One of the popular stories going around today is that of Nielsen/NetRatings (an online measurement service) scrapping the basing of their ratings being based upon page views and moving to tracking how long people spend on a site. This is as a result of the increased use of video and tools like Ajax which mean people are not refreshing pages all the time.
There’s been some interesting observations being made around the blogosphere and Richard MacManus gives a few observations as it relates to blogging:
“Blogs are a good case where ‘time spent’ is more meaningful than page views. Especially since the blogosphere is particularly prone to the ‘quantity over quality’ problem. It’s easy to pump out 20+ posts a day – and that tactic garners a lot of page views. But are those blogs actually writing for their readers, or writing to get page views? In other words, check the ‘time spent on site’ figures for those blogs and I think you’d find it is very low – because users click through, find nothing of value, and quickly leave. Is that good for advertisers on those sites? No it isn’t. So in the case of blogs, I’d argue that ‘time spent on site’ is a better measure than the easily gamed (or at least cynically exploited) page view model.”
It’s interesting that he says this because I’ve noticed in tracking my own blog’s statistics over the last six months that I’ve been looking less at the page views count and more at stats like bounce rate (how many people leave the site without surfing deeper into it) and time spent on blog.
Page views still are something I do like to build in that they are still related to income (many of the income streams still have an impression based focus) – but I think we’ll see more changes in what the ad networks are doing also. Google’s been moving more focus to ‘Cost per Action’ and I still think we’ll see some attempts at some sort of an ‘Cost per Time’ ad network – or at least an ad network that refreshes ads over time.
So page views are not completely dead for me – but they’re definitely less important than they once were.
What metrics do you look at? Are page views still important to you?
Page views are only important as they relate to ad views. For me, the most important metrics are things like del.icio.us bookmark counts, links from other sites, etc.
I’m of the opinion that page views are still a very useful measurement, as long as they are taken with several other metrics.
Time spent on site is a useless measurement unless you know how many people were there spending that time. By the same token, page views can be worthless if they are all a few seconds long.
All of the various analytics work best if they are taken together so that it’s possible to get the whole picture.
My favorite measurement is definitely click-mapping and visitor paths. I find that stuff fascinating.
– Mason
You make a good point. I think it’s still all about getting the traffic there, but the hope is that once they get there they will stick around. To that end, page views are still important, but only in as much as they drive potential readers (those who stick around more than a few seconds) to your site.
Here is a page view question that I have wondered about. Perhaps you, or someone else who reads here, could answer it. Let’s say I had 200 page views in a day and at the end of the day I have 50 spam messages in my Akismet filter. Were the viewers who created the 50 spam messages included in the 200 page views? (I’ve sort of been assuming that they were, but I really don’t know.)
I look at time spent on the blog itself and how many pages that particular person looked at…I want them to spend time and read the critiques and hopefully visit some of those reviewed.
Laura: depends on how you’re tracking them. If you’re using Analytics or Site Meter, no. If you’re using a log analyzer, then probably yes.
I have to say that the page views are important to me. Someone read through 42 pages/posts the other day and that was exciting for me. I think all the stats have their place.
I like to know a both the page views and time spent. The way I figure it the more information I have the better. From an advertising perspective if a site has a ton of hits but almost no time spent, the likelihood of my ad being seen is far less than a site with less hits and more time.
aha. Some synergy is going on here. Just two weeks ago I switched to performancing metrics. It seems to place a higher value on Visitor count. When I think about it, increasing visitor count is a good goal to shoot for.
So, if you use Analytics, think about the time spent, maybe pageviews is an criterial of php-refresh age, if you have some developement in Ajax you must think in time spent.
Of course, build a high number of readers first and then further growth of your blog/site is much easier to be done by modifications to increase time spent on site or decrease bounce rate.
The problem with “Time Spent Online” is that it needs to be applied VERY carefully. For example, the average for today on my blog is 3:35. However, this includes 21 referrals of 0:00 (i.e. bots/crawlers). If I took away those 21 from the 250 I recieved yesterday, my average time would probably be much higher (had 21 visitors who viewed on average 7:51, and 35 visitors who viewed on average 4:20). So to do a “time spent online”, selling to begin with, could be unfair to the publisher.
