|
Now that the journey towards standards and transparency has been outlined in part 2 we can now examine the first metric - traffic.
 On the surface traffic should be quite easy to define. It's a person that looks at a web page isn't it? Well, not exactly. For example, is it a person or an IP address? If it's an IP address then across what period of time? Is it graphics traffic, html, php or exe traffic? AS you can see just defining what traffic is or isn't can be a challenge!
From experience each parking company defines traffic in their own way. Some use an IP address over an 18 hour window others over a 24 hour window while yet still others don't use IP addresses as all. Each also has their own fraudulent traffic detection algorithms.
Let me first say that I have some sympathy for parking companies as they quite often have to extrapolate and guess information so that domainers can receive up to the minute data on their domains. Domainers need to appreciate that the pressure that we place upon parking companies to deliver up to the second stats creates a situation where the data being reported is inherently inaccurate.
I know that many of us like checking our statistics a number of times per day but I believe that we need to resist the impulse. I often wonder what's the point of looking at stats which are wrong and what would we do with the information anyway?
To overcome this situation we need to separate what the parking company knows (Green) as to what they will know in the future (Red). For instance, the parking company knows what hits their web pages but what they don't know is what Google or Yahoo counts. When purchasing a domain, domainers often just want to know if the domain has traffic. This IS reportable in a timely fashion.
Since there is a direct relationship between traffic and revenue we need to also break the definition of traffic into the stages that the traffic goes through to be counted as "real". I believe that these definitions need to be reported upon on a domain by domain basis.
Please note that these definitions are what I believe needs to be reported and that they do not require a parking company to indicate what their anti-fraud technologies are but rather to report on the results.
URL Traffic
The total URL hits are the total number of requests for ALL traffic made to the domain name. This means everything with NO filtering. The reason why you want to know this information is that you may find that a particular domain has a huge volume of traffic that is unsuitable for parking but great for affiliate programs or development.
Web Traffic
This is the total amount of traffic to a domain that would normally appear in a browser. In other words, it's the total traffic less the graphics, .exe, sound, video requests.
Other than file type, no filters should be applied to this traffic. What this figure will indicate is the level of interactivity that a particular domain receives. For instance, let's imagine that we have a domain with ten uniques (see definition below) but 50 Web hits. This would indicate a high level of interactivity that could be capitalized on as well as influencing the value of the domain.
Traffic Sent to Advertising Aggregator
This is the traffic that is sent from the parking company to Google/Yahoo. Most parking companies do a level of filtering at this point so that any "suspicious" traffic is not sent onto the advertising aggregator. The main reason for this is that if all traffic is sent on then potentially the whole portfolio of domains will be "smartpriced" downwards.
Each parking company has there own levels of aggressiveness for their filtering systems and having the true volume of traffic sent to the advertising aggregator helps a domainer make a choice of where the domain should be best parked. As a side benefit if every parking company reported the total number of URL hits then RPM numbers would not need to be normalized between each of them.
Total Uniques
It's one thing to have sent an advertising aggregator traffic and it's quite another to have them acknowledge that it is real traffic. From my understanding of the NameMedia contracts with Google and Yahoo they are able to receive the total number of impressions on a domain by domain basis. This number needs to be reported back to the domain owner so that they are now clear on how much of the traffic is being paid for.
We now have:
1. URL Hits - Traffic received by a domain.
2. Web Hits – Traffic normally seen in a browser.
3. Traffic sent to Google/Yahoo
4. Traffic acknowledged by Google/Yahoo
These four steps complete the traffic cycle for the domain name owner and provide transparency into what is happening with the volumes of traffic at each level. At no stage has the proposed reporting standard asked for what is inside the filters at the various levels, as this could potentially open up the whole system to increased levels of fraud.
In the next article on standards I wish to examine one of the problems associated with the current system of traffic reporting.
Wiki: IP Address, HTML, Google, Yahoo, URL, RPM
Related Articles: Parking Company Standards - Part 1, Standards - Part 2 - The journey to transparency
Trackback(0)
 |
TrackBack URI for this entry
Subscribe to this comment's feed