Formerly, the Internet rankings company used page views as its metric for determining a web site’s popularity. It will still measure page views, but it will cease ranking web sites by them. Given the changes in technology, many feel that it’s about time the company made this change.
And then there are sites with streaming video. YouTube comes to mind, of course, but it’s only the most prominent among countless others. If you’re watching a video for a long time, a web site might fit several ads into your viewing time – but if you use the page view metric, you’ve still viewed only one page.
Online gaming sites have the same problem. I can play my favorite game on PopCap for an hour to unwind after work, and the frame surrounding the game may cycle through several ads. But guess what? That still counts as just one page view, if I understand the metric correctly.
Is it any wonder that Nielsen/NetRatings decided to change the metric for this online popularity contest? The old page view metric was no longer reflecting the reality of how web surfers used the web and engaged with their favorite web sites. The move seems to make good sense from a technological point of view, but it’s important to consider what effect it will have – and to remember that this metric may have its own flaws.
The differences between time spent and page view metrics may be significant for some categories of web sites, but not others. Let’s look at search engines, for example. Nielsen/NetRatings said the ratio of total minutes spent on Google Search versus Yahoo! Search was 3.3 to 1. Their page view ratio is 3.1 to one. That’s not terribly different.
But when you look at social networking sites, the numbers change considerably. Two of the most popular social networking sites are MySpace and YouTube. The difference in the way web surfers engage with the sites becomes clear when you look at the ratios. When you look at the time spent, you see that the ratio between MySpace and YouTube is 3.6 to 1. When you look at the page views, however, the ratio is downright huge: 10.4 to 1. What’s going on here?
You probably don’t need me to tell you. YouTube’s main focus is videos, and every time someone watches a video, that counts as only one page view for as long as the video is being watched. If you looked solely at the page views metric, you would assume that MySpace was far more popular than YouTube. But YouTube users spend an average of 46 seconds on each page, while MySpace users spend an average of 16 seconds on each page. Whose users are more engaged? Hold that thought; we’ll be coming back to it in the final section of this article.
Search engines actually take it on the chin with the new ranking system. Consider that Google drew the biggest unique audience for May of 2007, according to Nielsen/NetRatings, with 110.2 million users. But its raison d’etre is to help people find what they’re looking for and send them somewhere else, so they rank fifth in total minutes. Yahoo and MSN/Windows Live both did better, thanks to their more widely used instant messaging and email applications (yes, Google has both of those, but they aren’t as popular).
And which web site now holds the top spot under the total minutes metric? Why, AOL of course. No, I’m not kidding. Ironically, the site ranks sixth in number of page views. Its wildly popular instant messaging application now counts toward the “total minutes” metric used by Nielsen/NetRatings – and I don’t know about you, but I use AIM for work purposes. I’m far from alone. You do the math – or, more precisely, you don’t need to do the math since Nielsen/NetRatings just did it for us.
We have to ask what this means for advertisers. Up to now, page views have been a very important metric for many if not most sites, in part because advertisers often pay for ads based on a cost per thousand views (CPM) rate. With such a public statement about the change in the importance of the page view, marketers may need to reconsider how much of their advertising dollars they choose to spend with any particular web site.
Advertisers – and advertising-supported web sites – will not abandon the page view completely in favor of the new metric. If they’re smart, they’ll consider both of those metrics along with others, such as past campaign data, when weighing where to invest their marketing money. Jeff Lanctot, senior vice president for global media at Avenue A/Razorfish, reflects on this matter. “I think what we’ve overlooked is the importance of third-party ad serving, because as you look over time what’s most influential is the rich historical data that digital provides,” he said.
Other companies that measure the popularity of web sites seem to get this. For example, comScore uses monthly unique visitors as an important part of its web site traffic ratings. But it also considers other metrics, such as time spent. Last year, in a nod to the increasing popularity of sites such as YouTube, it began collecting data on video streams.
In theory, it could be argued that web sites with a higher “time spent” metric may be able to charge more for ads on their sites. For those who can remember that far back, this new metric is a reflection of the old efforts in the early days of the web to make your web site “sticky” by finding ways to keep your visitors on your site, whether it was by engaging them with interesting things to do or – sometimes – by making it difficult to leave. But you can’t really tell just how engaged a site’s visitors are unless you have page views as well, to give you some idea of how long per page each visitor spends on the site (assuming that’s a valid measure of engagement). So once again we have to ask: how will this affect advertising? And there are bigger question with the new metric.
Reactions in the industry to Nielsen/NetRating’s move have been mixed. While most agree with the idea that page views are becoming outdated as a metric of a web site’s popularity, some think the change may be a little premature. Sheryl Draizen, general manager of the Interactive Advertising Bureau, thinks caution is warranted because “I don’t think we’ve done enough work to come up with what is the replacement for the page view…It may be that time spent is the right one, but I don’t know if we know that yet.”
The difference you need to keep in mind is that, while time spent can measure engagement with a web site, so can page views. Page views specifically measure user-initiated actions (clicks on a page). If you’re measuring the time someone spends on a site, how do you know they haven’t just left the site open while they go to the restroom, do chores, leave to run some errands, or do any of a number of things that don’t translate to site involvement? Could it be that minutes spent on a website don’t translate as smoothly into engaged users as some would like us to believe?
The problem multiplies when you consider the effects of increasing broadband access, tabbed browsing, small desktop applications such as instant messengers and stock ticker feeds, and larger high-resolution monitors. As a result, it’s easy to have lots of different web pages and applications all open on your computer at the same time. As a writer, I’ve done that any number of times while researching an article, and I know plenty of people who multitask more than I do. Is it really valid to count all of that?
In the end, it may not truly matter that much. Marketers must decide for themselves if a particular site is a good advertising investment, and the total minutes spent at the site ranks as simply one piece of information. No one is going to stop advertising with Google because its standing with Nielsen/NetRatings has slipped; the search engine clearly knows how to bring in money for itself and its advertisers. As the landscape changes and becomes more complicated, perhaps we’ll all benefit from realizing how important it is to use more than one metric, and – even in an industry that reinvents itself all the time – to take a quick look at historical performances before letting any one piece of information draw us up short.