analysis Price Breakdown of .COM Sales (2018)

Domain name sales of value $100,000 or more represented only 0.10% of all publicly-recorded .com sales in 2018, but they accounted for almost 30% of the dollar volume of sales for the year. While average sales prices can be useful, they mask the fact that most sales are at much more modest prices. The typical .com sale, at least as publicly recorded, was about $265 in 2018.

Study

I used the NameBio searchable database for all 2018 .com sales of $100 and greater. While I considered including all domain extensions, it seemed wise to restrict the analysis to the more homogenous dataset of just .com sales. I used the full calendar year 2018, rather than say the most recent 12 months, to have a dataset that is pretty stable from additions and deletions. The .com sales for 2018 include 69,896 domain sales.

Each line in the table includes sales of that dollar volume or more. For example, there were 129 sales at prices of at least $50,000 accounting for a total sales volume of $29.3 million. Those constituted only 0.18% of the total number of sales but 33.4% of the total dollar volume of sales.

Key Points

Here are highlights from the analysis.

• The top 0.1% of .com sales represent 29% of the sales dollar volume for the year.
• Only 16% of .com sales are at prices of $1000 or more.
• Almost 69% of sales occur at prices of $500 or less (even when we exclude all sales less than $100).
• Half of all .com domain sales in 2018 (>$100) were at sales prices of less than $265.
• For the same year, the average .com price was $1256. The average price is highly influenced by a small percentage of high-value sales.

Interestingly, the median .com price of $265 in 2018 dropped significantly to only $227 during 2019 year-to-date. I am not sure of the reasons for this change, but it appears to be statistically significant.

Keep in Mind

It should be stressed that not all venues report to NameBio. Sales that take place at DAN, Afternic, Efty sites, smaller Sedo sales, and most brandable marketplace sales are not included, nor are many private sales. Some time ago, NameBio outlined in a NamePros thread the sales venues that are included in the database. Also keep in mind that the database includes both wholesale transactions between domain investors and retail sales to end users.

It is hard to predict the potential bias due to the sales venues represented in the database. While many high-value sales are not publicly reported, there is also a bias against lower-priced sales based on the Sedo sales reporting and the venues that do not report to NameBio at all.

Small Value Sales

The NameBio subscription plans allow one to look at sales of less than $100. I took a look at how including them would change the results. In 2018, there were 168,600 .com domain sales at prices of less than $100, with an average price of $25. If we lump together sales of all value, there were in total about 238,500 .com sales in 2018, but the median price was just under $24. The majority of .com sales are between domain investors, or sales from expired domain auctions, and at very modest prices.

Final Thoughts

It is clear that a small number of high-value domain sales account for most of the sales volume pie. The vast majority of domain sales are at much more modest prices, however.

This analysis provides guidance of the pricing levels that are accepted by most domain buyers and account for the vast majority of domain sales.

While the average sales price is an important metric, keep in mind how much it is influenced by the few high-value sales. For example, in 2019 so far, the average sales price of $1459 drops to only $925 if the single $30 million sale of voice.com is removed. The top three sales of 2019 represent more than 41% of the total sales volume at this point in the current year. This means that while my earlier post showed that the first half of 2019 was strong in both numbers and average prices, the latter would change dramatically had a few large-value sales not taken place.

The drop in median domain prices from 2018 to 2019 is potentially important, and I plan to look into median price trends in a future post.

—

Thanks to @Michael Sumner, CEO of NameBio, for creating and maintaining the domain sales resource that makes analyses such as this one possible!

 

[Keith DeBoer]

While I appreciate the effort, I don't see much practical value in these stats because, as you point out, these sales are a mix of wholesale and retail. In fact I would say, that 80% or more of the sales recorded in NameBio are wholesale sales. Because of the mix, its hard to find insight into either wholesale or retail pricing from this analysis. That said, I appreciate all your articles, Bob and hope you will keep up the good work!

 

[WatchDogue]

Thanks Bob, for an insightful and well documented price breakdown of .COM Sales in 2018!

 

[Recons.Com]

While I appreciate the attempt, the analysis is misleading.

You attempt to exclude the outliers in $100K+ category, while you include the purchase price of domains as the sale price.

Majority of low $xxx are simply auction wins by resellers from the registrars/dropcatchers.

