Hierarchy for Domain Names — A Fallback
Joseph Peterson takes a look at fallbacks if the domain name you want is taken.
In a recent article on the expired domain market, I was struck by NameJet’s top sale for the week: Insurance1.com, which had been a developed website since 1996. What’s striking is that this domain showed its age immediately so that verifying the “vintage” was a mere formality. Many of you will share my gut reaction. We’ve been browsing the web such a long time, following all, that we recognize a dated naming convention as belonging to a certain era.
Many years ago, adding the numeral “1” following some juicy keyword was a recurrent workaround. If someone judged Insurance.com or Office.com or Trailer.com to be out of reach, then they went following Insurance1.com (1996) or Office1.com (1997) or Trailer1.com (1996). Those domains remain registered today, partly since of their very age and the back links they accrued. But – and this is crucial – contemporary domain buyers are not inclined to add “1” as a default short cut. Not to the same degree or at the same frequency as mid 1990s registrants.
Back then, cutting this little corner allowed us to obtain a faux single-word .COM simply by adding 1 character – quite a memorable character as well, this “1”, with its connotations of being first, best, primary. That branding logic would still apply today but for 2 reasons:
(A) We’ve had more TLD options at our fingertips ever since .INFO and .BIZ were introduced in 2001, continuing through periods of increased .ME and .CO and .IO adoption, all the way to our present surfeit of hundreds of nTLDs. So a “1.com” suffix faces more competition.
(B) The Keyword+1 convention once so apt now strikes us as belonging to the past, and such datedness will not scrub off. Instead fresh fads and conventions are constantly arising. Rather than choose Trailer1.com or Office1.com, by the fresh millennium we were hankering following Trailerz.com (2002) and Officez.com (2003). Insurancez.com actually had a head start in 1999!
Again I hardly need to check the vintage. In my previous article, a number of “Z” misspellings were noted sold: Themez.com (2000), Smokerz.com (2003), and ActiveGamez.com (2007). (Naturally it took longer for naming pressure to build up until the fashion for “Z” spilled over into 2-word domains.)
These trends intrigue me. In part, that’s since I study the domain market as an analyst. But I also work regularly with clients on naming and domain selection; so I’m constantly confronted by the myriad ways in which their domain priorities have changed … or haven’t.
For example, I know a serial tech entrepreneur whose advice to others in 2015 remains a firm policy: “If .COM is taken, then use .CO until venture capital arrives to rescue the .COM.” I won’t say that he’s wrong, but his subjective range of options is obviously limited. Actually, it’s easy to date his own startups to the period of greatest .CO promotion. Had his projects launched a few years later, then his rule might be identical except that .CO might have given way to a knee-jerk .IO.
And my point is?
One of the most interesting aspects of naming domains is what I’d call the “fallback hierarchy”. Think of it this way: When your ideal domain name seems out of reach, what is your first recourse? What’s the first alternative adaptation that springs to mind? And if that’s unavailable, then what? And so forth in order of priority, moving through conditioned habits and instinctive good taste, from reflex to gut reaction, until we run out of ideas completely. At that point, we’ve passed a reluctant buyer’s domain of last resort and are plunged into the wasteland of the unthinkably bad or utterly unknown.
You might imagine this fallback hierarchy as a loosely ordered sequence:
1. Cars.com
2. Cars.net / Cars.org / Automobiles.com
3. Carz.com / TopCars.com / Cars.info
4. CarsSite.com / Cars.biz / Automobilez.com
5. Cars.website / Carz.net
6. TopCars.net /Automobilez.net / Top-Cars.com
etc.
My choices above are fairly arbitrary. Rearrange them if you like. What I’m getting at is, of course, very familiar. Registrants will add keywords or extra characters; we’ll pick a different extension; we’ll introduce misspellings or hyphens; we’ll look at longer words or less familiar words; or else we’ll shift gears and pursue a different naming style entirely. Some alterations are preferred, while others are undertaken more reluctantly. And the fewer transformations the better.
This fallback hierarchy varies based on demographic – especially across ccTLD and language boundaries. German buyers will favor .DE, of course; but they’re also much more receptive to hyphens and value .NET more than some American domainers expect. Tech entrepreneurs who don’t get out much will overestimate the familiarity of .IO compared to mainstream registrants and rank .IO higher within their fallback hierarchy. And, as we’ve discussed, these go-to naming habits vary over time as well. Adding “1” would have ranked high up in people’s fallback hierarchies during 1998. The same option exists nowadays, but a “Z” misspelling will likely have supplanted it. And for many buyers some fresh TLD (.CLUB perhaps) has superseded the “Z” misspelling.
Priorities change. In fact, priorities change most below the surface. The uppermost layer of people’s fallback hierarchy, a person’s first-choice domain name, has scarcely changed at all from the beginning of the internet until now. If you ask most people, their ideal domain would occur in whichever TLD dominates their audience’s attention – .COM globally, ccTLDs locally, perhaps .ORG for nonprofit causes. Usually the buyer wants the domain to be as simple, short, and familiar as possible and with as few artificial concessions (e.g. hyphens, misspellings, extraneous characters) as may be.
This top slot in the fallback hierarchy is almost constant over time. Yet underneath it, the workaround options are constantly shuffling positions. We walk on a firm crust of ideal names beneath which the fallback options are a molten swirl. Old fads are submerged by fresh trends. And, as fresh extensions are discovered more and more, they float nearer the top. Indeed, the whole nTLD initiative can be regarded as a fierce competition by registries to reorder the status quo of consumers’ learned fallback hierarchies.
Don’t misread this as my saying “.COM is King”. For chance historical reasons, .COM is quite dominant. But I’m referring largely to intrinsic naming values. Whatever your first choice is, it will feel firm, whereas runner-up ideas are far more volatile. Less desirable options are more numerous; and, as we’re less attached to them, they dislodge one another more readily with the changing fashions.
Maybe this all seems fairly obvious. And in a vague way, it is. But fallback hierarchies could be made explicit and rigorously studied. This is a field ripe for consumer research. ICANN has conducted opinion polls regarding the nTLD program, intending to gauge global awareness and receptivity. That’s all well and good. Yet I’d say what we really need are studies of registration patterns – real and simulated (so as to control for pricing), both accomplished and attempted.
When I say attempted registrations, I’m addressing registrars specifically. For it’s at the registrar that buyers actually pursue their fallback hierarchy as a definite ordered sequence of actions that can be traced. When they hit an obstacle, we can learn a great deal from the angle at which they rebound – toward keyword + 1? toward creative misspellings? toward added words? toward some other extension? If so, then toward which TLDs in which order? Who and in what context? How big a price discrepancy causes how much of a deflection in fallback preference? Registrars hold this data, but I doubt they analyze it adequately.
