New Website Measurement System Just A Little Less Useless Than Previous One
from the keeping-score dept
In a bid to improve the relevance of its ratings, Nielsen/NetRatings has announced that it will no longer use page views as its primary metric for measuring the popularity of websites. Instead, it will focus on the amount of time that users spend on the site. Obviously, there have been a lot of problems with the current system, as the use of page views grossly inflates the popularity of some sites, like MySpace, while penalizing sites that aren't refreshed or reloaded as often. As the above article notes, the new system will give YouTube a boost, but will ding Google's main site, which isn't designed to keep users around. Of course, therein lies the flaw with this new measurement system. Google is incredibly profitable and successful, precisely because it does a good job of whisking users away to other sites, either through ads or its search results. The idea of penalizing it because users don't spend a lot of time on the site is absurd. When it comes to TV shows, it may make sense to adopt a uniform measurement system, because all TV shows have the same purpose: to sell ads. Websites, however, have a variety of different business models, so trying to define a standard metric of success is going to prove impossible. Ultimately, the most meaningful measure of a site or service is its profitability, which, unlike page views or time spent, isn't so easily gamed.
Reader Comments (rss)
(Flattened / Threaded)
Why either-or? by Nasty Old Geezer on Jul 10th, 2007 @ 9:11am
If I am evaluating a site (or anything else) a single metric will rarely if ever give me a realistic picture of success.
Humans have excess enthusaism for "the best" without saying "for what purpose".
(reply to this comment) (link to this comment)
what about visitors? by Ryan on Jul 10th, 2007 @ 9:30am
Why not just use visitors? Whenever I've tried to buy or sell a website, people only want to know two things:
Monthly unique visitors
and
Monthly Net Profit.
(reply to this comment) (link to this comment)
by Joel Coehoorn on Jul 10th, 2007 @ 9:30am
Neilson ratings are used by advertisers to determine what rates they will pay to what sites. If your site isn't interested in selling ads, or if you use a business model that cuts Nielson out by selling limited text-only ads like Google, than it doesn't really matter how Nielson rates you.
Advertisers seem to be hung up on the idea that internet ads need to turn into an instant sale. Certainly the internet makes that possible in many circumstances, but I think this is a bad idea. An internet ad needs to be viewed more like a billboard placed next to a freeway, where the goal is often to build brand recognition or just plant a seed for a later sale. Anything more aggressive and you just annoy people. If someone clicks an ad then great- maybe pay bonuses for 'clicks' and 'actions', but a single view like a motorist driving past a billboard on the interstate should be the baseline.
(reply to this comment) (link to this comment)
Absurd? by ChurchHatesTucker on Jul 10th, 2007 @ 9:35am
"Google is incredibly profitable and successful, precisely because it does a good job of whisking users away to other sites, either through ads or its search results. The idea of penalizing it because users don't spend a lot of time on the site is absurd."
Actually, that makes a lot of sense. If a user breezes through a page, an ad is less likely to catch someones attention.
Google *is* a bit different, though, in that the ad may be exactly what the person was looking for, but if it's not they may not even notice it. A better metric would be click-through rates, or click-through-and-buy rates, of course.
(reply to this comment) (link to this comment)
Not so absurd by Tony on Jul 10th, 2007 @ 9:56am
Well if they are going o base the ratings on time, I think it could just work.
For example x amout of users spend x amout of time at google versus yahoo...
or x amout of users spend x amout of time at myspace versus facebook.
It obviosly makes no sense to compare search engine to social network as it is said in the article. However with time you can get a really accurate system for rating groups of websites instead of every website.
(reply to this comment) (link to this comment)
damn by Tony on Jul 10th, 2007 @ 9:58am
i misspelled 'to' by the way :(
(reply to this comment) (link to this comment)
by CodeMaster2 on Jul 10th, 2007 @ 10:20am
Here's an absolutely crazy idea for measuring websites:
Unique Visitors.
(reply to this comment) (link to this comment)
LOL by Buzz on Jul 10th, 2007 @ 10:37am
They should rank sites based on who has the smallest HTML file sizes. ^_^
(reply to this comment) (link to this comment)
by Matt Bennett on Jul 10th, 2007 @ 10:44am
Joe, you speak as if the "measuring the worth" of a site via Nielsen ratings is somewhat important. It's not. Nielsen ratings are a tool for advertisers to judge how ell their money is spent on none-click through ads, nothing else. Citizens bank homepage does not care, it is there to serve bank customers, not place ads. Google does not care, they have a bidding system, and their users can directly measure effectiveness through the quality of leads they get thru click-thrus, the only thing they pay for.
(reply to this comment) (link to this comment)
Re: LOL by Mr. Vage on Jul 10th, 2007 @ 10:53am
Then we'd see more of those crappy all Flash sites. All that's in the HTML is an embed tag.
(reply to this comment) (link to this comment)
I call BS on this one too by OKVol on Jul 10th, 2007 @ 11:47am
Page views - too easily spoofed.
Time viewing site? - HTML is connectionless, Google downloads quickly being lean and looks bad. And, if you are still using dial-up, you spend forever waiting on YouTube. How is Neilson going to see that I may be waiting on a page to load in one tab, downloading a second, and looking at a third?
Neilson is getting to be more and more irrelevant. They seem to be spending most of their time proving that they are relavent.
(reply to this comment) (link to this comment)
And furthermore by OKVol on Jul 10th, 2007 @ 11:49am
Unique visitors? You better use cookies and hope folks aren't security paranoid. We have 200+ users behind one IP address here.
(reply to this comment) (link to this comment)
by Anonymous Coward on Jul 10th, 2007 @ 12:56pm
I measure business success by bottom-line, net profit.
Anything else is like measuring your cock by starting at your knee. The numbers sound great, but it ain't gonna rock anyone's world.
(reply to this comment) (link to this comment)
And still more by OKVol on Jul 10th, 2007 @ 12:59pm
We use a proxy here that caches the web sites, and updates the cache automatically for many sites that are frequently hit. So, 10 people could look like one, or the cache could be updating when no one is hitting the site.
And, so many sites are coding themselves with specific cookies for tracking that my hit rates on the cache are going down.
(reply to this comment) (link to this comment)
WHY NOT OFFER BOTH ... by JOHN DOE on Jul 10th, 2007 @ 5:20pm
Serious, why would I use a company that only offers one why smaller better companies like hitwise offer more.....
(reply to this comment) (link to this comment)
Two points by Ambrose on Jul 11th, 2007 @ 12:53am
If they're going to measure sites by time spent there, they need two completely opposing scales: for search engine websites, LESS time means you're better. For most other sites it's the other way around.
And you know what? There is still such a thing as public service broadcasting! Maybe not so much in the USA but it makes me sad to see anyone say "all TV shows have the same purpose: to sell ads". They do have ratings for the BBC in the UK, strangely, despite there being no ads at all on the BBC...
(reply to this comment) (link to this comment)
Add Your Comment