Here is an illustration of the low usefulness of this: if NameBio decided to report $50 sales, you'd suddenly now come to conclusion that the average price of domain sales in 2018 was under $200. And then if you'd include all closeout sales from Godaddy, oh miracle, your average price is well below $100.

 

[Ivan Rasskazov]

Thank you for the post. I would mark this point down for the future and revise for any sales that are discovered through public statements (often by George Kirikos, Jamie Zoch also notes possible acquisitions which can be tracked through domainIQ). This way we can go back and see how the numbers may change with discovery of additional data. Feel free to look through GGRG's reports as well, which we help put together, as Escrow and ShortNames contribute key sales data there.

- Sorry for any grammar errors, on the road at the moment -

 

[The Durfer]

Ty

 

[Donnyd]

If you had private sales ones that where NDA. You would have figures over 50% in the five and six figures. No one gets an nda or cares about sales below 10k.

 

[NameSpirit.com]

Thank you for taking the time to analyze all of that data Bob. I strongly believe reading your posts has lead to at least one sale for me already as I revised my prices.

Thanks again!

 

[Karluch]

Thank you. Appreciate you taking the time and doing the research and sharing.

 

[Bob Hawkes]

Because of the mix, its hard to find insight into either wholesale or retail pricing from this analysis.

Your point is well taken @Keith DeBoer if you are wanting to look at only retail or only wholesale sales. I think a fair number of domainers actually do sales both to end users and to other domainers and looking at the total dataset may (keeping in mind reporting biases) be relevant for those who do both retail and wholesale sales.

You attempt to exclude the outliers in $100K+ category

Sorry I don't understand this. No data was excluded that is in the NameBio ($100+) database for the year.

Thank you for the suggestions, @Ivan Rasskazov. I do follow both George Kirikos and Jamie Zoch on Twitter and admire the great work they and others do in sleuthing out unannounced major sales. While everyone is interested in the highest value sales, and they get lots of attention, if the sales do not end up getting in the NameBio database I am hesitant to use them in an analysis. I think there is much to be said for using a publicly accessible database so that anyone could check/replicate, or do their own extended analysis. Once one starts adding additional sales that did not pass the requirements for NameBio inclusion one opens any analysis to criticism why some were added and not others, and make it impossible for someone else to check or do further analysis on the same dataset.

The issue of dividing out wholesale and retail is both important and challenging. While one can exclude certain venues, or exclude certain price ranges, neither is a perfect or even very good divider. We saw this week even $1 million sale can be to another domainer. It is clear that the sales below $25 are virtually completely domainer purchases from expired auctions. If you look by prices, other than that, the mix of retail/wholesale while undoubtedly more wholesale at lower prices does not seem to have a clear demarcation. Keith may well be right that it is 80% wholesale, but I think there is a pretty high percentage wholesale even at sales well past mid $$$. And some of the sales at low to medium $$$ are indeed end user retail ones.

Anyway my main point is that the sales volume is really concentrated in the small number of high value sales, and that these high value sales have a big impact on average prices. If one looks at all NameBio recorded sales (admittedly some between domainers or to domainers) a large number are at modest levels. Since the NameBio stats give average, but not median (they did for a little while but it was slowing up results too much is why it was dropped) I think there is some lack of appreciation for how many sales are happening at modest price levels.

Finally thank you everyone for reading the post and for your comments and encouragement.

Bob

 

[equity78]

While everyone is interested in the highest value sales, and they get lots of attention, if the sales do not end up getting in the NameBio database I am hesitant to use them in an analysis. I think there is much to be said for using a publicly accessible database so that anyone could check/replicate, or do their own extended analysis. Once one starts adding additional sales that did not pass the requirements for NameBio inclusion one opens any analysis to criticism why some were added and not others, and make it impossible for someone else to check or do further analysis on the same dataset.



Bob

Agreed Bob, many a time a big sale when delving into public filings can show that a site or intellectual property was tied to the sale. If it's not on Namebio I would not include it.

 

[Recons.Com]

Sorry I don't understand this. No data was excluded that is in the NameBio ($100+) database for the year.

Anything about the rest in my post? The confusion created by mixing in purchases with sales?

Basically, it would be the same as reporting fashion sales for Macy's which features both sales to customers and what they paid to Ralph Lauren, Tommy Hilfiger and others to purchase those items and then claiming that the average of all that is the average sale price for clothes.

