Recently, I looked at how respondents to our Content Marketing Survey Report are measuring the results of their content marketing efforts.
Following on from this, I thought I’d provide some insight into the content marketing objectives for the blog, the metrics we look at to measure success, and the lessons we learn from them.
Why we blog
Our audience is the digital marketing community (if I can call it that), which encompasses a range of roles, areas of expertise and levels of knowledge.
Econsultancy’s brand is built on helping marketers understand and make use of the web, thereby making their jobs easier. For this, we have a range of great reports, training, events, and the blog.
Another important factor for Econsultancy is to take seemingly complex issues and present them in plain English. We hate unnecessary jargon, and you won’t find us leveraging anything. Not if I can help it anyway…
The blog is about getting the brand known and providing articles which digital marketers will find useful, which will create informed debate and perhaps inspire people to investigate the site further and check out our splendid reports, events and training.
In that sense, we have been content marketing ever since the blog was launched more than six years ago, even if we haven’t always called it that.
The quality of the content is the prime consideration. While we do sell ads on the blog, this is not necessarily the primary source of revenue for us. Hey, we use a bit of linkbait now and then, and we know the value of lists online, but it’s not all about the pageview numbers.
Content marketing metrics used by publishers
Our Content Marketing Survey Report found that the most common metrics 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.
Which metrics do you use to measure performance and engagement with your web property or web properties?
What do we measure?
We look at most of the metrics covered in the chart above, though I would add SEO and linkbuilding to that list.
Traffic metrics help us to identify hot topics and issues that are interesting to our readers. One such example from this year is the EU Cookie Law. One article we published in March, a few months before the ICO ‘deadline’, attracted tons of page views and lots of comments.
This brought the scale of the problem home to us, as well as the fact that, despite the ICO’s own guidance documents, people really didn’t know what to do about the legislation.
So, we produced a report providing guidance and suggested solutions for compiiance, we interviewed the ICO, conducted surveys, and created useful blog content on the law, all of which hopefully helped businesses to decide what to do (or what not to do).
We can also use the blog as an early warning system, as with the Cookie Law example I mentioned above. If a particular post gets more views, comments and shares than you would expect, then this tells you that readers are concerned about a particular issue.
If that’s the case, then we can write more on the issue, look to produce a report, or make sure it receives some coverage at our events.
One of the most popular posts earlier in the year was this on the number of users Zeebox gained during the screening of Dancing on Ice.
You can see the spike here. Just look at it:
It was published on the Monday after the Sunday night show, and caused a traffic spike that morning, but check out the bounce rate:
Clearly, people were more interested in celebrity gossip than digital marketing. Not the kind of traffic we’re looking for.
On the other hand, the aforementioned post on the cookie law attracted a decent amount of traffic, but also kept people on the page for longer. Thanks to links to related reports and content, the exit rate is far lower.
I can also tell by looking deeper that it sent a good number of users to the report page on cookie compliance. An important metric for us.
Page views are useful, but only in context, and in relation to other metrics.
For us, an article with 100,000 page views is great, but we’d rather have a tenth of that if it was a post that led people to stay on the site longer, and download a report or two.
We share these stats with our writers, as I think it’s important for writers to have access to the various metrics for their posts. If nothing else, they’re curious to know how many people read or tweeted their post, but they should also be learning from this.
I would never criticise someone if they wrote an excellent post that just didn’t happen to get the traffic it deserved, but it is important to learn from what works well. Also, I also think a little competition doesn’t hurt.
We have experimented on the blog from time to time. Some things work and some don’t, but the important thing is to learn the lessons.
For example, we did try to focus more on news at the end of last year. On reflection, this didn’t really work for us. Perhaps we’re not seen as a source of news, or perhaps our audience prefers to see more analytical insight from Econsultancy.
We have since changed tack and, ‘non-news’ works much better for us than news did. The stats showed that, when we changed the focus away from news, the traffic went up and the bounce rates dropped.
One metric which wasn’t mentioned in the content marketing survey was links generated by content. This isn’t necessarily something we measure religiously, but it’s good to see which posts are generating links and what we can learn from this.
It isn’t necessarily a question of popularity either. Some of our more popular posts have generated relatively few links, while the reserve is often the case.
For example, this post on the 2009 Innovation Awards winners has generated 2,440 inbound links, yet had a relatively modest number of page views. Shows the value of ego-bait.
This post, on 25 reasons why I’ll leave your website in 10 seconds many more page views than the awards announcement, more than 2,000 tweets, 190 comments, yet this produced 383 links.
So what’s the secret of a ‘link-able’ post?
These three posts all had 1,500+ links:
It’s not about traffic. The social media stats post did attract a lot of traffic, but the other two a modest amount.
I think the common theme is that they were all timely and useful posts.
The social stats post is from 2010 when people were desperate for information on social media marketing, Dave Wieneke’s excellent Pinterest post (one of a series of three) provided information on Pinterest when people needed it, while Matthew Curry’s round up of the issues from his Digital Cream roundtable provided useful information for people who couldn’t be at the event.
