Defaults PersistNarrow moat

The most valuable thing VeriSign owns is a habit.

When somebody who has never registered a domain thinks about getting one, they think about a .com. It is the suffix in the example, the one assumed when none is given, the one a customer tries first. Nobody decided this and nobody can revoke it. It is the accumulated residue of thirty years in which .com was, for most purposes, the only sensible answer.

ICANN has run a genuine experiment on whether that persists. The 2012 round produced more than a thousand new top-level domains, many of them shorter, cheaper, more descriptive and more available than anything left in .com. Fourteen years later there are 401.6 million domain registrations worldwide and 179.1 million of them — roughly 45% — are .com and .net, while the entire new-gTLD field accounts for 52.9 million.¹ The alternatives did not fail; they took a third as many names as the incumbent in fourteen years of trying.

The honest counterweight is that defaults are generational rather than permanent. The new gTLD field grew 34.0% year over year against 5.1% for .com and .net. Growth rates that different, sustained long enough, change what the next generation assumes.

Two things would signal the default breaking, and neither has: a large, recognisable company choosing a non-.com as its primary address, and the .com renewal rate falling while the total market grows.