 

[karmaco]

I think this data is useful in knowing what to price your names at if you are looking to sell to fellow investors which is useful to domainers who like fast flips.

I agree this is largely wholesale pricing we are looking at is my point. I don’t apply limited data to end users. End users are not the main buyers being reported. Domainers are.

So for actual end user pricing, we really have no specifics or rule of thumb aside from the ones we decide personally when approached about a domain. We also have this forum to ask for help with questions on pricing to end users.

I think one of the bigger takes from this type of data is that a large portion of domainers are predominantly fast flippers and sell under 1K.

 

[ksusha64]

Great article. Thanks!

 

[Bob Hawkes]

Anything about the rest in my post?

My apologies, while I had not tagged you I had intended the third paragraph of my reply making reference to the difficulty of defining a dividing line between retail and wholesale as a response. I should have made it clear I was also responding to your point.

The issue of dividing out wholesale and retail is both important and challenging. While one can exclude certain venues, or exclude certain price ranges, neither is a perfect or even very good divider. We saw this week even $1 million sale can be to another domainer. It is clear that the sales below $25 are virtually completely domainer purchases from expired auctions. If you look by prices, other than that, the mix of retail/wholesale while undoubtedly more wholesale at lower prices does not seem to have a clear demarcation. Keith may well be right that it is 80% wholesale, but I think there is a pretty high percentage wholesale even at sales well past mid $$$. And some of the sales at low to medium $$$ are indeed end user retail ones.

Re your other point:

Here is an illustration of the low usefulness of this: if NameBio decided to report $50 sales, you'd suddenly now come to conclusion that the average price of domain sales in 2018 was under $200.

After your response I did that calculation. If one includes all sales of .com of $50 or more that are in the two NameBio databases the average price is $948 (the median price is $182, that is there are as many sales above that value as between $50 and $182). Adding in sales between $50 and $99 adds 24,600 sales (and $1.8 million) still the vast majority (about 90%) of the under <$100 sales are below $50. There is no doubt we should exclude the under $25 expired auction sales, but after that a dividing line is subjective.

Your point that NameBio data mixes what are revenue with what are costs I totally agree with., and have said the same numerous times such as this 9 months ago in my analysis of the Verisign post:

"...a significant part of the $83.1 million was not to end users, but it was purchases of domains by investors hoping to resell the names. Many drop-catch, GoDaddy or Flippa auction sales reported on NameBio are sales to domainers. So they are counting what is a cost to domainers, and treating it like revenue!"​

Thank you for your comments. I really appreciate the multiple views made in the replies, as adding to our understanding of what any particular data or analysis means, or may not mean. We are fortunate to have the vibrant NP community that makes it so easy to share these different perspectives in an effective way.

Bob

 

[Super-Annuation]

Domain name sales of value $100,000 or more represented only 0.10% of all publicly-recorded .com sales in 2018, but they accounted for almost 30% of the dollar volume of sales for the year. While average sales prices can be useful, they mask the fact that most sales are at much more modest prices. The typical .com sale, at least as publicly recorded, was about $265 in 2018.

I love this Bob! Great comprehensive analysis of Nambio .com data. I have to admit, it has now influenced my attitude towards domain purchasing. I'll be more conservative, and asking myself "would I hold this domain for 10 years." I think it's important to be pragmatic and adopt a systematic approach from now on, given the weak potential for major profit.

To compliment your work, I noticed you could see this kind of pattern in DNPric.es sales lists too! Compared to total .com sales, there isn't an attractive amount of sales that actually pass the desired $10, 000 amount, which is unfortunate.

I've been exposed to the domain industry since say early this year, and didn't understand what I called a Mindless Mass Domain Peddling of $1 - $50 Licenses.

I'm closer to deciding that people purchase mediocre domains, peddle them to make a small profit, ensuring they have made enough fthroughout FY to cover costs of their portfolio (at least break even), playing the numbers game of holding many 'subjective-quality' domains, thus giving them a greater chance of making those occasional large sales, which makes the whole ordeal a somewhat profitable one.

Given the figures and my want for a few of those 10k - 50k domains, I'll have to realise I need look harder, and not buckle at something mediocre, think ahead, and work on marketing, because I'm not hustling low value domains for a hobby. It's the equivalent to panhandling!

Plenty of great analyses there Bob

 

[falez]

appreciate it

 

[Ali Adil]

Thanks Alot...