Domain Names Portfolio Management Theory

Not open for further replies.


Level 8
Jul 10, 2007
Reaction score
Feedback: 10 / 0 / 0
I wrote the beginning of this paper after a night of drinking & smoking. It would be great if this can be written the wiki-style, with folks knowledgeable with the diff aspects of PM contributing to this effort. I truly believe that DNForum is a great place to contemplate this effort. All meaningful contributions are welcome! All contributors will be acknowledged.

Domain Names Portfolio Management Theory
Simplified version with emphasis on educational & [FONT=&quot]entertainment[/FONT] aspects of DNPMT and some historical notes

Domaining is becoming a business in its own right. Literally thousands of investors all over the world are putting time, money and efforts in collecting and trading domain names. This trend is somewhat similar to the stock and bond investing mania of their parents’ and grandparents’ generations. The following Table lists estimates of the total value of the world financial markets in the trillions of dollars:
Treasuries: $27 T
Corporate Bonds: $15 T
Municipal Bonds: $25 T
Stocks: $50 T
There are also commodities, FOREX and the real estate markets. And probably a few more.

In comparison, the entire size of the ‘visible’ domain name market is no greater than $3 Billion. We are in the midst of the new ‘industrial’ revolution – this time info-lution. It’s taking the entire world economy by the storm. The old ways of doing business are going the way of Dodo. The entire infrastructure of the sales and marketing force in most industries is being either discarded or totally revamped. And just like Titanic-the ship saw only the tip of the iceberg, it was the ‘invisible’ part that did the trick. The size of the economy that critically depends on the web presence and web transaction capability is already in the Trillions. Par example, multimillion-dollar businesses are built around,,, Mapquest, Amazon,, Travelocity, eBay, etc. The future wannabe: For the new large-scale enterprise the actual dollar price paid for the domain, whether it’s a $10 reg fee or $10 million on the secondary market is becoming irrelevant.
In this paper we will try to bring about and apply the ideas and methods of investing and portfolio management which are already in existence in other markets.

People were collecting stuff for fun and pleasure since they were monkeys. The Egyptians were collecting the pyramids (Were those pyramids any good for the pharaoh’s contemporaries other than to starve and kill the slaves employed in the pyramid’s construction?) , the Greeks – scientific knowledge, the Romans – slaves & the legal code, the Jews – Talmud wisdom and moral code, the Dutch in XVIc. – Tulip flowers (1), the Germans in the 1940s – gold teeth & hair from the prisoners (2), Russians and East Germans in the 1970s – Olympic Gold medals with the help of anabolic steroids (3), American and European investors – Treasuries & stocks, including billions of dollars of worthless stocks (1950s – current), baseball players – home runs with the help of the same steroids (2000s), Geeks an Nerds - domain names (1994 - ?), etc. As you can see from this short list many of the items people collect are of dubious value.

The sheer volume of DN is mind-boggling. As of 2007 the numbers are:
Biz 1,860,669
Info 4,981,597
Org 6,194,878
Net 10,476,009
Com 73,433,353
Total 96,946,506

In 2008 it will be well over 100 million.

By default Domain Names became an investment class. Many wealthy investors and financial institutions started gobbling up this product. And just like with any investment people found out that some are real pearls, some are ‘working horses’, but most are rotten apples. In the absence of any meaningful guide investors turned to the forums to exchange ideas and products. A number of simple ideas on what constitutes a ‘good’ domain became public knowledge. Investors started to buy not just single domains, but small and large portfolios. Currently there are pools with 50,000 and 300,000 domains in them. But most of the portfolios are still 100 to 5,000 domains.

The issues that these new-found managers facing are formidable:
Am I collecting names for fun only?
Can I stand to lose money every year?
How much can I stand to loose?
Can I make money every year?
Shall I trade?
Shall I buy at Land rush, secondary markets, drops, auctions or just reg it myself?
How do I optimize my portfolio?
Can I swap portfolios?
How do I price portfolios?
Shall I diversify?
Can I make big money?

In the mid-XIX century American economy started to generate humongous wealth. The titans of that period invested their family fortunes with the financial horizon of several generations, probably a 100-year horizon. These days many bankers at the major Wall Street firms play an arbitrage game where the ‘investment horizon’ is 10 seconds or shorter. We will talk about the investment horizons and how they affect the financial picture.