I will break my analysis in two areas the visitors arriving and visitors leaving.
I) Visitors arriving: Where they arrived from, where they clicked on my site and how long the stayed.
2) Visitors Leaving: Bounce rate is the key and exit page.
Vijay
I look at where they come from and the page views from that source in Google Analytics. So, if they come for a HD format war post, I ‘m sure they look at other post related to it. I rarely look at the overall page views.
I can see the internet turning into TV. Time spent on a site should only be part of a sites measurement of success.
Quality of traffic is king over quantity. But what if I put hypno-toad on my front page and visitors are hooked for hours at a time, will my site shoot up the rankings?
Page Views are still important, but perhaps their importance is starting to fade slightly. Time spent on site may be a more accurate judge of the blogs quality but not how much advertising revenue can be generated. I.e it could be aruged that a blog with fantastic content will have a lower adsense click through rate than a low quality blog, because the readers are interested in the content and not looking for another site to go to.
I think there really needs to be a combination of both. I know that the way I track many of the sites I read is through my rss reader. When a title or topic catches my eye I click, read, and then usually I’m done. I don’t surf any deeper and since I read fairly quickly don’t spend much time. I have found looking at my metrics that traffic from my rss subscribers tends more toward this than does referall traffic, or search traffic.
In fact today I read this article after clicking the rss link and then left. Later I decided, after thinking about what I had just done in relation to what I have noticed on my own site, to return and post this comment.
While I agree that I do tend to look at time spent, and bounce rate, I also look at page views as it relates to new unique visitors, and return visitors. I don’t think you could base a rating solely on time spent and totally ignoring pageviews.
later all and have a profitable productive blogging day
If I keep up with a blog then I don’t need to look at other pages on the site. And if the post is short then I’m not going to spend any more time there than to read the post.
If they are going to be looking at time in seconds then this might be effective — someone spending two seconds at a site is probably barely reading the title, let alone the post.
Otherwise I think this could penalize popular blogs where the posts are short and readers are familiar with the material.
I agree with Suzanne. I think it depends on the type of site. If my blog is successful and ppl keep up with it, I gotta think that my pageviews are going to be low as visitors will probably only visit 2 pages if that. On content sites however, I should expect things to be a bit as users will have to go through many pages to get the full story on a certain topic.
So for me, it depends on the kind of site.
It is interesting that Nielson is changing its ratings, but for most bloggers (or any websites) this is just another distraction?
You’d better be tracking what matters to your site’s goals. If you sell impression based ads, then don’t use ajax on your site. If you sell a product then, you will go broke with 10,000 visitors a day who all spend an hour each on your site, but none of them buy your widget.
If you make money from AdSense, then clicks is your metric, and so on.
Darren,
Would love to see what kind of numbers you have for time spent on site and bounce rate. I’m currently at 2:59 and 63%, but its always good to see what I should be striving for. Thanks!
Brian
I’m still a big fan of unique visitors. This gives you a better feel for your ‘reach’ in terms of the number of different sets of eyeballs that are looking at your stuff. From a monetization perspective, visitors that just hit a single page are often quite profitable… They come in from a search engine, find what they want, and then act on it (click an ad, follow an affiliate link, etc.). Of course, such visitors aren’t as available for CPM programs, but I don’t care as much about that as I earn far less from CPM programs than from other sorts of revenue streams.
I hardly pay attention to my pageviews anymore. I tend to focus much more on the amount of visitors my blog receives. The only time I give pageviews much attention is when I want to see the average amount of pages each visitor views.
Both the metrics have their own importance, but like Trent said… I count more on Del.icio.us and other social link backs
“Time spent on site is a useless measurement unless you know how many people were there spending that time.”
I (partially) agree with this comment. Until we know what people are doing during the time spent on a site, it’s hard to determine how valuable this metric is. I wouldn’t go so far as to say that it’s a useless metric, but I know that I’ll often open several pages in individual tabs, and then close those tabs as I read them.
If it takes me 10 or 15 minutes to finally get to the last tab I have open in my browser, it’s going to appear as though I spent a lot of time on that page, when, in fact, it’s entirely possible that I only spent 30 seconds actually viewing the page.
Like any other metric, it must be combined with others to provide real value. When looked at in isolation, “time spent on page” is no more useful than “number of page views” is.
