new gtlds The Potential For Big Jumps in New Extension Prices

[Bob Hawkes]

For some time I have been bothered by statements that the worth of a domain name is 'reg fee' simply on the basis of a limited (or nonexistent) sales record in that extension. Since I have been closely following the ngTLD sales for about a year and a half I have constantly seen cases where a high value sale represents a big jump from previous sales.

That happened again in the most recent (Sept 17) NameBio daily report, where smile in the direct extension sold for $7018 on Sedo. There had, before this, only been ever 4 sales in the extension, and the previous one was almost two years earlier and just $111 (can see the list here).

Now I know that the ngTLD sceptics will jump in and say it is a fluke and the purchaser paid too much (see I said it for you so you don't need to! ). That is a possibility, I readily admit, but I think if you take a multi-faceted detailed appraisal look that the price of the smile sale is about right.

I do such an analysis in this new post on my blog.

I also addressed the question of whether this was the only time a big jump happened. Some of the more dramatic ones I list below (there are many others, but I think these are the biggest jumps)

• When home.loans sold for $500,000, the previously highest sale in loans was just $7500 and there were only 4 previous NameBio listed sales at any value in the extension.
• When autism.rocks sold for $100,000 on Sedo in 2015 there had not been a sale at all previously in the extension.
• Vacation.rentals sold in late 2017 for $500,300. The previous highest sale in that extension was just over $1500, and even today there are only 5 other sales in the extension listed on NameBio.
• The .online extension has sold many times, but when casino.online sold for $201,500 in March 2017 the previous highest sale in the extension was just over $16,000 (there have been some higher sales in the extension since them, but not nearly at the casino level).
• The great domain name video.games sold for $183,000 in 2017. The next highest sale in the extension is only $1101.

Thanks for reading!

Bob

 

[NYJimbo]

anomaly

a·nom·a·ly
noun
"something that deviates from what is standard, normal, or expected.​

You are citing anomalies in the sales of domains in new extensions. That is usually because of very strong word pairings or keywords with new TLD's.

Thanks for reading!

 

[Bob Hawkes]

But when dozens and dozens of anomalies, I wonder if we should really view them that way. I agree certainly that the odds of a huge sale, at any time, in anything including com, is low. But there are dozens of cases of order of magnitude jumps in ngTLD sales, so I am resistant to calling them all anomalies (although totally agree with you that it is arguable to do so).

Not sure if you read the full post, but I point out that this is not just in ngTLDs. One example I use is the summer sale of the sleeping (.) com domain name. If prior to that sale I look on NameBio at previous sales using the word "sleep" anywhere there are 251 sales, mainly in .com. But the total value of all 251 COMBINED is less than this one sale.

My argument is not to discount previous sales as one measure in an evaluation. But to over-simplify on only it is misleading. Thanks for your input.

 

[NYJimbo]

Not sure if you read the full post, but I point out that this is not just in ngTLDs. .

I try to dude, but a lot of your stuff is "TL;DR"

Not a dig against you, but sometimes your posts go on for 2 pages and its a lot to take in.

Regardless, I stand by my post that most of what you are seeing is solid word/TLD combos that skew analytics.

 

[bmugford]

Outlier

An outlier is an observation that lies an abnormal distance from other values in a random sample from a population. In a sense, this definition leaves it up to the analyst (or a consensus process) to decide what will be considered abnormal. Before abnormal observations can be singled out, it is necessary to characterize normal observations.

 

[Bob Hawkes]

I know I write at too much length! Sorry (comes from wanting to justify and give assumptions)

I think this excerpt summarizes my main point, and I think we agree solid word/TLD combos, just want people not to look only at past sales and extension.

"It is important to take a multi-faceted approach to any domain name valuation, looking at the strength of the term, previous sales in the term, previous sales in the extension, search and advertiser statistics, the aesthetics of the name, competition with similar names, and the pool of potential end users, among other factors."
​

from post here

 

[NYJimbo]

I know I write at too much length! Sorry (comes from wanting to justify and give assumptions)

I am not trying to diminish the value of your postings, its just that sometimes I cant read that much in one response and go all "Man, this post is just TL;DR".

 

[Bob Hawkes]

I certainly agree that ANY large sale is an outlier, although as @bmugford wisely and concisely reminds us you need to define what is normal before being able to decide what is abnormal.

I am not at all arguing against using past sales data. I spend at least an hour of every day on NameBio (I know, I should get a life). But past sales, especially past sales in extension, are not the only or necessarily even the best predictor of future sales.

