Atom provides rich data for each domain it values.
This is part of Domain Name Wire’s review of automated domain appraisal tools. See all reviews: Estibot, GoDaddy, Atom, Appraise.net, GoExpired, Dynadot, NameWorth, Saw, Appraise.software, Humbleworth.
Atom is the fastest-iterating domain aftermarket in the business. How does its appraisal tool stack up?
Atom provides perhaps the most robust data backup for its appraisals among all the tools we tested. Building on its other tools to help domain investors, you see cross-TLD registration information, signals like whether the name matches an existing app, sell-through rates for the root words, and very good comparables data.
The company has been testing showing comparables on domain landing pages, which helps justify the asking price. As we’ve said before, comparables are the number one way to justify a domain’s asking price.
Let’s take a look at how Atom stacked up to the competition.
Two word brandables
We evaluated two domain names in this category: MakeMatter.com, which sold for $15,000, and PressBridge.com, which sold for $5,000.
Both of these domains are in the sweet spot for two dictionary word brandables of $3,000-$15,000. While some sellers hold out for higher amounts, their data will show these domains sell for much more. But looking at overall sale data, it’s reasonable for an appraisal system to return anything in this price range for this type of domain.
Atom appraised MakeMatter.com at $4,499 and PressBridge.com at $8,799.
Even though it valued the domain that sold for less for more than the other, recall what we pointed out in our intro post: a lot of this has to do with how the sellers priced these domains.
More importantly, Atom backed up its valuation: PressBridge is registered in more TLDs, its keyword sales data points to high STRs, and Atom lists comparables. In other words, it shows its work, and that’s the most important thing for a domain appraisal.
Atom also smartly judged that our unregistered test domain, CloudToaster.com, wasn’t worth much. It said it was worth <$1,000.
One word, high value .com
Once again, we ran Dragonfly.com as a test of high value domains.
Atom returned a valuation of “> $1 million”.
This is a good example of an appraisal tool not getting too specific and understanding its limitations. It reports all seven figure domains as merely being worth at least $1 million.
Popular ccTLDs
We ran a solid one word .io, and a plural .ai of lesser quality through Atom’s appraisal tool.
Mike sold expedite.io for $14,995. Atom valued it at $20,500
We also valued kickers.ai, which Andrew sold for $8,000. Atom valued it at $34,000.
On DNW Podcast #580 in March, Atom CEO Darpan Munjal told Andrew that a quarter of the company’s sales revenue comes from .ai domains. So we don’t want to question its valuation, especially with .ai’s trajectory. That said, Atom’s .ai appraisals for recently sold .ai domains are often much higher than the sales prices.
Exact Match descriptive
Exact match, category defining domains aren’t as popular as they used to be. These days, it’s all about the one word brandable dictionary term.
But there’s no doubt WaterFilters.com has a lot of value. It’s for sale for $3.5 million. We don’t know what it will sell for, but this is a huge business, especially for online sales. It’s a great domain.
Atom valued the domain at $66,000, which was in the middle of the pack of the tools we tested.
Three and four letter domains
Three and four letter domains are some of the most liquid domains on the market.
We tested a pronounceable CVCV .com domain, dujo.com, that is listed on Afternic for $36,000.
Our goal was to see if the algorithms caught that this was not just a random set of letters. Pronounceable, brandable four letter domains are generally worth more than unpronounceable random letters.
Atom valued it at $21,000, toward the lower end of the valuation tools we tested.
We also tested a four letter, non-pronounceable. Logan Flatt sold MOTG.com for $14,888. Atom valued it at $10,499.
These valuations seem fair. It’s good that Atom, which has focused on brandable domains, recognized the difference between these domains.
For a three letter domain, we tested VJN.com. It’s listed for sale on Afternic for $39,000. V and J aren’t great letters, so this is likely on the lower end of three letter .com values.
Atom valued the domain at $114,766. Given that the domain is listed for sale at a cheaper price and hasn’t sold, this seems high. That said, names like this do sell in the six figure range. It would have helped if Atom had shown comparables for this domain, but the appraisal stated it didn’t have any.
New TLDs
New top level domains are very hard to value because there is very little sales data for most of them.
We tried a couple of new TLDs.
Andrew sold voicemail.app for $5,000 last year. He bought it in Early Access when the TLD launched. Atom valued it at $2,799. Atom didn’t provide comps.
Mark Levine sold timber.homes for $2,899. Here again, Atom understands its limitations: it said it couldn’t value .homes domains. Frankly, given that NameBio has only 24 .homes domains in its database, we don’t know how anyone could value these domains.