Keeping people on site
While more than half of our total traffic comes to the blog, and we hope that people stick around for a while. One way to do this is to provide recommendations for other content, based on the page or article they visited initially
Of the publishers surveyed in our Content Marketing Report, 69% said they use internal recommendations to provide inspiration for where to go next.
50% are using third party technology for this, 35% in-house solutions, and 15% are providing manual recommendations.
We use a mixture of these at Econsultancy. We are also big on a strong internal linking strategy which not only provides relevant articles for people to view, but also allows us to use anchor text wisely and helps us rank for target keywords.
We also have the most popular / most commented post table which you can see on the right of this page, with the aim of providing further inspiration.
In addition, we use automated recommendations for further reading which are provided by Idio.
Bounce rates is one indicator of whether we are encouraging people to stick around. High bounce rates, such as on the Dancing on Ice post mentioned above, tell us that very few visitors decided to explore further.
Other posts have relatively low bounce rates, such as this one, which is a compilation of useful posts on Google Analytics. It’s basically a bunch of links, but you can see how giving users ideas for what to read next can keep them on site.
I also look at the navigation summary in Google Analytics, which is great for seeing where certain posts are leading users.
In this example, a post on EU cookie law compliance has generated 811 pageviews of the related best practice report, as well as related blog posts.
This tells me I’ve done something right with the post. It wasn’t a sales pitch for the report, it was an article which works as a stand-alone and provided useful advice for readers.
What it did do is demonstrate to people looking for advice on the topic that we knew what we were talking about, and made a good proportion of readers check out the report itself.
Econsultancy generates a decent amount of traffic via social media, and Twitter especially. Social media, while it is rarely the last click before a conversion, does very well indeed when you look at assisted conversions.
We have a Social Media Manager who is excellent at analysing our social traffic and optimising our social accounts for the best results, but the blog team hopefully helps him out.
First of all, we try and produce content that is worthy of sharing on social sites, but that isn’t always enough. You have to be smart in the way you promote your posts.
As Stephen Croome (@firstconversion) pointed out on Twitter last week, most of the most popular posts listed in the sidebar started with numbers.
In fact, they all started with numbers. We just fooled him by typing the number five…
It isn’t always so, but the reason we use lists is to make this content more appealing to people. It can make it easier to digest, it can provide a way to quickly get up to speed on a given subject, and it makes the article that bit more shareable.
In fact, we even have a list post which shows why list posts work…
This is part of the tactics we use to make our content more appealing and easy to share. We can write very descriptive, almost academic titles describing post, but what’s the point if no-one reads them?
It’s not just about lists. If you want people to be able to share your posts on Twitter for example, then they need to be short enough to allow room for a retweet and a comment.
The need to make things easy to share is the reason for the 65 character headline rule. We don’t always follow this religiously, but it’s a good guide. It also applies to email subject lines or Google search listings.
To take Twitter as an example, there are a number of things we can do to make it more likely that our posts will be retweeted.
Timing of posts
When we publish a post can make a big difference to its success. Twitter has a lot to do with this (and more on that further down) but, if we publish at an optimum time and day, push out via Twitter, this can give a post legs and ensure that it does as well as it can.
So, if I had a post I really wanted to push, i’d publish it mid to late morning on a Tuesday. This is what the stats tell me. I think it’s because this is when our UK audience is switched on (they tend to be less so in the afternoons, and traffic tails off on Fridays).
So, a good, shareable post published at this time will ping around Twitter and elsewhere for an hour or two and, hopefully is still there when the US comes online, giving it a second wind.
There’s no formula for success here, but definitely pays to experiment with publish times. It can make a big difference.
Timing of tweets
One way to maximise retweets is timing. The map below shows when our followers are most active.
Those black dots represent 23% of our top 5,000 followers. That is, the ones with the biggest networks (A couple of these are brands obviously).
This means (and we haven’t actually been doing this for long) we can target tweets when the maximum number of people are likely to see them. In this case, mid-afternoon UK time seems to be the sweet spot.
The blog enables Econsultancy to rank well for terms that are relevant to our content and services, and that drive traffic to the site. Roughly 75% of our unique visitors and just over half of total page views come to the blog.
We can measure our search traffic, which accounts for roughly 35% of all visits. We’d love to look at the key search terms that send people to the site, though this is becoming more difficult, thanks to encrypted search:
We will target search keywords connected with our content, and this enables us to have a strong search footprint for these issues. For example, we published several articles on the EU cookie law and as a result, we occupied high positions in Google for terms related to the issue.
I’m not totally convinced by the Queries section on Google Analytics, as the numbers are suspiciously round, but here they do give an indication of our search visibility on cookie-related matters.
It shows the average position for the top ten terms related to the law, between April and June this year.
What do we do with this information?
Gathering this data is all very well, but it’s no use unless you learn from it and use it to improve what we do.
To summarise, we use these metrics to:
- Let our writers know what works.
- Identify issues of concern for our readership.
- Improve timings of posts and tweets.
- Learn more about what works for us.
- Find ways to keep people on the site for longer.
- Improve our search rankings.
I’d love to read your views on the most valuable metrics that you use to measure your blogging / content marketing efforts. What works best for you?