 

[bmugford]

You are pointing out are a handful of sales that are way outside the norm. That is like the definition of an anomaly or outlier.

Legacy extensions have sales in every price range. It is not like there is one sale of $500,000 than the next closest is $1,500.

That is far more impressive than some random, isolated registry sales. All it takes is one buyer, or sucker, in some cases. These sales are not repeatable, or really that relevant to a domain investor.

Brad

 

[Bob Hawkes]

All you are pointing out are a handful of sales that are way outside the norm. That is like the definition of an anomaly or outlier.

Legacy extensions have sales in every price range. It is not like there is one sale of $500,000 than the next closest is $1,500.

That is far more impressive than some random, isolated registry sales. All it takes is one buyer, or sucker, in some cases. These sales are not repeatable, or really that relevant to a domain investor.

Brad

I totally accept that com the normal/abnormal is well defined since there are so much sales data. For even the top selling ngTLDs like top there is not nearly as much, so to some degree in most extensions there is no normal.

I am not sure if you noticed that I did talk about sales in com as well though. e.g. NameBio sales experience would perhaps suggest a 'sleep' name ending in 'ing' would perhaps sell for much less than it did, but a more nuanced look at the market and other factors arguably support the price.

"when Buckley Media Group sold sleeping.com summer 2018 for $502,225 that price exceeded the combined sales of all 251 previous NameBio listed sales that include the term 'sleep' anywhere (the next highest sale price was $14,888)."
​

(from my post, at very end)

Bob

 

[Bob Hawkes]

By the way right after the NameBio daily report came out I checked with GoValue and Estibot on their valuations of the domain (hopefully fast enough that they had not adjusted their values). For what it is worth, or not, here data for smile(,)direct:

Sales Price: $7018
GoValue: $2456
Estibot: $7200

 

[bmugford]

I totally accept that com the normal/abnormal is well defined since there are so much sales data. For even the top selling ngTLDs like top there is not nearly as much, so to some degree in most extensions there is no normal.

I am not sure if you noticed that I did talk about sales in com as well though. e.g. NameBio sales experience would perhaps suggest a 'sleep' name ending in 'ing' would perhaps sell for much less than it did, but a more nuanced look at the market and other factors arguably support the price.

"when Buckley Media Group sold sleeping.com summer 2018 for $502,225 that price exceeded the combined sales of all 251 previous NameBio listed sales that include the term 'sleep' anywhere (the next highest sale price was $14,888)."
​

(from my post, at very end)

Bob

I would say Sleeping.com sold for far higher than I would have expected, however there are many .COM sales for mid $XXX,XXX. So seeing a single word .COM sell for mid $XXX,XXX is not that far outside the norm IMO.

Brad

 

[deez007]

@MetBob ...spot on!

Awesome blog post as well.

 

[Bob Hawkes]

I realize in retrospect that by giving the examples people are naturally focussing on them rather than my central point of why you should use multiple ways to evaluate a domain name worth. I summarize that part of the post below:

• One component, but only one, is to look at previous sales in the extension.
• A second question is to look at previous sales in the term.
• A third aspect to evaluate is how popular the term is in searches and for advertisers.
• Possibly the most important thing to evaluate is how many potential users are there for the domain name.
• An additional aspect that needs to enter into the valuation is the inherent quality in the domain name. This is a matter of the intersection of aesthetics and marketing, and cannot be evaluated in an exact way.
• For ngTLDs, a key consideration is how well the term and extension match.
• How much competition is there for among similar domain names?
• While it should be well down your priority list, I do believe that it is always worthwhile to look at the automated estimates of domain worth.
• Does it have special value to certain potential end users?

It is longer on each point at the blog but does take you through the example with the specific domain name on doing each of those factors . If you want the long version, go here.

 

[JB Lions]

For some time I have been bothered by statements that the worth of a domain name is 'reg fee' simply on the basis of a limited (or nonexistent) sales record in that extension. Since I have been closely following the ngTLD sales for about a year and a half I have constantly seen cases where a high value sale represents a big jump from previous sales.

That happened again in the most recent (Sept 17) NameBio daily report, where smile in the direct extension sold for $7018 on Sedo. There had, before this, only been ever 4 sales in the extension, and the previous one was almost two years earlier and just $111 (can see the list here).

