I hope to write this up more fully, but I recently completed doing a valuation on 89 domain names in my portfolio using both
Estibot and GoDaddy
GoValue (Beta). Here is what I found:
(a) Overall there now (see next point) seem
little systematic bias between them.
For the total of the 89 domain names Estibot gave $49,505 and GoValue $53, 226. In finding these totals I took the <$100 that GoValue gives to low value domains as $0 (perhaps I should have counted it as $50).
(b)
GoValue estimates are very much in flux. I rechecked today all 89 (yes, it let me) and compared to values I had taken one to three weeks earlier. In some cases they changed by as much as a factor of 4 and most changed at least a little. While in general the new GoValue estimates are lower than a few weeks ago, a few did go up (and a number stayed constant or nearly so).
(c) Estibot and GoValue
do agree remarkably on some domain names - e.g. for frugal(.)design one gave $360 and the other $364; for BCT(.)pw the estimates of $500 and $573 are close, and even robotics(.)gdn at $1400 and $1198 are pretty close.
(d) They
completely disagree on some others. For example AdvantageHappy(.)com is listed at $0 on Estibot and $1302 on GoValue, and codely(.)ca is $10 on Estibot and $1674 on GoValue, nanoAZ(.)com is $0 on Estibot and $1263 on GoValue, while XXL.science is $9100 on Estibot and only $361 on GoValue.
(e) It seems that
Estibot is pretty kind to single generic word in the .top extension (e.g. cabins(.)top at $1700, compassion(.)top at $2600)i , while
GoValue is less discriminating about extensions among the new gTLD. For example I own the genetic DNA editing acronym CRISPR in multiple extensions, and it valued most of them the same, even though some work better together (like it tells me CRISPR.science and CRISPR.trade are equal in value, but they are not in my opinion).
(f) At least for the domains in my portfolio,
Estibot seems unfairly harsh to .ca domain names and any .com that is not a single generic word. For example DIYI(.)ca which could stand for Do It Yourself Investing or various other possibilities, just four letters and the .ca extension, they value at $60, but GoValue gives an estimate of $2350! (which is too high, but not by the same factor that Estibot is too low)
(f) The information they provide makes it clear what the valuations are based on. To get a
high Estibot you need a common word or phrase that has great search statistics and has sold and most extensions are taken.
For GoValue the really important thing seems to be other extensions taken and especially that there is at least one sale in exactly that word at a large value.
(g)
Both over-estimate to some degree on average I would say. I certainly would happily sell my portfolio for much less than the overall valuation
.
Overall conclusions:
- I think both are helpful and provide guidance but clearly are not precise and sometimes make bad errors.
- I feel that if GoValue stays freely accessible it will help us selling domains when we can point to that as a third party valuation (at least when the value is much more than what we are asking).
- I like that GoValue have a category <$100 for low value domains, and don't try to estimate precisely high value domains (simply give >25,000). GoValue should round all values, as the implied precision of for example saying $1391 is not supported.
- Estibot seems to care about extensions more than GoValue, but not in a completely logical way. They are too harsh to the last wave of extensions, and too kind to things like .top that have been around for 3 years and had more time to accumulate sales.
- Be careful mentioning GoValue estimates, since they vary day by day (really!). I hope that is because they are in beta, and the changes will become less once it gets more stable.
- The comparator domain name sales that GoValue give are in general excellent, and it provides in many cases information that does not seem to be in NameBio (the huge database of the Afternic network gives GoDaddy a big advantage).
- Overall I think these two are far better than any of the other estimation bots I have found on the web. In a way, it is good that they do emphasize different things.
- If I was giving grades to them, I would give GoValue a B+ and Estibot a B, and everything else I have tried an F. I think there is room to improve in both of them.
Sorry this is so long!
Bob