Metrics Values | 1 thru 10 | 11 thru 20 | 21 thru 30 | 31 thru 40 | 41 thru 50 | 51 thru 60 | Main Traffic Page

Traffic Reports
Problems With Ad Click Count Accuracy

2cooleysphoto.gif (3642 bytes)

Do you have links in various Web sites that give you a user name and password to view a statistics page for your particular link or banner? Does this statistics page give you page view counts (impressions), click through counts, etc? If so, expect your real traffic report numbers to show less click throughs than what these statistics pages show. The issue of over-reporting is a well known problem within the online advertising industry.

Over-reporting is what got Google in hot water that resulted in a law suit with Google settling for $90 million. To me it is astounding that any site would still offer these little statistics pages.

Why It Happens
Internet Technology and telephone technology are closely related. How many times in your life have you dialed someone's phone number and heard a busy signal? How many times have you heard the phone ring many times and go unanswered? How many times have you heard that nice woman recording saying, "We're sorry. All circuits are busy now. Please try your call again later."?

These are just a few examples of how your call does not go through. And similar issues with computers talking to each other occur that cause inaccurate counting on those little "stat" pages.-

The world's top authority on this issue is the Interactive Advertising Bureau. This non-profit organization sets world standards for all types of interactive and online advertising. Their members sell 86% of all online advertising. You can visit the IABs Web site at http://www.iab.net


Anyone who knows anything about traffic metrics knows that what IAB says here is true and correct, and therefore will never argue that statistic pages are accurate.


Quote from IAB's Ad Impression Measurement Guidelines:
"A click does not guarantee that a visitor actually arrives at the requested target URL, it only measures the opportunity for the visitor to be transferred to the target URL. This means that a click will be considered valid even if the visitor hits "stop" or otherwise aborts before arriving at the target URL. The click will also be valid if the target URL is busy or not available. In practice, a click will typically be recorded when a Web server or Ad server executes a program designed to redirect the visitor to a target URL.

It is important to note that clicks are not equivalent to web-site referrals (visits) measured at the destination site. If an organization complies with the guidelines specified herein, there will still be measurement differences between originating-server measurement and the destination site (advertiser). The use of 302 redirects helps to mitigate this difference because of its easy and objective quantification. However, differences will remain from measurements taken at the destination site because of various issues such as latency, user aborts, etc. The subject of the magnitude of this difference may be a subject for future phases of this project." End IAB quote.

A complete traffic report shows the number of successful hits, the number of failed hits, what part of the failed hits were server errors, and what percentage of failed hits were client errors.

In all traffic reports I see anywhere from 13% to 24% of total hits as failed hits. Remember, hits are not visits, or visitors. However, even a small percentage of failed hits, either client or server, will cause clicks from other sites to fail. When this happens the click is recorded in the site of origin as a click, but that click does not result in the person who made the click actually seeing the site. It is this event to which the IAB refers to in the above quote.

What It Means
Okay, I know. Technobabble. The "originating server" is where your link sits on another site. The "destination site" or advertiser, is your Web site. The important point is that IAB is stating "there will be differences".

Many sites still offer these little statistics pages. 2Cooleys does not, and will not unless the technology improves. These little statistics pages are almost always going to over-report the click count, and usually by substantial amounts. I've seen this for years.

Yet whenever I try to point out this problem I get accused of being "sour grapes" and "not being a team player". The errors caused by the statistic pages is such common knowledge within the online advertising industry that anyone who still advocates use of the statistics page does NOT know what they are talking about. I don't care who they are, or how "Official" they think their word is. It is the equivalent of saying the world is flat.

The problem is with the "redirect" files that must be used to run the little statistics pages. IAB states in its standards that the redirect should be on the same server where the requested Web site is hosted. The redirect should NOT be hosted on the same server as the originating site. Yet folks selling banners and links want the redirects on their server so that you will see the pumped up numbers. As you can tell, this is a pet peeve of mine.

Here is one example of how redirects got Google in hot water: 
Lane's Gifts and Collectibles, an Arkansas merchant, filed a suit against Google in February of 2005. Lane's claimed improper charges from the Google AdSense advertising system. The Google "click counter" was over-reporting click throughs. Google offered $90 million to settle the case without going to court.

Why Do They Still Use Redirects?
The common sense question here is, "If the statistics pages are over-reporting, why would organizations keep using them?". The fact that so many sites employ this faulty reporting means does not make it right and good. The answers are varied, but here are the most common reasons:

1. They just don't know any better. They have not done their homework, they have no idea who, or what, IAB is. The test is to ask them very simply, "Do your redirects comply with IAB guidelines?". If they ask, "Who is IAB?" you know they don't know any better.

2. They know better but are hoping you do not. The over-reported numbers impress you more, and therefore you are encouraged to keep buying their program.

3. They know better, but have so much time invested setting up all those statistics pages they don't want to stop using them.

4. They know better but are embarrassed to admit it. It makes them look bad.

5. They disagree with IAB. Not smart. That is what Google tried to do - and lost. Now Google complies with IAB guidelines. Imagine that.

It Is No Big Deal
In the end all that matters is that you do not rely on what you see on your statistics page. Just be aware of the problem and rely instead on your real traffic report.

Written August 2008 by Gary Cooley

Metrics Values | 1 thru 10 | 11 thru 20 | 21 thru 30 | 31 thru 40 | 41 thru 50 | 51 thru 60 | Main Traffic Page