Final analysis
We agree with many of the numbers Atom provided. But more importantly, Atom provides a rich dataset for each domain, supporting its valuation. It doesn’t provide comps for all domains, but when it does, this is great ammunition for negotiations.
I like it. It likes my domain names. Bunch of them > $1M.
Any of those “appraised” > $1m two words or more? If not, any of them not even words?
I would be interested in checking that one out when I get a chance, out of curiosity, and seeing how it goes. Maybe it’s at least better overall than most, I don’t know. However…
Already I see the notion of “Comparable Sales” mentioned there.
I have always considered the idea of “Comparable Sales” in domain investing (aka domaining) to be extremely *fallacious* and problematic.
Moreover, it seems to be a conceptual outgrowth of “the real estate analogy” – with the problem being it’s the wrong real estate analogy.
The right real estate analogy is commercial real estate. The problem in the industry is that so many tend to practice and be stuck in residential real estate thinking instead.
“Comparable Sales” really has little and closer to no true applicability when it comes to “commercial assets” like domain names, most especially good and valuable domain names.
————————————————————————————————
That said will also just add what I wrote at Sully’s recently again here too for the topic at large:
May 8, 2026 at 05:25 PM
The “automated” tools for “appraisal” are generally garbage, and harmful to the industry.
Most especially bad is when they are put next to your listed domains whether you want them there or not. Plenty of that going on.
Let’s take Crypto.com for example. Even at the height of Bitcoin being on top of the world, Estibot was declaring Crypto.com to be worth $48,000. Even after Crypto.com sold and people knew it was at or near $12,000,000 – Estibot was still saying for a long time after it was worth $48,000. Since they require extra steps to do a quick check, I haven’t checked to see if they finally revised the figure to something in the same universe since.
I know a .com that had a real offer of just under $7,000,000 on it in the late 90’s. It’s normally “appraised” by the likes of Estibot at about $1,400 give or take. One of the most famous and iconic phrases ever, worldwide.
Personally, the only “appraisal” tool I have ever found to be both “honest” and brilliant is Oceanfront domains. AI can never be conscious, self-aware, or have any actual mind or sould, hence the “quotes.” Nonetheless, surprisingly “honest,” and brilliant.
Is that merely because I like what it said? That’s obviously a fair question and it’s fair to hold cynical thoughts about that. But I can tell you before God it’s not that. It’s my honest and true conviction – that the service has been both genuinely (and finally after all these years) “honest” and “intelligent.”
I haven’t used it in months though, so I also know these things can change, and I have mentioned that before. People and things can become corrupted, especially when aggressive people with an agenda get involved or worm their way in.
Well I could say more but that will do.
I’ve been commenting in the blogs for many years now as “John,” but only just recently changed my screen name to the above. Some of the other major bloggers are already long and well familiar with me. I also used to use a different email, but this is the one I’ve been using for a while now if you all compared notes. As I’ve mentioned before, I’ve become 99% end user and only ~1% domainer for many years now, fyi.
It’s funny you should mention commercial real estate. Before working on NameBio I was a market researcher for CoStar, which tracks sales and leases of commercial real estate.
CRE is absolutely valued based on comparable sales. I don’t know where you got the idea that it isn’t, but you’re wrong. It’s just the attributes that make one property comparable to another are slightly different than residential real estate.
Comparable sales are how you price any unique and valuable property, whether that is real estate, vintage cars, art, antiques, sports memorabilia, domain names, etc.
Case in point, CoStar owns Comps.com and ComparableSales.com. I even sold them the latter. Odd thing for a commercial real estate data provider to own if CRE isn’t valued based on comps like you claim.
You can certainly argue the point, just as one can argue anything, but with physical “non-digital” real estate, residential or commercial, analogies can only go so far. If what you say is accurate, there is still a certain dimension of “fungibility” that does not apply to “digital” or “electronic” real estate at all.
If you have a shopping center property on one side of Route 1 in Virginia just outside of DC, for instance, and another one directly across the street from it, generally and absent any truly compelling difference the two are going to be essentially “fungible.” Even if one of them isn’t yet set up for getting Krispy Kreme donuts hot off the press, which are super bad and harmful for you and should be avoided to begin with.
Not so with the “commercial real estate” of domain names. The uniqueness factor of domain names is not just a cliche, but a critical component that can make what some wish to assert or what some automated blob of code wishes to assert to be “comparable” anything but.
Each domain, as a potential or actual commercial asset and means of commercial revenue, must be evaluated on its own commercial merits. One may appear to be just another shopping center property on Route 1 across the street, but in reality be a sprawling mall property in the very heart of NYC’s Manhattan or Florida’s Miami.