Now I know that the ngTLD sceptics will jump in and say it is a fluke and the purchaser paid too much (see I said it for you so you don't need to! ). That is a possibility, I readily admit, but I think if you take a multi-faceted detailed appraisal look that the price of the smile sale is about right.

I do such an analysis in this new post on my blog.

I also addressed the question of whether this was the only time a big jump happened. Some of the more dramatic ones I list below (there are many others, but I think these are the biggest jumps)

• When home.loans sold for $500,000, the previously highest sale in loans was just $7500 and there were only 4 previous NameBio listed sales at any value in the extension.
• When autism.rocks sold for $100,000 on Sedo in 2015 there had not been a sale at all previously in the extension.
• Vacation.rentals sold in late 2017 for $500,300. The previous highest sale in that extension was just over $1500, and even today there are only 5 other sales in the extension listed on NameBio.
• The .online extension has sold many times, but when casino.online sold for $201,500 in March 2017 the previous highest sale in the extension was just over $16,000 (there have been some higher sales in the extension since them, but not nearly at the casino level).
• The great domain name video.games sold for $183,000 in 2017. The next highest sale in the extension is only $1101.

Thanks for reading!

Bob

This is such a ridiculous post, addressed already, outlier. 1 reported sale in almost 2 years, then trying to make something of it. 1 sale is just that, 1 sale, not a jump. A jump would be something like when the LLLL.com prices went up awhile ago. That was pretty clear, you had hundreds/thousands of sales data to look at.

Let's just take a look at 1 you mentioned, autism.rocks that sold for $100,000 in 2015.

https://namebio.com/?s==UjM1IDM3gTM

Only 2 reported sales after that for a whopping, $100, $190. Outlier sale, just like this one.

How about the video.games that sold for $183,000 in 2017. Before that the biggest sale was $1,101. Was it a jump? Look at the sale after that $101. Outlier.

Again, these big sales are nothing but 1 person/company in the world wanting that 1 domain. Freak, outlier sales, many times by people have no clue of what domains are selling for. The buyers of that .loans or .rentals admitted that.

Somebody who is into statistics should know something about sample size.

 

[Bob Hawkes]

With due respect @JB Lions there is nothing in my post in any way claiming a statistical likelihood of a big sale. My point is that looking ONLY at past sales in an extension, without considering other factors (see above), is not a complete way to evaluate sales prospects of a domain name. All of the factors mentioned above, and others, should be considered.

Just in case you did not read the linked post this is what it concludes with.

"Now let me stress that this post does not mean that any domain name might sell for a large amount. The key issue is always the quality of the domain name. Let's not over-simplify the complex task of evaluating the worth of a domain name by saying that just because there is not a history of large value sales in an extension so none are possible."
​

Looking at only one part a post is analogous to the problem of looking only at past sales in an extension. When we over-generalize or over-simplify we may make less than optimum decisions.

 

[JB Lions]

"But a much higher sales price is fairly routine in the ngTLD world."

"I have constantly seen cases where a high value sale represents a big jump from previous sales."

It's fairly routine in every extension.

It's also fairly routine for a much lower sales price, see the examples I posted above.

Prices are all over the map. The sample sizes in new extensions are usually too small to make any conclusions.

"But when dozens and dozens of anomalies, I wonder if we should really view them that way."

It's usually 1 or 2, or very small amount, not dozens. See video.games. See autism.rocks. 1 in those extensions.

Sales over $100,000, using your examples:
rentals - 1- next closest $1,584
rocks - 1- next closest $190
games - 1 - next closest $1,101
loans - 1 - next closest $7,500
online - 1 - next closest $19,500

You used 5 different examples in your first post, only 1 sale over $100,000 for each of those extensions, the next closest sale, not even in the same neighborhood. All the examples you just used, 1 freak sale.

"Now I know that the ngTLD sceptics will jump in and say it is a fluke and the purchaser paid too much"

Because it's true, that's what the data tells us, that's what the examples you just used tell us.

 

[Domains - Wanted]

One other thing to add here about such outliers, flukes, no matter the extension... some of these may be financial transactions between business partners where the domain name was only an instrument, rather than the object of sale

Speaking from personal experience here, BTW. Due to their inherent valuation ambiguity, domains are a convenient vehicle for all manner of financial transactions. Seems I had been a little conservative in my dealings way back when (late 90's). Then again, had no support from such convenient Namebio listings back then

Great food for thought here, thanks @MetBob

 

[Bob Hawkes]

I totally agree that ngTLD prices are all over the map. By the way the "much higher" in the line you extracted, probably not clear from the short context, is than previous sales in that extension, not much higher than domain prices in general.

