tips Simple Domain Valuation Equation

Can you translate to English?

Genius~!

johnn said:

Can you translate to English?

Click to expand...

A value of a domain can be estimated by taking into account is neighbors.

So, if you have namepros.com ; the value could be estimated by:

1. The second level domains : .net, org e.t.c
2. The number of times it has been resold.

This is a simplified explanation but it in principle, that's what the paper is talking about.

A more simple but preferred and recommend technique would be to use the median values (median and media absolute deviation). Not the mean and mean absolute deviation/standard deviation.

Thies Lindethal uses real estate theories and research as a proxy to analyze the domain market. As you know domains are no different that physical plots of land.

Domain names - digital plots of land
Real estate - physical plots of land.

See:

1. A Guide to Aggregate HousePrice Measures by Jordan Rappaport
2. A better method to estimate price change in single family housing by Donald Epley (this compares the median technique with the technique Thies Lindenthal employed in the paper @DnameAgame quoted.)

DnameAgame said:

Hi All,

Been a good week, just had a sale close on BB today, in a good mood & wanted to share:

With so many confused about domain valuation, we figured offering a simple valuation equation would make it much easier for all.:

See below and life will be easy and figuring out your domain value should be a breeze!

Nice and easy now? hahahaha, I shouldn't be giving this secret away.
______________________________

The simple domain value equation:

Conceptually, the sales price P of domain i is split into three components:

Show attachment 115459

where SLD and TLD capture the quality of each SLD and TLD. Time dynamics enter the equation through Dt. Differencing transactions that share the same SLD cancels out the SLD component:


Show attachment 115461

which leads to an equation that can be empirically tested in a regression:

Show attachment 115462
______________________

Look up Thies Lindenthal, paper (this is from)called: Valuable Words: The Price Dynamics of Internet Domain Names. interesting stuff to learn.

Enjoy,
Brian

Click to expand...

Interesting stuff!

Could you pls explain AsSeenOnTV(.)com through that formula?

PS: sold for $5M

Is this joke, April fool? if not then you have no chance makeup in domain!

ValleyRock said:

A value of a domain can be estimated by taking into account is neighbors.

So, if you have namepros.com ; the value could be estimated by:

1. The second level domains : .net, org e.t.c
2. The number of times it has been resold.

This is a simplified explanation but it in principle, that's what the paper is talking about.

A more simple but preferred and recommend technique would be to use the median values (median and media absolute deviation). Not the mean and mean absolute deviation/standard deviation.

Thies Lindethal uses real estate theories and research as a proxy to analyze the domain market. As you know domains are no different that physical plots of land.

Domain names - digital plots of land
Real estate - physical plots of land.

See:

1. A Guide to Aggregate HousePrice Measures by Jordan Rappaport
2. A better method to estimate price change in single family housing by Donald Epley (this compares the median technique with the technique Thies Lindenthal employed in the paper @DnameAgame quoted.)

Click to expand...

Can you translate back?
This one is even more complicated than the other one.

April fools was two weeks ago.

There is no formula that means anything, let's look at similar CVCV that repeat the same two letters, so a lot in common. Dudu.com sold for $1million and LoLo.com sold for $37,000 and then a $16,000 loss 3 years later. I would argue LoLo is better than DuDu but at the end of the day the only appraiser of a domain that matters is the buyer with the greatest offer.

I was not able to access the full article (well without paying for it which I was too cheap to do ).
I mean the idea that the word, the extension and the point in time all contribute makes sense and is not rocket science. The difficulty of course is to place value on the word and TLD. From the bit kindly provided, it is not obvious to me that the interaction between them is taken into account.

Let me give an example - yesterday on GD auctions butterfly.net sold for a good price. To a human it is obvious why that is. But if we split it into the three parts of the equation butterfly is good but probably not great, while of late .net is good TLD but not as great as it once was. Of course the integration with extension is critical in evaluating new extensions.

Now maybe the regression properly considers all the cross terms, but without reading the full article not able to decide.

Thanks for drawing out attention to the article.

Bob

The domain market is not a science.
You can buy one name here on NP for $20 and sell to small business for $2,000 or a big boy for $20,000.

Using any formula will make you buy crap names and stuck with them for a while.

johnn said:

The domain market is not a science.

Click to expand...

Because it is not an exact science does not mean that there is no value in using scientific analysis to study trends, values and variability. The reason domain prices are not exact is because ultimately the price will be what it is worth, or perceived to be worth, to one or more organizations. Nevertheless looking at how similar words sold, values that TLD generally sold at, whether names of this kind are going up or down in price, are all worthwhile to do, imho.
Bob

Bob Hawkes said:

I was not able to access the full article (well without paying for it which I was too cheap to do ).
Bob

Click to expand...

As attached

1. A better method to estimate price change in single family housing by Donald Epley
   
2. The price dynamics of Internet domain names by Thies Lidenthal
   
3. A Guide to Aggregate House Price Measures by Jordan Rappaport

This is an quick overview regarding 2 word "dictionary" domains.

