strategy Demand Based Domain Pricing Strategy - Part 3

[Whizzbang]

In the previous articles in the series I discussed how much a domain is really worth and some of the potential traps that some domain investors fall into in selling stock-item domains (ie. multiple keyword domains). In this article I plan on tackling both the pricing and portfolio models from the perspective of supply and demand.

What we do know about the supply/demand curve is that once a domain is acquired the cost of supply is constant for .com domains. Each year there is the same renewal fees (assuming no increase) and the renewal fees are identical to each domain whether it’s google.com or fredspizzashop.com. This line is represented in the blue colour in the below chart.

Read more >

 

[TRY]

Thank you for sharing, As always pleasure reading your blog.

 

[Whizzbang]

Thank you for sharing, As always pleasure reading your blog.

Thanks for that!

 

[frank-germany]

to summarize:

BS domains sell better below $1000 USD
and
good domain may sell higher

great info
thanks Michael

 

[Whizzbang]

to summaries:

BS domains sell better below $1000 USD
and
good domain may sell higher

great info
thanks Michael

LOL! I'm never sure whether your comments are complimentary or sarcastic....I'll assume the more positive.

The importance of the article was that every market vertical has it's own demand curve which with enough data points can be calculated. For example, two domains from two different market verticals may have a completely different price, even though they are in relatively the same position on their respective demand curves.

For example, a loans domain may command a far higher price than a gaming domain due to the nature of the market vertical. What we should be able to do is actually calculate the correct BIN price for each domain. At the moment both domains may be set at $1,000 and the domain owner is leaving a lot of cash on the table for one of them.

I think that I need to follow-up the article with a lot more detail around my thinking on this...

 

[frank-germany]

What we should be able to do is actually calculate the correct BIN price for each domain. ...

I must have misted the part where you explain HOW

 

[TRY]

One way you can get a general idea of what value a keyword has relative to another is to use the Google adwords tool. Of course many other factors should be used.

So say for example loans is $10 a click and shoes is $2 a click. So the visitors that visit loans site is 5 times more valuable than the shoes site visitors.

Then you analyze the quality of the domain name and see where it fits in regards to its desirability in the vertical and price accordingly.

 

[frank-germany]

One way you can get a general idea of what value a keyword has relative to another is to use the Google adwords tool. Of course many other factors should be used.

So say for example loans is $10 a click and shoes is $2 a click. So the visitors that visit loans site is 5 times more valuable than the shoes site visitors.

Then you analyze the quality of the domain name and see where it fits in regards to its desirability in the vertical and price accordingly.

thats a strategy for keywords domains only

are we talking about keyword domains?

 

[Whizzbang]

Yes - we are talking about multi-word keyword domains that I refer to as stock-items.

I must have misted the part where you explain HOW

The demand curve can actually be plotted through a combination of factors, a number of which I outlined in the article. See below:
"Obviously the more data points that I can put into the mix the better the result. For example, what is the demand for Google adwords for 3D printer related keywords? If I put the domain names into the keyword search tool, then out pops a cost to buy that keyword. Since this is a dynamically changing auction it provides a real snapshot of the demand curve for customers. Once I match this up with the other data I’ve collected I should know which domains are actually my diamonds and which are my straw."
The next article will take these thoughts one step further towards implementation as there are a number of other factors that need to be considered in pricing.

 

[frank-germany]

Yes - we are talking about multi-word keyword domains that I refer to as stock-items.


The demand curve can actually be plotted through a combination of factors, a number of which I outlined in the article. See below:
"Obviously the more data points that I can put into the mix the better the result. For example, what is the demand for Google adwords for 3D printer related keywords? If I put the domain names into the keyword search tool, then out pops a cost to buy that keyword. Since this is a dynamically changing auction it provides a real snapshot of the demand curve for customers. Once I match this up with the other data I’ve collected I should know which domains are actually my diamonds and which are my straw."
The next article will take these thoughts one step further towards implementation as there are a number of other factors that need to be considered in pricing.

please give an example how you calculate a BIN out of this data

 

[TRY]

I think this type of information is more so important for buyers (corps) so they can justify paying what the sellers are asking for to their bosses, as well as using the info as a negotiating tool.

 

[Whizzbang]

I think this type of information is more so important for buyers (corps) so they can justify paying what the sellers are asking for to their bosses, as well as using the info as a negotiating tool.

Completely agree with you that the information can be used on both the buyers and the sellers side of the negotiation.

 

[mhdoc]

Two thoughts.

1. For some reason your article makes me think back to my biometrics class long ago when the subject of the Poisson distribution came up. As I remember, it was useful for describing random and rare events. Domain sales certainly fit that model. Have you looked at applying this kind of distribution to your analysis?