As the financial markets started to grow and mature the new financial theories abound. Harry Markowitz from Baruch College in New York did a fabulous job in the 1950s on Portfolio Theory: hm
Michael Milken mm was pioneering the modern high-yield (junk bonds) market at Drexel in the 1980. He pissed off a lot of well-entrenched Ivy-league professors and the Wall Street old-timers by showing that his investing methods are way superior. They were superior to the point that the old guard decided to destroy Drexel & put Milken away. At his predicament the issues discussed in Greenwich, Connecticut were ‘class, ethnicity & entitlement’. Milken is a thorough portfolio manager and a scholar. But he didn’t publish any scholarly papers. He did talk about his management methods at various meetings with potential clients. His methods of maximizing portfolio performance are directly applicable to the DN industry.

Black, Scholes, Merton, Cox & Ho applied equilibrium methods from Mathematical Physics to Finance to price financial derivatives: bs.
A great number of finance scientists were awarded Nobel prizes in the last 20 years. These studies are distinct from Economics in a major way: They talk specifically about the mechanics and underlying math of the specific investments. But it is interesting to note that many of the latest crops of the Nobel folks in finance theory didn’t do so well in the financial markets themselves as investors. As a matter of fact, many of them invested in the Long Term Capital Management Fund (ltcm), which almost crippled the entire world economy! They weren’t just passive investors. They came in as advisors and gurus with their Nobel prizes and the Harvard, Berkeley and MIT tenures. They put on a $2 Trillion bet with John Meriwether, the hero of the book Liar’s Poker (lp). And they lost. If the Government didn’t step in and helped to unwind the positions they took the 1928 stock market crash would’ve looked more like another ripple.

24 hours before the LTCM meltdown the Federal Reserve Bank of New York folks called on the S&P brass and requested that they look into the situation. S&P hastily organized a late meeting in the basement at 25 Broadway. It was scheduled to go at 8 p.m., after all the analysts gone home. It was important not to give anybody any clue and not to freak out the Wall Street community. The cleaning crew was told that no cleaning needs to be done there. Professors from MIT, Stanford, Princeton, Yale, & Cornell were flown first class or on charter flights. The NYU folks just walked across the street from the Stern School. A year earlier S&P has developed a sophisticated computer model which puts the entire economy under dynamic stress. It was written by two Russian mathematicians who were bored to death at their day job. The game these folks played that night resembled the simulated war games that our brave generals play at the Pentagon all the time. After just a few simulation runs it became apparent to everybody that all the financial markets will collapse, there will be no liquidity and the US economy will take a nosedive.

In the 1980s the discount brokerage was introduced in the USA. Instead of paying 5% on the total transaction price many investors could trade for just $9.99 or even less. After the market turmoil in 1987 many stock and bond brokers found themselves unemployed. The smartest of them have figured that instead of pitching themselves as discount brokers they would rather become money managers or portfolio managers to their clients. And so there is a new breed of money managers who instead of running a mutual fund end up running hundreds of tiny portfolios for their clients. Many portfolio managers wouldn’t take a portfolio that’s smaller than, say, $100,000. They charge their clients a percentage of the portfolio’s total value. The old-fashioned stock broker is gone for good.

We can foresee some of the trends in domaining which will follow in the footsteps of other financial markets:
Portfolio leasing
Portfolio repurchase agreement – repos
Introduction of the DN Portfolio Duration

In simple terms, some of the deals will involve:
One party giving up portfolio revenues for a fixed period of time in exchange for fixed payments from another party
One party exchanging DN portfolio for another portfolio with or without remuneration

To correctly describe DNPT one needs the tools of stochastic analysis, partial diff equation and sigma-algebra. These tools became the everyday staple of many players on Wall Street, the analysts at the US Treasury Department and probably the classroom toys in places like Wharton & MIT. For our purposes we will use here only the simple methods or none at all.

Profit & Loss Scenarios

The following Table represents diff revenue scenarios. We look at portfolios which generate NO revenue at all, and portfolios with 1 penny a day per domain, up to portfolios with $1 revenue daily

P/L 1c
P/L 5c
P/L 10c
P/L 50c
P/L $1

A small portfolio with 100 domains generating 1 penny a day per domain will rack up $640 losses a year. Not that bad. But if your portfolio has 5,000 domains making a penny a day you should be prepared to shell out $32,000 just to stay even.

Now, if you’re lucky to get 10c a day per domain and there 1,000 domain in your portfolio you can expect to get $12,000 profit! And if you have 5,000 domains making $1 a day each you’re a fat cat with $1.75 million revenue a year. Congratulations!
We made a number of plausible assumptions in constructing this Table: your average expenses per domain are $10 a year and you pay no other expenses to maintain your portfolio.
A trivial, but important result:
The break-even point: 3c a day per domain. In lay terms it means that inn order to stay put and not to loose a shirt your portfolio should generate AT LEAST 3c a day per domain.

