In the first part of this two part article, I challenged 15 domain appraisal firms to put a dollar value on one domain name, LiquidTrading.com. Their valuations ranged from $200 to $160,000. In the second and final part of this article, I will suggest some ways in which appraisal firms might look to improve their appraisals, so that one name does not support such a dramatic range of valuations.
Since the domain name appraisal industry is so young, perhaps there are other industries that it can look to as a way forward. Real estate maybe? Domain names are often described as ‘virtual real estate’. But is there really any similarity between real estate and its virtual counterpart?
- Real estate exists at a physical location, and there are likely to be many similar properties nearby, some of which will have recently changed hands. This makes real estate valuation a relatively easy task. Domain names are by their nature, unique, and any attempt to locate ‘similar’ names is laden with traps and pitfalls.
- Real estate has a cost associated with building it, which gets built into the valuation figure. Domain names can be registered for $15 in 5 minutes flat, so have no essential value. They are intellectual property in its purest form.
So it seems that any similarity between domain names and real estate is superficial at best. Perhaps domain names should be compared to something more individual; works of art for example? Here the similarities seem more striking:
- Pieces of art are necessarily unique, as are domain names.
- Paintings do not exist at any particular physical location, neither do domain names.
- Neither domain names nor paintings can be valued in terms of ‘cost to build’.
- Domain names have objective ‘physical’ characteristics which can be used to help value them e.g. length, number of words, spelling, etc. With painting, the identity of the painter goes along way towards valuing the work.
So there are some distinct similarities, but let’s not fool ourselves, domain names are not paintings.
Improving the Appraisal Industry’s Validity
Domain Names are a unique piece of intellectual property born of the Internet age, and must be treated as such. As one web site puts it very nicely, "domain names are a new asset class". So the domain name industry must develop its methods of valuing names, and only draw on other industries where appropriate. Here are a few suggestions which might help improve the consistency and credibility of domain name appraisals, both in the short and the long term.
1. Create a Statistically Valid Sales Database
The single biggest problem with the appraisal industry at the moment is that reliable statistics for sales of domain names are very hard to come by. Every market relies to some extent on knowing previous sales values to predict future sales values. The art market only knows how to value the next Van Gogh sale because they knew what the last one sold for.
However, there are several problems with the sales databases in use by domain appraisal firms today:
- Most databases are very small, and this makes them statistically error-prone. Again, if you really want to know the value of Football vs. Basketball names using sales values, knowing the value of three transactions is simply too small a sample – the likely error rate for such a sample size is huge. In order to reduce the error size to an acceptable range (+/- 5%), a sample size should be 500+ similar names.
- Most sales values are out-of-date. On the Internet, things change very quickly, and that includes sale prices. Is the sale price from 2 years ago for DomainName.com relevant to its value today?
- The concept of ‘similar’ names is a very difficult one for domain names. They do not have a physical locale like real estate and there are no ZIP codes for domain names. Domain names have no ‘authors’, so you cannot say, "this is a Van Gogh", and use that to help assess similarity. For some names, similarity is a great help. For instance, if you have a recent sales value for the name FootballScene.com, and you also know the relative value of football names vs. basketball names, then it is possible to put a value on FootballScene.com. Even then, you would have to know whether the sale price for BasketballScene.com was a fair one, whether either buyer or seller was unduly motivated etc.
- The prices need to be ‘normalized’. For example, if the purchaser of a domain name is known to be a reseller, then it is fairly clear that the name has not been bought at market value, as most resellers will not pay more than around 25% of market value for a name. Again, trying to identify the type of buyer is a difficult proposition, but without such normalization, the validity of any database must be called into question.
- They do not contain a good ‘spread’ of names. Most names will come from Afternic.com, a domain name auction site favored by resellers. Sales at this site tend to be low-cap (below $5000). Other sale values will come from GreatDomains.com who regularly publish details of large-cap sales. However, in many ways, the most important sales to register are the mid-cap sales, where private individuals sell names to corporations. These statistics are very difficult to collect.
So ideally, appraisal firms should share their sales databases. More realistically, new firms will appear specializing in collecting, storing and selling domain sales information.
2. Use Real Appraisers
There is an obvious economic temptation to minimize the use of people in the appraisal process and maximize the use of appraisal ‘algorithms’. Some appraisal firms might even believe that they can do appraisals 100% without human intervention and consider this goal the holy grail of domain appraising.
In this author’s opinion, appraising domain names will always be a job for humans, with just a supporting role for technology. Domain names are examples of ‘natural language’. No computer algorithm has ever come close to understanding natural language and its infinite subtleties. In the main, any attempt to replace people with programs will lead to highly inaccurate and sometimes embarrassing valuations.
3. Use Qualified Appraisers
Just as art appraisers specialize in valuing works from certain painters, styles, or periods, so appraisers should consider specializing in categories, be it finance, sports, entertainment, or whatever. Ideally, they should have some professional qualifications or at least some real experience in the fields in which they specialize.
Also, what about all the different domains? We already have the three top-level domains, .com, .net, .org, plus the adopted global TLDs, such as .TV, .cc, .fm, .am, .to, .ws, not to mention the country domains. Can any single appraiser expect to know the value of all these different domain markets? Probably not. Again, some degree of specialization looks to be the appropriate answer.
4. Be Neutral
It should be obvious that appraisal firms should aim to be a neutral party sitting between buyers and sellers. But in today’s market is this really the case? Many appraisal firms have been set up by domain speculators and brokers who tend to have a financial interest in maximizing sales figures. For instance, if a domain broker makes 10% on all domain sales and acts as a domain appraiser, it seems fairly obvious that they have a vested interest in appraising names as highly as possible.
5. Form an industry body to set up and police standards
Without an industry body, any drive to standardization is doomed to failure. It is in the best interest of the appraisal firms, particularly the smaller ones, to get together and create a non-profit organization, whose job would be to work out a code of practice for members. In return, members would be able to display a special logo on their site.
All industries develop standards as they mature. The question is this: From where do these standards come? If no attempt is made to create a common set of standards, then ‘de facto’ standards will rule, which will naturally come from the few big domain appraisal companies. If the smaller appraisal firms want to be part of the process of defining how domain names are to be appraised, they need to get together and form an organization and really work towards implementing the ideas.
Wrapping Your Head Around Python
Introduction to Swift
Jump Start Git
Level Up Your Web Apps With Go
Jump Start MySQL
Jump Start Git
- 1 SitePoint/Flippa Hack Day: Hacking our First IoT Project
- 2 Learn to Create D3.js Data Visualizations by Example
- 3 5 Things Every Designer Needs to Know About Conversational UIs
- 4 How Pointer Events Will Make Cross-Browser Touch Support Easy
- 5 Bringing Componentization to the Web: An Overview of Web Components