.com versus the new TLDs, honestly compared

.com is still the default, and defaults are expensive. A decent one-or-two-word .com trades in the mid four figures to low six figures. The equivalent .io, .ai, .co or .xyz is often a tenth of that, sometimes a hundredth. That gap is real money for an early company, and the honest question is not "which is better" but "what am I actually buying with the premium".

What the .com premium actually buys you

Three things, roughly in order of how much they matter.

• Type-in and recall insurance. When someone hears your name at a conference and types it later, they will append .com unless they have a reason not to. If a competitor or a parked page sits on the .com, you leak a slice of traffic you will never measure. The leak is small for a company with a paid acquisition funnel and large for one that relies on word of mouth.
• Email deliverability and legitimacy signals. Some spam filters and procurement teams treat unfamiliar extensions with more suspicion. This is uneven and getting better, but it is not zero. If you sell to enterprise, ask your first ten prospects what they think before you commit.
• Resale liquidity. A good .com has a deep buyer pool. Most alternative extensions have a shallow one, dominated by people in the same niche. That does not make them worthless, but it does mean you should price them as an expense, not an asset.

What the premium does not buy you: search ranking. Google has been explicit for years that generic top-level domains carry no inherent ranking advantage. Any correlation you see between .com and rankings is a proxy for age, links and budget.

Where new TLDs genuinely win

There are situations where the alternative is not a compromise, it is the better name.

• You get the exact word. A one-word .co or .io that names your category beats a three-word .com that describes nothing. Clarity in the string usually beats familiarity in the suffix. If ledger.com is a nine-figure fantasy and ledger.xyz is affordable, the second is a stronger brand than getledgerapp.com.
• Your audience already lives there. Developer tooling and .io, AI products and .ai, design studios and .studio. Within these audiences the extension reads as a signal rather than a compromise. Outside them, it reads as "could not afford the .com".
• You are not chasing type-in traffic. If your growth is app store, marketplace, sales-led or community-led, the recall gap costs you less. Nobody types the URL of a company they found through a Slack link.
• You are pre-product-market fit. Spending £40,000 on a domain before you know what you are selling is a bad trade. A cheap extension now and a .com later is a legitimate sequence, and plenty of companies have run it.

Where new TLDs quietly cost you

The costs show up later, which is why founders underrate them.

The .com owner becomes your permanent landlord. Once you have traction, the price of the matching .com goes up, because the seller can see your funding round. Buying it at seed is cheaper than buying it at Series B. Some companies never buy it and are fine. Some end up paying several multiples of what an early purchase would have cost, plus the cost of a rebrand consideration they never wanted.

Confusion in spoken contexts. "Dot io" needs spelling out on a phone call. "Dot ai" gets heard as "dot A I" or "dot eye". Say your full domain out loud fifty times before you commit; you will do that in real life within a month.

Registry risk on the long tail. Country-code extensions are governed by national policy, and policy changes. Some ccTLDs have residency requirements, some have had pricing or eligibility shifts. Read the registry terms for anything that is not a generic extension, and check the renewal price for year two and beyond — some new TLDs renew far above their first-year promotional cost.

The trademark problem does not go away. A distinctive name on any extension is defensible. A generic name on an obscure extension is not. Your extension choice does nothing for your legal position; the strength of the name does.

A decision framework you can actually apply

Answer these four in order.

1. Is the .com owned by an active business in your space?

If yes, stop considering that name entirely, regardless of extension. You will fight for search results and email forever. Pick a different word.

2. What is the .com actually priced at?

Founders assume six figures and never ask. Plenty of good .coms sit in the low four figures. Check before you rule it out — see what is genuinely available with a keyword domain availability check rather than guessing from a hand-wave.

3. Can you spread the cost?

If the .com is right but the lump sum is not, leasing or instalments change the maths. A monthly payment structure turns a capital decision into an operating one, which is often the difference between getting the exact name and settling.

4. Would a different .com beat a compromised alternative?

This is the question most founders skip. Instead of yourword.io, consider an invented or compound .com that is available outright. The Name Studio generates brandable .com options and verifies each against the live registry, so everything it shows can actually be registered. A clean coined .com often outperforms a borrowed word on a borrowed extension.

What to do if you buy the alternative

Assume you go with .ai or .io. Protect yourself:

• Find out the .com price now, in writing, before you launch. Even if you do not buy, knowing the number lets you budget for it.
• Register defensive misspellings sparingly. Two or three, not twenty. Redirect them.
• Use the extension consistently in all marketing. Never write your name without it. The habit trains recall.
• Set a trigger for revisiting. "When we hit £1m ARR or raise a Series A, we buy the .com." Write it down so it survives the founders forgetting.

The short version

If your name is generic and descriptive, the extension matters more, because the string does nothing to distinguish you — this is where a true category-killer .com earns its price. If your name is distinctive and coined, the extension matters less, and a good alternative TLD is a rational trade.

The worst outcome is neither of these: a mediocre three-word .com bought purely because it ends in .com. Familiar extension, forgettable name, and you paid for the wrong half.