When you're looking at a pair of ripped underpants with the brown stains - it's WORTHLESS. If you own in any tld it's not just
Let me explain. You reg it for $10 a year. You keep it for 10 years. You charge it to your credit card at 18% annual. Your estimated cost is $ 200, and PV is little
diff. Say, you have a portfolio of 1,000 domains which you intend to keep for 10 years. There are great names like, say, FreeCreditCardIsHere.XXX in your portfolio.
Then the portfolio's NEGATIVE VALUE is around $200,000.

Star Performers
It turns out that one doesn’t need star-quality domains to make a decent living. As a matter of fact, some of the stars can literally kill your business.
Let’s take a look at the financial workings of This domain was purchased for $12,000,000 with a commercial rate loan at 11.75%/10 year.
At this rate the owner has to pay $170,435 in monthly payments. It was reported that attracts close to 3,000 visitors a day, with half of those converging. Several years ago the adult industry was paying publishers anywhere from $2 to $5 per visitor. These days the numbers are $0.25 to $2.00. Assuming is getting the top dollar we get
3000 x $2 x 30 days x 50% = $90,000 estimate.
$170,435-$90,000= $80,435 shortage!
Yes, there is a good chance that another shmuck will come along and shell out $25 million for this domain and rescue the current owner. Meanwhile, the owner is bleeding.

Synthetic Financial Instruments
DN with no revenue stream – can be approximated by the Black
DN with sporadic revenue stream
DN with high revenue stream – can be approximated by Merton’s model for the fixed-income derivatives

TLD Valuation
Lets talk about premium and generic. One word. Huge traffic. The good stuff.
This is how the tlds look today - the pecking order:

.com High Value TLD




Low Value TLD
Remember, some tld's have astronomical reg & renewal fees:
.mobi - from $15 to $75
.in - from $0.25 to $15
.cn - from $2 to $50
.cc - from $27 to $60
.jp - from $20 to $50 - from $25 to $50
.tv - from $30 to $5,000
.tm - $2000 for 10 years upfront
.kz - the land of Borat

You need to factor it in. Many tlds where either introduced or marketed by the shameless registrars. As an example, thousands of totally useless, & were sold at $25 a pop. Millions of dollars was wasted. Those tlds will never take off. Search engines ignore those turkeys.

There are a lot of well-intentioned folks losing money in wasted registrations / renewals, and even "sales" made in the aftermarket ... due to irrational and misguided hype. There is a sales thread that just recently closed where the seller stated the domain would "easily be worth $XX,xxx in 5 years from now" (and it sold on Sedo for $65.00 (only 1 bidder)!

There is another sales thread in which the Asking price is $XX,xxx ... and the person selling it acquired it for $8 less than three months prior, as well. These are just two examples of extremely wishful valuations ... versus cold reality.
The moral of the story then ... to save dollars and cents, don't register stuff based purely on hype and what others who stand to benefit the most say domains will be worth / values in the future, and don't keep renewing domains year after year that shouldn't have been registered in the first place.

Secondary Markets
Friction and Liquidity
Let’s take a look what brokers in diff industries are charging their clients:

TABLE: Brokerage Commissions
Art 50% to 75%
Real Estate 1% to 10%
Domain names 5% to 20%
Stocks $10
Derivatives $0.50
Futures $0.10

Let’s compare the sale of a $1 million stock or bond position (same CUSIP) with the comparable size sale of a domain at the TRAFFIC auction: 0.0001% vs. 20%. This is friction.
Friction costs: Costs, both implied and direct, associated with a transaction. Such costs include time, effort, money, and associated tax effects of gathering information and making a transaction.

When friction is high then market liquidity is low. High friction prevents the markets from developing and renders markets inefficient. In our Brokerage Commissions Table Art industry corresponds to the lowest liquidity, and derivatives & futures markets are the most liquid.

It is imperative for the DN industry to become a viable force is to develop exchanges with the minimal or no cost. The 10% to 25% fees domainers pay to SEDO or TRAFFIC are not just obscene. Those fees indicate to the other markets that the DN business is a clubhouse with backroom dealings.

Suggested topics:

Portfolio Pricing

Portfolio Duration

New Trends in Domaining


Portfolio Optimization

IDN Revolution


Distressed Situations

How ICANN Affects Prices

Government Regulations

Legal Issues


Please email to
No attachments please. All emails with attachments will be discarded


Level 10
Mar 28, 2007
Reaction score
Feedback: 226 / 0 / 0
I thought your paper was actually very funny in some parts and somewhat informative. Didn't have a chance to read the whole thing atm.

What was your purpose in writing this paper?