For me, page views are a point of departure. I use them to see if they landed on my blog, how they arrived at my blog, and then look to see how much time they spent there.
Some of the search parameters they use through Google, only bring my page up because I used a specific phrase in my post, and I am surprised they still clicked on the site. However, they don’t stay long.
I’m very happy if they convert that first initial visit to a long term one where they keep coming back. That return visit number is growing…
Completely agree with Adam. He beat me to making the point about tabbed browsing and “dead” time when browsing sites, and I would like to reiterate it. Most browsers will keep tabs open till they get to them. Even worse, there are those tabs that often remain open indefinitely.
For example, when I start up my browser in the morning, I always check my email in my first tab, and open up the administrative areas of my blog and a couple of other sites I take care of in the others. Thrown into the mix might also be some other blogs and sites I like to keep track of on a daily basis. Now my browsing session usually involves me clicking on a mail in my inbox and then switching to another tab, this switching between tabs happens everytime I load a new page, because this way I am doing something else while all that loading is happening, and I find I use the time more efficiently this way.
As a result, my mail window might be open for 30 minutes to an hour. And a similar situation can arrise whenever I am going through a whole list of things on a site — like a forum, for example. This means my actual eye-ball time on the site would be something like 5 minutes while it would technically show that I had stayed on for 50 minutes. That is quite a large discrepancy.
What it comes down to is that time spent on a site is as fallible a measure of site popularity as page views. In fact you could even claim that page views is a more real measure in the sense that it measures an actually measurable quantity, i.e. – how many times a page was accessed on the server. Unless someone uses blatant and deliberate spoofing techniques such as scripted robots to hit a site, page views are a real measure of how many people arrived at a page. On the other hand there are many situations when time spent on a site can be the untruth even without any planned deceit. What happens if I keep a site on while I answer the door or go to the kitchen to make myself a cup of coffee?
Sure in a completely controlled environment where the viewer is almost a test subject in a lab (which might be the case for services such as Neilsen specifically), these sort of specific abnormalities and gaps in time measurements can be factored in and actually not measured, using advanced instruments or systems. But, when you’re talking of standard metrics methods involving a little code inserted into your site or server software, there is no way to measure time spent on a site accurately without drastic breeches of privacy for the user.
For that reason alone, I would say page views are not going anywhere.
I think what’s really important in analytics is that no metric can really be looked at in isolation. Time spent can be valuable to know, but it doesn’t account for the person who opens your page, gets a phone call, and when they get back to your page leave immediately. That person spent no time interacting with your site, but that’s not what your analytics is going to tell you.
Similarly with Page Views. Did someone view a lot of pages on your site because they enjoyed the site so much they needed more or because they couldn’t find what they were looking for?
Statistics are very useful, but you can prove anything you want with the same statistics if you look hard enough.
I think it’s good that people are looking at time spent, but it still needs to be seen in conjunction with what your other metrics are showing.
with my mates the bragging rights about who has a bigger readership has always come from unique visitors and the amount of time spent when they came. I think any intelligent advertiser should be asking for that metric. Unless the bouncing = converting. It would be great if their was a wordpress plug-in that showed average time spent by visitors. Does anyone know of one?
Okay, this is an encouraging post. I’ve been writing quality content for about a month and a half on my blog.
My total page views are small (I’m almost embarrassed to say) but I’m working at making the content top quality. I’m doing interviews with people who are knowledgable in their fields and can leave the reader with an important idea.
I have a small fan base but I can see I can get a core group of the same visitors back . . . now, I feel one or two really strong posts once a week or once every couple of weeks and those will get passed along . . . and my readership base will grow and strengthen.
i would agree with one of my friends here who said that sites with good content recieve fewer clicks. I have a good, informative travel blog myself and good page views with average time spent around 21 mins with over 8 page views per visitor on an average. Some have viewed over 25 pages and spent over 40 minutes (its from another country, so i know its not me). But i have hardly any clicks. That’s what happens if your content is good and engaging. If you just have some crap, visitors click the ads just to get out of your site. And that is exactly how you make money out of adsense. i haven’t tried it though. But from what I have seen, having about 8 – 10 blogs with three blocks of adsense in them and no content will get you the much needed clicks