2. That class provided my first introduction to computers and I was in analysis heaven when I discovered there was a program running on the mainframe that would let me run stepwise linear regressions. I was calculating nutrient budgets in an aspen ecosystem and spent many happy hours trying to define every last bit of the variance. I had large funnels collecting rainwater as it dripped down through the trees after which we would measure the chemical composition. These were milligrams per liter of Calcium, Potassium, etc. measures with most values less than 500 mg/l.

On a collection trip after a big rain I discovered that one of the 12" collection funnels had gotten plugged with leaves, filled with water and a chipmunk had fallen in and drowned. That one body contained way more nutrients than all the other samples I had collected. It was an experience that went a long way toward convincing me natural systems are WAY more heterogenous than I had appreciated and not well described by equations that assume things like random samples and normal distributions. Means (of all types) and other statistics have their place, but I suspect most math describes a world that doesn't really exist.

 

[Whizzbang]

Two thoughts.

1. For some reason your article makes me think back to my biometrics class long ago when the subject of the Poisson distribution came up. As I remember, it was useful for describing random and rare events. Domain sales certainly fit that model. Have you looked at applying this kind of distribution to your analysis?

2. That class provided my first introduction to computers and I was in analysis heaven when I discovered there was a program running on the mainframe that would let me run stepwise linear regressions. I was calculating nutrient budgets in an aspen ecosystem and spent many happy hours trying to define every last bit of the variance. I had large funnels collecting rainwater as it dripped down through the trees after which we would measure the chemical composition. These were milligrams per liter of Calcium, Potassium, etc. measures with most values less than 500 mg/l.

On a collection trip after a big rain I discovered that one of the 12" collection funnels had gotten plugged with leaves, filled with water and a chipmunk had fallen in and drowned. That one body contained way more nutrients than all the other samples I had collected. It was an experience that went a long way toward convincing me natural systems are WAY more heterogenous than I had appreciated and not well described by equations that assume things like random samples and normal distributions. Means (of all types) and other statistics have their place, but I suspect most math describes a world that doesn't really exist.

I haven't heard about Poisson distributions for a long time....it's good to be reminded about them! Much of what you described is directly applicable to the domain industry. Many thanks for your thoughts.

 

[frank-germany]

On a collection trip after a big rain I discovered that one of the 12" collection funnels had gotten plugged with leaves, filled with water and a chipmunk had fallen in and drowned. That one body contained way more nutrients than all the other samples I had collected.

wow!

domaining at its best !

great
thank you

 

[TRY]

wow!

domaining at its best !

great
thank you

"Sarcasm is said to be a low form of humour as its intent is generally to get laughs at someone else's expense." ...I am also guilty of using Sarcasm, trying to cut down though.

 

[frank-germany]

On a collection trip after a big rain I discovered that one of the 12" collection funnels had gotten plugged with leaves, filled with water and a chipmunk had fallen in and drowned. That one body contained way more nutrients than all the other samples I had collected.

"Sarcasm is said to be a low form of humour as its intent is generally to get laughs at someone else's expense." ...I am also guilty of using Sarcasm, trying to cut down though.

no scarcasm from my side this time

I honestly think that's domaining as it happens

 

[Dondomainer]

First,Thanks for sharing, now what I have to say.

1 ) I think you describe only one side of the block. and ignores factors that add more complexity.

2 ) Everyone talks about "Speculators" but who is free from sin.

3 ) You can make money by registering domains for $ 1 ? The answer is : hell YES!

 

[SBS66]

@Whizzbang Thanks for the articles!

 

[Whizzbang]

@Whizzbang Thanks for the articles!

Not a problem....it's my pleasure.

 

[garptrader]

Perhaps for a future article....

How many end users / small businesses / developers are willing to pay a premium price (>$100 which would permit investment to cover investment and administrative costs) for a domain name say annually (market)?

Of those, what percentage are willing to pay more than low $XXX?

Of those, what percentage are willing to pay more than high $XXX?

Of those, what percentage are willing to pay more than low $XXXX?

For each category, how many would be willing to pay that price for a non-.COM domain?

Here in downtown West Palm Beach, it is common to see restaurant and retail stores close. There is ample foot traffic, but lease space is expensive and apparently demand is inadequate to cover the cost of operating those businesses.

Every week we see aftermarket sales reports but those sales are a small percentage of available domains for sale. If there are hundreds of available aftermarket choices for the rare end user willing to pay a premium for a domain name, then many investors are going to be unable to pay renewals from sales. Many developers and businesses are content to use social media platforms andfreebie development platforms for their web presence while most are content with what experienced domainers would consider newbie quality handregs.

 

[Whizzbang]

Really good questions.....in fact, I'm halfway through an article to answer these very questions.