DNForum Moderator
DNF Staff
Sep 4, 2002
Reaction score
Feedback: 166 / 0 / 0
not bad

now factor in "cost of acquisition" above reg fee


Level 1
Mar 17, 2008
Reaction score
Feedback: 0 / 0 / 0
I have been working on a business model for Internet Domain Portfolios and think I have a good working model. It's almost complete. <no promotions>
Last edited by a moderator:


Level 8
Jul 10, 2007
Reaction score
Feedback: 10 / 0 / 0

What exactly are you asking?
There are many diff way of approaching the problem.
The 'correct' way is to use the 'synthetic instruments', assume that the
"cost of acquisition" was covered by the loan (LIBOR+, say, 2%) + lots of other assumptions.
Mathematically - these are trivial problems.
I do have a problem, though, pasting simple 2-D tables here. Any suggestion?


Level 20
Feb 28, 2004
Reaction score
Feedback: 723 / 0 / 0
I liked reading it, entertaining. Not so sure about the mathematics involved ( you already said smoking and drinking was involved so I won't argue with that ).

Perhaps the scientific model can be presented with a more strict formula? Not my forte, just a suggestion.

Nice work regardless :)


DNForum Moderator
DNF Staff
Sep 4, 2002
Reaction score
Feedback: 166 / 0 / 0

What exactly are you asking?
There are many diff way of approaching the problem.
The 'correct' way is to use the 'synthetic instruments', assume that the
"cost of acquisition" was covered by the loan (LIBOR+, say, 2%) + lots of other assumptions.
Mathematically - these are trivial problems.
I do have a problem, though, pasting simple 2-D tables here. Any suggestion?

i was just throwing in another variable

as is, i interpreted that your data is based on 'reg-fee', as being the COA
(cost of acquisition)

image pasting may be disabled in this section, as we have had spammers post here in the past.


Level 8
Jul 10, 2007
Reaction score
Feedback: 10 / 0 / 0
Got it. Thanks.
Meanwhile, I set up a site

< no promotions>
so all kinds of stuff including tables could be resolved
Last edited by a moderator:


DNForum Moderator
DNF Staff
Sep 4, 2002
Reaction score
Feedback: 166 / 0 / 0

had to remove your link in post, as no promotions are allowed here.

you can however, add "one" link in your signature.


Level 8
Jul 10, 2007
Reaction score
Feedback: 10 / 0 / 0
Under the name chessmaniac I started a new subtopic at wiki titled, you guessed it right,
Domain Names Portfolio Management Theory

Probably, it's a better venue to pursue the topic.

i got several requests to estimate DN values based on the current & projected revenues. Actually, this is the kind of effort that the DNF community can do communaly, i guess.
Here's the write up:

Portfolio Valuation

Presently there are literally hundreds of models for asset valuation. Asset could be any instrument: security (bond, stock, CD, domain name, derivative, variable rate annuity, lottery payments, etc.) or portfolio of securities. The simplest models use deterministic values for all the variables. The more sophisticated models use the stochastic approach, where variables like interest rateby the term structure of interest rate, which is beyond the scope of this paper.

Present value is the value on a given date of a future payment or series of future payments, discounted to reflect the time value of money and other factors such as investment risk. Present value calculations are widely used in business and economics to provide a means to compare cash flows at different times on a meaningful "like to like" basis.

The most commonly applied model of the time value of money is compound interest. To someone who has the opportunity to invest an amount of money C for t years at a rate of interest of i% (where interest of "7 percent" is expressed fully as 0.07) compounded annually, the present value of the receipt of C, t years in the future is:

The expression (1 + i)−t enters almost all calculations of present value. Where the interest rate is expected to be different over the term of the investment, different values for i may be included; an investment over a two year period would then have PV (Present Value) of:

Present value is additive. The present value of a bundle of cash flows is the sum of each one's present value. In fact, the present value of a cashflow at a constant interest rate is mathematically the same as the Laplace transform of that cashflow evaluated with the transform variable (usually denoted "s") equal to the interest rate. For discrete time, where payments are separated by large time periods, the transform reduces to a sum, but when payments are ongoing on an almost continual basis, the mathematics of continuous functions can be used as an approximation.

I'll set up another table in a few days


Level 4
Feb 25, 2008
Reaction score
Feedback: 0 / 0 / 0
I liked it thanks :). Although, could yo fix those images?

My two questions:

-What's "DN"?

-Are info. really that expensive to renew? I didn't know that.


Level 8
Jul 10, 2007
Reaction score
Feedback: 10 / 0 / 0
DN - Domain Name
.info - can be registered for $0.99, re-new for $8
images can't be fixed. See Adam's comment on that.

You can see the full text + images on my site. Just click on it.
Not open for further replies.
Top Bottom