager Matt Owen looks at regularly and, as he explains, measuring social media via Google Analytics isn't always straightforward

For a quick overview, you can use a social media dashboard like this one, which you can download here (when logged in to GA). 

In summary

It's not all about the pageview, but the most important thing is that publishers are using analytics tools. There is much to be learned by studying the behaviour of visitors and how different types of articles perform. 

I would add that there is extra value in providing the whole writing team with access to analytics, and this is what we do at Econsultancy. The writers can see the various reports, custom reports, segments and dashboards I've mentioned here. We also provide analytics training if required. 

It makes them more aware of how their work is performing, so they can absorb the lessons themselves. 

If they write a post and it doesn't go so well, they know and can address any issues, or they can see how changes they make to style and layout affects things like bounce rates and time on page.

This has the added benefit that our writers become more clued up on the use of analytics, a vital skill for writers and marketers today. 

Have I missed any key metrics for publishers and others? Please let me know below. 

Graham Charlton

Published 10 September, 2013 by Graham Charlton

Graham Charlton is editor in chief at SaleCycle, and former editor at Econsultancy. Follow him on Twitter or connect via Linkedin.

2566 more posts from this author

You might be interested in

Comments (12)

Save or Cancel
dan barker

dan barker, E-Business Consultant at Dan Barker

thanks for the post, Graham.

Here are some vagueish thoughts on some of the bits I think are really important:


I v much like the point that setting targets can be a way to help people to pay attention to the data & the feedback it gives them, rather than necessarily being about the targets themselves.


I also like the "risk of putting pageviews ahead of decent content" point which is the big risk for many publishers. You only have to look at your 'most viewed content' report to see there is huge appetite for one particular type of post - great that you also publish other types of posts (like this) alongside. I suppose that really underlines the importance of having overall goals & objectives.


'Author' is a great example here of using other attributes to add nuance to the data. Adding other attributes (or dimensions) like 'content length', 'category', 'purpose' from the point of view of the content itself, and dimensions like new/returning, 'member type', (or even more detailed information as used in your 'project arachnid' thing: http://econsultancy.com/uk/labs/arachnid), and then things like 'shares', 'reads', 'comments', 'sign ups', (maybe) 'resulting ad revenue' from an action point of view too. Having said that, it's best to take it step by step & look at the most important one or two most useful first.

almost 5 years ago

Graham Charlton

Graham Charlton, Editor in Chief at SaleCycle

Thanks Dan - the targets thing has worked well for us, but I do try to emphasise that it isn't just about pageviews.

I find that using Google Analytics, even in my own amateur way, has had a lot of benefit for the blog. We're a small team, and analytics does help to give us an edge.

almost 5 years ago


Joseph Franklyn McElroy

I noticed you don't have any metrics for traffic by channel. That is, for particular posts is traffic from search, social, campaign, etc. This can be a very effective measure - certain channels will have different conversion patterns - so understanding what will drive channel traffic can be used strategically for conversion goals.

Overall a very good article!

almost 5 years ago


Ian L

Good post Graham, thanks for sharing!

Another thing I find useful for content / informational websites is scroll reach tracking using events in GA - this allows you to determine how far down people are scrolling on a page. There are a few articles out there that detail how to do this, check out Justin Cutroni's "Advanced Content Tracking" articles parts 1 & 2.

Also, couple of little nitpicks below.. ;)

Should dimensions replace metrics here?
"metrics such as traffic sources, or the location of the visitor"

In your first GA data table (where the dimension is page title), some questions could be raised about the inclusion of the 2 goal-related metrics. The reason I say this is due to Avinash Kaushik's post "Align Hits, Sessions, Metrics, Dimensions" ('hit-level dimension' vs 'session-level metrics') on Occam's Razor - you will understand what I'm referring to if you have a read. Shuki Mann's comment and Avinash's reply under the main article addresses this point even further. The concept is thought-provoking and one that I'm still getting my head around in some instances!

almost 5 years ago

Graham Charlton

Graham Charlton, Editor in Chief at SaleCycle

@Ian L Thanks - corrected now. I am a bit of an amateur, and though I attempt to understand Avinish's posts, some of it goes way over my head ;)

almost 5 years ago

James Gurd

James Gurd, Owner at Digital JugglerSmall Business Multi-user

Hi Graham,

Great post and some good pointers for people getting to grips with using data to help them focus their efforts on what adds most value to the business.

The key point i always make is this: start by defining what your core goals, objectives and KPIs are before you even consider doing any analysis on content performance. By tying the data analysis back to strategic goals, you're measuring against your success criteria, not a metric that may be useful but won't drive the right decisions.

The Daily Mail model was actually incredibly successful because initially it was about monetising content through ad revenue - so the more impressions, the higher the revenue. It may have led to a dirge of soulless content but heh, it paid the bills and I've seen some of the data. I'm not saying that's the right focus or it will endure but it achieved the goal.


almost 5 years ago


Matt Lovell, Head of Customer Data, Insight & Analytics at Eurostar International Ltd.

Really interesting article and some great comments above.

One additional thing that definitely worth considering is the issue with tracking time on pages and the option Google offers in the way of their Event Tracking API to correct this by capturing an event every 10 seconds that someone spends on a page. By implementing this (there are numerous articles and blogs on how to do this) there are two major advantages:-

* It actually limits bounce rates to anyone who truly bounces from the site (i.e. arrives and leaves straight away as they are no longer a bounce if they are there for greater than 10 seconds) - A separate goal can always be created to show anyone who only views a single page if you want to be able to see both metrics

* The time on page is much more accurate irrespective of whether it is the first or last page they view on the website. This means you can start to look at people who will have read the entire article versus those skim reading or simply bypassing the page all together

The only issue that is occasionally hit is that customers who leave their browsers option can seem to spend hours upon hours on a given page. This can be resolved with the help of a clever developer to either set it to only continue to fire when the tab / window is being viewed or to stop capturing the event data after x. iterations...

almost 5 years ago


Tudor P.

Hello Graham,

Another thing to look at would be visitor flow from page to page - Analytics has a very good visual representation of this. If you find out where most of your visitors are dropping off it can tell you which pages need to be improved OR if visitors on those pages have trouble finding their way to the next step which ultimately leads to the conversion page.

almost 5 years ago

Graham Charlton

Graham Charlton, Editor in Chief at SaleCycle

Thanks Tudor, I'll try that.

almost 5 years ago


Mark Hansen, Founder and President at Megalytic

Nice summary Graham! At Megalytic, we've got a lot of clients that like to measure the total size of their engaged audience. You can get that information out of Google Analytics, but it takes some work exporting and manipulating data. Here are the details: http://blog.megalytic.com/measuring-engagement

Disclaimer: that blog post also shows how we automate the process in Megalytic.

over 4 years ago


Matt Lovell, Head of Customer Data, Insight & Analytics at Eurostar International Ltd.

@ Mark

That's an interesting approach however the difficulty I always hit with the basic GA engagement data such as bounce rates and average durations is that given any timings don't take into account the last page viewed and that if someone moves away from the page (to another tab or program) they are still tracked as viewing the site, then those figures aren't overly accurate...

over 4 years ago


Mark Hansen, Founder and President at Megalytic


Yes, that's true - the basic metrics are not super-accurate. Have you found any better way to measure, say - avg visit duration?

I find, however, that they don't need to be. We look at them directionally over time - are they moving up or down?, how do they look for visitors from cohort A vs B? Used like this, and looking at the changes in the overall size of the engaged user audience, we find to provide good insight.

over 4 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.