It is certainly true that the high price sales are dominated by com and will stay that way for the foreseeable future.

If one WERE to talk statistics, I think the only ngTLD extension that currently has a consistent and significant long term record of major sales is TOP (and it sells mainly only in one geographical region)
So far 2018 YTD on NameBio 435 sales with an average price of $4569
3 sales above $100k 2018YTD
28 sales above $10k 2018YTD
287 sales above $1K 2018YTD

A few others like CLUB have some long term record of significant sales. 2018 YTD on NameBio 73 sales with an average price of $5878.
1 sales above $100k 2018YTD
3 sales above $10k 2018YTD
37 sales above $1K 2018YTD

In the past XYZ had big value sales, and still occasionally. The GLOBAL registry sales have high average price, but not much reseller sales in that extension. ONLINE has a few large sales. The rest are largely scattered with as you note one or two big sales, and a smattering of small ones.

When I do my monthly analysis of ngTLD sales it is true that there are only a few large value sales. For example in the most recent report there were only 12 of the 167 sales in the 30 day period that were above $10k in value, and 69 that were $1k or more. It is still about a week before I will produce the next report, but I suspect there may be fewer large sales in it, unless a lot next few days.

 

[Bob Hawkes]

I don't think it necessarily is a fluke. In some cases (like the autism one) clearly yes. In others I would argue that an analysis that looked at the importance of the field, the quality of the word, and the match with the extension could have predicted the big sale, even if not the size of the sale. The data is still sparse enough in most new extensions, and many cc too actually, that there is a lot of variation in prices.

 

[JB Lions]

I don't think it necessarily is a fluke.

You used 5 different extensions in your first post. All 5 had 1 reported sale over $100,000, the next closest sale, nowhere near that. It's the definition of a fluke/outlier/freak sale.

Again, sales over $100,000:
rentals - 1 - next closest $1,584
rocks - 1 - next closest $190
games - 1 - next closest $1,101
loans - 1 - next closest $7,500
online - 1 - next closest $19,500

 

[Bob Hawkes]

You used 5 different extensions in your first post. All 5 had 1 reported sale over $100,000, the next closest sale, nowhere near that. It's the definition of a fluke/outlier/freak sale.

I used those examples because the title of the post was "Big Jumps" . I had thought if I was going to write about cases where the values in an extension were at low prices, and then there was a high price sale, well I should include some cases where that happened.

But just in case someone thought I meant these were typical of all extensions, there are 646 new extensions where this did not happen (I just made up that number so no one quote it!)

 

[ecalc]

For some time I have been bothered by statements that the worth of a domain name is 'reg fee' simply on the basis of a limited (or nonexistent) sales record in that extension.

It’s worth noting that any given name has two distinct values, wholesale and retail, distinguished by buyer/seller motivation imbalance. Extremes occur when one party is desperate. Most names have $xxxx retail potential, music to registrant’s ears, while the same name is worth ‘reg fee’ or less wholesale including the subject smile name even after the $7k lightning strike.

 

[Kate]

Personally, I always try to consider the domain game in terms of statistical probabilities. I know that if I buy a .rocks domain, the likelihood of a sale is nil, regardless of past reported sales. A 100K sale means nothing to my bottom line. It is not actionable data.
Besides, those high-value sales in nTLDs are usually always registry sales. For domainers such sales remain elusive.

And you see outliers in ccTLDs. Not just .com & nTLDs.

 

[Bob Hawkes]

I gave your post a like @ecalc because as usual in your contributions it is clear, logical, and succinct (I struggle with succinct myself). While totally agreeing with two of your points (there are retail and wholesale - clearly the big sales mentioned are retail), and secondly to get a super high price you need a highly motivated buyer (in fact that was exactly what I discussed in the point of the full post called "Does it have special value to certain potential end users? where I even speculate on who I believe is the probably buyer of the smile name on this basis.

I do however, disagree, perhaps only moderately, that smile is just reg fee without this buyer. I think the fact that two different ngTLDs with this word smile have sold at two different venues each for more than $5000 is reason to believe that certain great matches with the word are valuable I would argue. The word smile sold in many country codes for >1000 (eu, cc, be, me, pl for example) and in pro for over $2k. While appraisals are always inexact, I would argue that there is solid support for the idea that smile in some other country codes and in new extensions where it is a great match is worth, at least retail, something in the $$$$ range.

Thanks for your contribution.