Let’s be honest, total domains registered is a vanity metric, not a KPI


In the startup world we talk a lot about KPIs or Key Performance Indicators, things that can tell you and other people how well you are doing. At the same time there are “Vanity Metrics” these are metrics that are much easier to make look good, but don’t actually tell anyone how your business is actually doing.

Techcrunch wrote a great article about Vanity Metrics, in it they say:

“Vanity metrics are things like registered users, downloads, and raw pageviews. They are easily manipulated, and do not necessarily correlate to the numbers that really matter: active users, engagement, the cost of getting new customers, and ultimately revenues and profits.” (Source – Techcrunch)

In the domain name world new gTLDs are still new enough that people are trying to determine what are the real KPIs that can tell you how a new gTLD is doing, and what are the vanity metrics that are just big numbers meant to impress the press.

I personally think that “total registrations” is a vanity metric since it is so easily gamed. If one person owns a million .bongo domains and then twenty people each own one. Saying that .bongo is incredibly popular wouldn’t really be true, but the vanity metrics make it looks like it’s on fire.

The real KPI that I think truly should matter to a new gTLD (and their investors) is number of unique registrants. The reason here is that one person who owns 20,000 domains isn’t going to build sites on all those names, put them to use, or do many of the other incredible things that they new gTLDs were created to do. In the end, also the person who owns 20,000 domains is less likely to renew all 20,000 each year vs. someone who owns two or three who will very likely keep renewing them.

The time has come to look past the vanity metrics and focus on the KPIs and for new gTLDs I’m going to come out and say it – total registrations is a vanity metric, total unique registrants is a KPI.

Agree? Disagree? Comment and let your voice be heard!

{ 4 comments… add one }

  • Leonard Britt June 23, 2016, 6:41 pm

    As an investor intEerested in ROI, registrations of a TLD particularly one which plays games to boost registration volume is not a meaningful metric. Registry sales of reserved names have nominal value as well because they likely do not represent the caliber of aftermarket name available to the typical investor. I also believe some reported registry sales are not true arm’s length transactions but promotional efforts to boost registration volume. Even aftermarket sales of a TLD may not be a relevant comparison if those sales are LL or LLL but one’s holdings are keywords.

  • Dot Advice June 23, 2016, 7:15 pm

    @ Morgan , great article . @ Leonard – totally agree . 1/3 of all registrations are not arms length – “sold” to an affiliated company , not a registrant. For investors looking to buy in or indeed looking to sell up and get out – as the marketplace is so far below original budgets, ROI calcs , its better to use financial + Saas metrics based on CAC and Churn – registrations volumes are totally distorting the marketplace and I agree pure vanity metrics.

  • Andrea Paladini June 23, 2016, 7:41 pm

    Totally agree with Techcrunch on those “vanity metrics”, it’s since the late 90’s that I’m trying to make investors understand that those data are not meaningful, besides being easily manipulable and uncorrelated to a company underlying profitability drivers and competitive advantages (if the company has any).
    And those mentioned are not the only ones … more junk metrics and stats around 🙂 …

  • Joseph Peterson June 24, 2016, 7:48 pm

    Distinct registrants is a better indicator of success than total registrations. But this “KPI” can be (and has been) gamed as well.

    A few weeks ago, when .XYZ dumped a limitless number of domains on the market for a penny each, .XYZ naturally scooped up a bunch of registrants who had a few pennies in their pocket.

    Those customers were willing to pay up with their spare pennies and play the scratch ticket for a minute or two. But should they count equally alongside customers who deliberately go after domains they want, paying $10 or $10,000 for a single domain, perhaps even developing it?

    Is a 12-day registrant equivalent to a 12-year registrant who has been through multiple renewal cycles? For instance, .XYZ might lose 1000 registrants who drop domains they bought last year at $1 while simultaneously adding 1000 registrants who are speculating on new .XYZ domains for $0.01. Rationally, that scenario ought to be a step down; but using the metric of distinct registrants, it would show up as a flat line.

    The number of distinct registrants is problematic. Many other metrics would be better. For example, we can look at registrants’ cumulative spending in a TLD – Q1, median, Q3. We can also look at the number of distinct registrants who have registered full-priced domains in a TLD. Or at renewal rates. Number of developed websites. And so forth. Many metrics to choose from.

    While the number of distinct registrants is quite informative, it isn’t reliable enough for apples-to-apples comparisons.


Leave a Comment