Big Brands Want Real Numbers

by Joey deVilla on October 30, 2006

As I've mentioned before, the online advertising world, even though it's leaps and bounds more accountable than the offline world, has a few challenges to address:

Internet companies have had great success selling advertising space, in part because the effectiveness of those ads is supposedly so easily measured. But marketers, even as they continue to push more of their ad budgets online, are starting to ask for better proof.

A group of large companies, including Kimberly-Clark, Colgate-Palmolive and Ford Motor have said that by the middle of 2007, they will demand that online publishers hire auditors to check their ad and viewer counts. And analysts say they believe that online ad growth over the long haul will depend on the eagerness of large advertisers like these to shift more dollars online.

Meanwhile, reacting to advertiser questions, online companies like Google, Yahoo and LookSmart have begun to meet with industry groups to answer basic questions on how click-based advertising works.

Nice of the Times to list LookSmart along with Google and Yahoo! One of these things/Is not like the others/One of these things/Just doesn't belong…anyway.

I'm not surprised that large brands want audited figures from the publishers (like, say, nytimes.com) and ad networks (like Google or DoubleClick). The trouble is, the state of the art in audience measurement relies on cobbling together IP addresses, logins and user accounts, user agent strings, and cookies to identify a unique visitor, all of which are difficult to actually correlate definitively to a single person. The unstable and temporary nature of the composition of a unique visitor is, after all, what makes it so difficult to clearly show clickfraud. Oddly enough, trying to pin down your audience by watching what happens on your website, simple as that sounds, may not be the best way to go.

With rigorous sample control, the panel-based services like comScore Media Metrix, or Nielsen/NetRatings should offer a far more accurate picture of traffic to a given network or site, and provide baselines for clickfraud detection, an issue that Fred Wilson (responding to a TechCrunch post) touches on here:

"One point of controversy was around Digg’s claim of 20 million unique monthly visitors and steep monthly growth, whereas the Comscore’s most recent September report shows only 1.3 million monthly unique visitors and flat growth since April (see chart below). Comscore is notoriously flaky, and these numbers are for U.S. households only. Comscore is almost certainly significantly under-reporting Digg traffic."

Michael is one of the best bloggers ever and I read Techcrunch every day. But I think he got this one wrong. Comscore is not "flaky". They are a third party measurement service. They don't always get everything right. None of the third party measurement services do. But they are the best of the lot in my opinion. Now I am biased as I have been an investor in Comscore since 1999 and have been on the board since then.

Even so, the sample may select to bias towards consumers versus, say, technology leaders. Men versus women, rich versus middle-income, etc. I'm not saying problems don't exist, but at least internet panels are based on more information than surveys and logbooks.

Link

Tags: , , ,

{ 2 comments… read them below or add one }

1 Anonymous October 31, 2006 at 9:21 am

George: This is a very complex problem with very few right answers (and there are no right answers even on the horizon). It is right of the big brands, or little web babies, to want accurate numbers. I think auditing might atleast light a fire under some of the current players to get more accurate and more open.
I have to admit that I dissent just a little bit with your recommendation of using panels. There is simply too little data (for comScore roughly 2 million panelists represent roughly 800 million people who are on the web, and they have to install a, shall we say, “monitoring software” on their computer).
I recently wrote a post comparing various options that are currently available on traffic measurement from the outside (covering Alexa, ComScore & HitWise). Perhaps it will be helpful to your readers:
Competitive Intelligence Analysis: Why, What & How to Choose
http://www.kaushik.net/avinash/2006/08/competitive-intelligence-analysis-why-what-how-to-choose.html
So what's the answer? Two thoughts:
1) Go for the external company, such as the ones above, that has the best sample and represents the users that you are looking for. Take a asprin and cross your fingers and then use the data.
2) Website tracking is extremely sophisiticated, and accurate now. Think of Google Analytics or Omniture or WebTrends or ClickTracks. Regardless of if you use javascript tags or web logs to collect the data they are increasingly more accurate and yes between vendors there is a difference but well within 10% (much less than what you will see in option #1 above).
The answer, for now, for “accurate” understanding of your numbers is to use both options above, stay consistent over time and you'll do fine (remember you've already had asprin!).
-Avinash.

2 Anonymous November 23, 2006 at 8:34 am

comScore does have alot of problems with accuracy, even in its sample audience. The most notable example of this is how comScore ( and almost every other third-party measurement source ) handles framelinks. I have an article on the problem here:
http://www.onedegree.ca/2006/11/15/big-problems-with-thirdparty-measurement

Leave a Comment

Previous post:

Next post: