Part 5 - Buying domains with CTR, EPC and Parking

Part 5 - Buying domains with CTR, EPC and Parking

We've managed to plough our way through a whole lot of information on buying a domain name but the best is yet to come. This is part five in the "Domaining 101" article and in this part we'll be discussing the impact on a domain's valuation caused by earnings per click, click through rate and parking companies.

Click through rate (CTR) is very important when purchasing a domain because it often indicates whether there are live, real people at the other end interacting with a parked web page.

I've seen many people on the forums trying to sell domains with thousands of unique visitors per day. Unique visitors do NOT necessarily mean an actual person, so beware!

For example, the domain may have previously been a virus launch site so that it is getting repeatedly hit by "Trojans" on different IP addresses around the world but none of the traffic is real.

Other methodologies of increasing traffic artificially include reporting not unique users but a total traffic number. This is often done by sellers who only report web stats and not screen captures of parking statistics. Web statistics often also include currently unmonetizable traffic such as graphics and applications. This makes the numbers look great (particularly for ex-photo websites).

In my view, other than if the seller is trying to exit a business (sell a website) the seller should always have parking company statistics. It is the rare sale that needs to be done so quickly that there was not time to get even a few days of stats.

When I see a "claimed" click through rate for a domain the first thing I do is check out what the keyword (if any) is being applied to the domain. I then compare this CTR against the average CTR for all our domains at ParkLogic for that keyword to make sure that it's a reasonable claim. If it is out by a lot then I either stop the negotiation or ask some more questions.

I also ask for statistics over the last couple of months as this will often show up any potentially fraudulent clicks. It's easy to add clicks for a week or so but much hard to consistently sustain clicks over a couple of months.

Likewise I also like to see earnings per click stats (EPC) and relate them back to a keyword. We've been tracking keywords for years and checking the existing keyword payout rate for a domain against our database has been invaluable in determining a potentially likely fraudulent claim.

A wild fluctuation in the EPC across a couple of months often indicates a category that does not have very much advertising. This strikes at the heart of the consistent return on your investment and should ultimately impact your offer to the domain owner.

For example, I've seen a domain that has earned 3 cents one day and $178 the next. So what's it worth? No idea. You can't average those two numbers as there isn't enough data to determine if the payment rate per day is then going to shift to becoming 3 cents or $178. Due to this uncertainty the domain purchase price needs to go down.

What I also look for is whether a keyword has been appropriately applied to a domain name. If a domain which is music related is using a "games" keyword then you can bet that you'll increase the daily rate that the domain earns with a new keyword.

I always check where a domain is parked. I then ask if the seller has any special relationship with the parking company that is artificially inflating the figures. If I have a special deal which is better than the sellers then it will help me adjust my return on investment calculation accordingly.

On the other hand if the seller has a better deal then me then it may cause me to talk with the parking company and seek under what terms I can also get that deal. Always remember, if you don't ask then you don't get.

So the CTR, EPC and parking companies all play a pivotal role in helping us determine the price we would pay for a domain name. The challenge is to create an algorithm that encompasses all of these variables that results in a meaningful result. I'll see if I can have a go at that in the next part.