As has been mention, all appraisals that currently exist are flawed. Even in the best possible situation, each domain name is unique, and it is not reasonable to expect any instrument to work well with all names, or even at all.
I think it is important to realize that the appraisal instruments that we have today work on several different principles. For example,
- Some are algorithms based on secondary indicators of worth such as number of exact registered extensions, total domain count for the term, extension, length, etc. These will always return the same values unless there is an algorithm change. Graen would be one example of this kind of instrument.
- Algorithmic based, as in 1, but include consideration of search characteristics of the term (search volume, CPC) as well as other characteristics. Estibot is an example.
- An algorithm that places significant emphasis on sales of the terms previously. GD Valuator is perhaps the best example here. They split up the domain name and look at how those terms sold previously, the 'worth of term' approach. Now they also consider other aspects, of course. The appraisal will yield a consistent response over short to medium periods, but over longer time, as new sales data added or algorithm tweaked, it can change.
- An algorithm that looks at many factors, and that has a particular emphasis on domain names as a company brand. It will look at things like the core terms, have they sold as brandables before and the sell-through rate, the number of characters, number of syllables, and many other factors. Atom Appraisal is the best example of this, but BrandPa, NameWorth, and others are also in this category.
- A true AI based instrument, such as OceanfrontDomains or one of the two instruments at agent.ai. These will place significant emphasis on creating a selling case for the domain name. Their evaluation of worth will vary from query to query to varying degrees. The current crop seem prone to state comparator sales that may not be valid, or even other aspects that are wrong. They probably are best at a comprehensive view of the ways that a domain name can be used, and best use case, but may fail in other aspects.
I think
which is best depends on the type of name and the purpose for your evaluation. Despite their serious flaws, the AI instruments are great for getting a handle on the ways a domain name could be used. If you are evaluating a product or service match domain name, an instrument that considers search volume is important. But for a created brandable, those instruments will be inappropriate. The extensive sales data that GD has can help make their instrument strong when the focus is on previous sales of similar names, but again each sale is unique and depends on situation.
You probably should
not depend on the number provided by any instrument, but they can be helpful in sorting to determine which names warrant a closer look. If you do really want a number, it probably makes sense to use several instruments appropriate to that type of name, and take the low to high as the range, with the worth somewhere in that. Almost always that range will be huge, from hundreds to tens of thousands, though, so not really helpful.
-Bob
️ the domain SOLD price varies according to buyers interest, high profit may come after waiting...
