Following my post the other week regarding online readership, sparked by some aggressive sales tactics by one of our competitors, I got talking to Andrew Smith of Escherman and we agreed to jointly carry out a more extensive piece of analysis looking at 50 different online news sites.
We selected ten sites each from the following areas – UK Nationals, Business, Marketing, Technology and Consumer. There was no particular selection process, just an attempt to have a reasonably representative sample and we both hasten to state that this is a relatively limited exercise which should therefore be taken with at least a pinch of salt. Particularly since indexed urls have been used as a proxy for content as this can be impacted significantly by site structure (as stated in slide 21), with some sites having sub domains and/or a more complex content directory structure.
However at the same time with the data that is readily available we think it provides some (arguably) valuable food for thought. So after several hours of research, number crunching and graph generation here are the results (I suggest you view in full screen mode unless you have excellent eyesight):
Site data is sourced from Google for the number of indexed urls via the “site:domain” command and from AdPlanner for the traffic data.
We are effectively looking at three areas:
Readership per article – average numbers of UK page views per Google indexed url per month. Where indexed url is a proxy for the number of likely visited pieces of content.
Engagement – time spent per page to indicate how long a reader is likely to be spending reading that content when they get there.
UK Relevance – what proportion of the sites readers as a whole come from the UK and would therefore be likely to be relevant if you were trying to reach a UK audience.
Andrew has provided his take on the results from a PR perspective
here. For my part the highlights are:
Readership
Unsurprisingly readership per article is much higher for UK Nationals and Consumer than the sector specific publications. However within the performance of UK Nationals and Consumer a handful of sites stood out for having particularly high UK traffic per article being, in order,
News of the World,
Heat,
The Sun,
The Mail,
Closer Online and
Marie Claire.
Interestingly though if you remove these six high scoring sites from the samples then the sector specific sites still achieve, on average, between 30-60% of the readership per article of the remaining UK Nationals or Consumer titles.
Engagement
This difference was further underlined with the Business and Technology sites achieving on average approximately twice the time spent as Consumer sites. Evidence for both more in depth content and greater engagement, which doesn’t seem surprising.
Relevance
There are significant differences within each group with regards to proportion of UK traffic. Within UK Nationals the tabloids are generally between 60-75% UK based with the qualities between 30-55%; the FT having the lowest UK traffic proportion with 31%.
Within Business titles the vast majority of sites are UK focussed and because of this their audiences are predominantly UK based also. The exception being
The Economist with only 7% of its audience being from the UK according to AdPlanner. Interestingly this seems to reflect the broad geographical interest of its content with even the US only accounting for just over a third of its traffic.
Its a similar picture within the Marketing titles with the vast majority of traffic to the sites selected being UK based. The marketing publication with the lowest UK proportion is
Econsultancy with 57% from the UK. This in in part due to around 20% being from the US which seems consistent with their having a US presence.
Finally Technology and Consumer titles have quite varied levels of UK traffic with sites such as
Techcrunch Europe and
Vogue.co.uk (clearly having the potential for interest from outside the UK) having lower proportions of UK traffic at around 20-40% compared to sites such as
T3 and Heat which are between 75-100%.
Conclusions
Though limited the analysis provides evidence for savvy PR people who already know that it is important to ensure that you understand the publications they engage with and their potential to actually reach the readers and communities most relevant to them and their clients and not be blinded by big traffic numbers.
There are many other points that could be drawn out of the results and we would love to get feedback from people on anything they observe or suggestions as to how to refine the analysis and improve the validity of the results.
Notes
The other publications analysed not mentioned above were:
UK Nationals
Business
Marketing
Technology
Consumer
‘engagement’, in digital/social media circles tends to mean people actually engaging with the content (rather than just consuming it).
So ‘engagement’ in digital display means the number of people who interacted with the ad (a subset of those that consumed it, which again is a subset of those that were exposed to it).
And in social media circles ‘engagement’ tends to come from the number of people who have bookmarked, tweeted about, commented on, linked to (etc..) the content.
So perhaps you should use the word ‘consumption’ (or similar) instead of ‘engagement’ to avoid confusion (and, let’s face it, justified criticism).
Hi Robin thanks for stopping by and “engaging†with the content
I take on board your points but I am not wholly in agreement with them.
Being engaged by something, such as a piece of news media content, is clearly as you would know better than I, a very difficult thing to measure in itself without asking the people who view it about their experience afterwards or without them providing evidence that they were interested in it.
The keyword being evidence. The things you mention as evidence of “engagementâ€, are only in themselves indicators not proof. I agree that a comment, such as yours, is strong evidence of engagement, but this is not always the case with others.
If I tweet a link to an article does that prove I was engaged by it? It could be that I just RT someone else because the article sounds like it might be of interest to my followers but don’t actually read or engage with it myself. With so many RTs often not incorporating any edits from the party RTing how can we know that the person was engaged?
The same could equally be true of bookmarking. I might bookmark an article to come back to read later and never do so. How can one be sure?
I would humbly argue that if engagement effectively means interest (as others more qualified than I have suggested) then spending a great amount of time on a particular web page is equally evidence that an article is of interest to someone. All things being equal if you’ve stayed with me until this point in my comment it will have taken you longer and arguably you are less likely to do that if you think my comment is not of interest to you.
Soon the one hand I agree. Is time spent *the* measure of engagement – no not in the slightest and perhaps we should be making that more clear. No more than our calculation of views per url is *the* measure of readership. On the other hand are tweets, bookmarks, links etc *the* measures of engagement – no, but like time spent they provide evidence.
But only through real in depth responses can we truly assess engagement and the issue there is the time and cost involved. Hence we all seek cost effective proxy evidence instead.
It would be interesting to expand the analysis to include other metrics such as tweets, links, bookmarks etc (perhaps using another free tool like Social Mention that anyone can access?) that provide further evidence of levels of engagement. If you were up for assisting with such an exercise that would be great.
Thanks again
Adam