Here’s some “interesting” data about how many .COM domains are unused

A friend of mine sent me a text with an article from a company called Singapore Data Company that put together an interesting little report about unused .COM domains. It seems like the person who wrote the article has no idea that people are buying and selling domain names and have been for years judging by this sentence in the first paragraph:

It can feel like people are registering every pronounceable combination of letters in every major language, and even the unpronounceable short ones. Is there rampant domain speculation, or do I just think of the same names as everyone else? (Source – Singapore Data Company)

You’ll notice that I put the word “interesting” in quotes above because the writer did a survey of 2,188 domains and used this to generalize stats about all .COM domains. The reality is this sample size is WAY too small. This would be like making a generalization about people living in San Francisco by taking a sample size of 2.

Still, they put together a little colored chart of what they found:


The question is, does this data actually represent how .COM domains are used? Are 4.8% of .COM domains parked? Are 7.1% for sale? Does the author realize that probably most of the parked domains are also for sale. Okay – this is probably completely meaningless…what do you think?

I want to hear from you, comment and let your voice be heard!

{ 2 comments… add one }

  • John McCormac January 31, 2019, 7:10 pm

    The sample size is far too small. However, most PPC parked pages are not for sale. Registrars often park undeveloped domain names on PPC parking using various services from firms like Sedo or Parking Crew. The “no webserver” % above is approximately in the right region. I ran a 1.4 million .COM survey (approximately a 1% survey) a few days ago. Usage in TLDs tends to be a bit more complex than the headings used in the blog post. Approximately 5% should be redirecting to the HTTPS version of the site. Redirects to ccTLD sites should also be present but the sample size might have been too small to properly detect them.

    The gambling one is an indication of Chinese registrations in the sample. Gambling affliate landers are very common in the Chinese market and can often be more common than PPC in some gTLDs. What people who do not do this kind of work professionally fail to understand is that the .COM, like many other large legacy gTLDs, is not a single market but is a set of country level markets with a much smaller global market. If one hasn’t a working knowledge of how usage patterns in various countries affect a sample, then it can produce some confusing results.

  • John February 1, 2019, 2:39 am

    Based on my own anecdotal experience visiting domains, I have a hard to believing the % of parked is not MUCH higher.


Leave a Comment