So we have 2 methods for price estimation:

1. Repeat Sales
2. Median Sales

For repeat sales, you need to find similar domains which have been sold more than once, hence the description "repeat sales", and you also take into consideration the "time between sales".

Using a domain the domain "AlleFinance.com" to illustrate in repeat sales you need to find;

• Similar names which have sold more than one time - SunriseFinance.com, CanyonFinance.com, WhiteFinance.com
• Must have the time between the sales - this is not easy to locate and in some case not available. This alone: the difficulty of locating more than one sale for similar domains discourages investors from employing the repeat sales methodology.

Now lets look at the Median sales, still using the domain "AlleFinance.com" - checking namebio.com you can see the following:

• WhiteFinance @ $245
• CanyonFinance @ $201
• SunriseFinance @ $980
• AimFinance @ $434

So using the median sales you price estimation for a "word + finance" in "com" can be calculated as :

johnn said:

The domain market is not a science.
You can buy one name here on NP for $20 and sell to small business for $2,000 or a big boy for $20,000.
Using any formula will make you buy crap names and stuck with them for a while.

Click to expand...

This is exactly how Henry Ford was thinking when he said "they can have it any color as long as it is Black".

And we all know how that ended.

posted another formula couple of months ago

https://www.namepros.com/threads/wh...-creating-a-domain-name.1125229/#post-7129309
imo

you should have written a domainguru help book and charged money for it like some domainers do hehehe

but I like your sense of humor... this shoudl probably be in breakroom though

biggie said:

posted another formula couple of months ago

https://www.namepros.com/threads/wh...-creating-a-domain-name.1125229/#post-7129309
imo

Click to expand...

Seems more like a principle than a method.

Bob Hawkes said:

Because it is not an exact science does not mean that there is no value in using scientific analysis to study trends, values and variability. The reason domain prices are not exact is because ultimately the price will be what it is worth, or perceived to be worth, to one or more organizations. Nevertheless looking at how similar words sold, values that TLD generally sold at, whether names of this kind are going up or down in price, are all worthwhile to do, imho.
Bob

Click to expand...

This isn’t right, very rare for a nome to sell so using similar sales will result in value too high!

Most new tld’s only have a couple of sales over 10,000+ registration. Better to compare to gambling/lottery.

johnnie018 said:

This isn’t right, very rare for a nome to sell so using similar sales will result in value too high!

Most new tld’s only have a couple of sales over 10,000+ registration. Better to compare to gambling/lottery.

Click to expand...

Exactly it's not right, you can look at 4L.coms that share several of the same components/style and the sales prices are nowhere close, I always use DuDu.com and LoLo.com Dudu sold for $1million, LoLo sold for $37,000 and then $21,000 three years laters, you have 2 CVCV that are the same CV repeating, sold within 18 months of one another. We can add in JoJo.com for $25,500.

equity78 said:

Dudu sold for $1million, LoLo sold for $37,000 and then $21,000 three years laters, you have 2 CVCV that are the same CV repeating, sold within 18 months of one another. We can add in JoJo.com for $25,500.

Click to expand...

You need to control for location.

ValleyRock said:

You need to control for location.

Click to expand...

Also must control for the theory being mess and worthless!

Glad this got a good discussion going & people were interested.

One thing that always comes to mind & think we all know it's coming - like it or not - AI will be playing a very active role in everything everyone does, in all industries, & in all aspects of life. But, lets just think about marketing for this post. Will AI be evaluating all aspects of branding and marketing and spitting out best ROI action plans - that will be followed blindly by the masses? Maybe? a good guess is probably.

To the point. If you dig deeper into the author behind that paper and the several related projects (Sed o, GD are 2 biggies) you will see what that ecosystem we will have to work within looks like as AI take more of the individual decision making process out of the mix.

These papers will lead you to so much info, including the methodology & metrics applied to the valuation algorithms used by - well just about all of the services, that automate things we investors know shouldn't be - but have to be conscious of & consider (for yourself only) are you more trusted by the customer or is GD on the value of a domain?

This is a real/serious question to ask yourself as it varies greatly: (e.g. known broker selling premium .coms in the 25k+ range (with reference and history) brings something different to the trust table than a new investor with a "white label" portfolio site selling hundreds of $xx & $xxx names.

I do agree formulas are not how "we" can work effectively as an established investor/biz. & hope that my OP showed it was a bit of humor with some interesting info.

But, we cant dismiss what we can see in writing (serious -massive- detailed -multi year - multi company- have to be amazingly expensive research papers) how the biggest players in the space will be playing the game.

So much to plan for, which is why this industry is so exciting.
Brian

The domain is more like a belief. Because they believe, they will pay.

So our goal is just to find people with the same faith.

a domain name is worth to a buyer
related to the enhancement of his future profit
caused by that domain name

so how can you calculate the future profit
of an unknown buyer?

BS