How to Choose Between Domain Name Candidates (What Data Settles, and What It Can't)

Three names on a whiteboard, zero ways to rank them

Every naming process ends the same way: a shortlist of two to five candidates that all survived the brainstorm, and a room full of opinions about which one is “better.” Most advice at this point is branding vibes — say it out loud, imagine the logo, ask your friends. This guide is about the other input, the one that has actual numbers attached: what the domain aftermarket says each candidate is worth as an asset. That input can’t make the decision for you — we’ll be precise about why — but it’s measurable, it’s free to check, and ignoring it is how founders discover too late that one shortlist name was a four-figure asset and another was renewal-fee filler.

What the sales data can actually settle

Reported aftermarket sales — the NameBio public database is the standard source — support three real distinctions between candidate names:

• Pattern class. A single dictionary word, a two-word compound, a brandable coinage, and an acronym trade in measurably different price bands. “cedar.com” and “getcedar.com” are not the same kind of asset, and reported sales say so with sample sizes attached.
• Extension. The same name is worth different money on different TLDs — measured medians, not marketing. Our startup TLD guide covers this half of the decision in depth.
• The range itself. Each candidate gets an appraised wholesale range anchored to reported sales in its class — the same itemized record our homepage appraisal produces, one per candidate.

Our name comparison tool runs exactly this: paste the shortlist, get each candidate’s range, class, and confidence grade side by side. Thirty seconds, nothing uploaded, and the vibes conversation now has a measured baseline under it.

Expect a tie — and distrust any tool that never shows one

Here is the part most comparison tools won’t tell you: run a realistic shortlist through the numbers and the usual result is a tie. An appraisal is a range, not a point — the honest summary of scattered comparable sales — and when two candidates’ ranges overlap, any confident ranking between them is precision the data doesn’t contain. That’s why our comparison names a leader only when one candidate’s low end clears the other’s high end — when even the most pessimistic reading of one beats the most optimistic reading of the other. Anything closer gets an explicit OVERLAP verdict.

A comparison tool that always produces a winner is manufacturing certainty, not measuring it. Single-score comparers rank overlapping ranges because a tie doesn’t sell a report — a tie is frequently the truthful answer.

A tie is not a useless result. It tells you something specific and freeing: the resale data has no objection to either name, so the decision belongs entirely to the judgment criteria in the next section — and nobody in the room gets to win the argument by citing an appraisal number that’s inside the other name’s range.

When the numbers do separate, ask the second question

Sometimes the comparison is decisive — a one-word .com against a hyphenated long-tail name will separate cleanly, because they’re different asset classes, not different flavors of the same one. When that happens, ask the question the numbers can’t: does resale value even matter for this purchase? If you’re building a company on the name and never intend to sell it, the higher-appraised candidate is better collateral and a better hedge against a future rebrand — real advantages, but indirect ones. The appraisal gap is a fact; how much weight it carries against brand fit is a choice, and pretending the fact makes the choice is exactly the black-box move this site exists to avoid.

Also check the confidence grades printed on each row before trusting a separation. A grade-A range comes from a directly measured sales cell; a grade-C range rests on a labeled judgment band and could sit almost anywhere inside it. A “separation” between an A-grade and a C-grade range is weaker evidence than the same gap between two A-grades — the grades exist so you can see when that’s happening.

The judgment half — labeled as judgment

What follows is naming craft, not measured data. None of it appears in sales records, which is precisely why no appraisal engine — ours included — can score it:

• The radio test. Say the name aloud to someone once and ask them to type it. Homophones, creative spellings, and hyphens all fail here, and the failure compounds on every podcast ad and word-of-mouth referral forever.
• Audience trust. Extensions and coinages carry different signals in different markets — developer tools live comfortably on .io; a law firm doesn’t. Your customers’ expectations outrank the aftermarket’s.
• Collision risk. Search each candidate before you commit. A similarly named competitor — or worse, one with a registered trademark — turns the “cheaper” candidate into the most expensive mistake on the list. A name that infringes a live mark can be worth less than zero once a dispute starts; WIPO’s UDRP guide explains how those disputes work. No appraisal models this; check it yourself.
• Room to grow. A name welded to one product (“cedarinvoices”) costs a rebrand when the second product ships.

The number the comparison can’t see: what each name costs to get

An appraised range describes the reported resale market. It says nothing about what each candidate’s current owner will actually charge you — and on a real shortlist, one name is hand-registerable for $12, one is listed at four figures, and one isn’t for sale at all. A leader you can’t buy loses to a runner-up you can. Get asking prices for the candidates that are taken, then read them against the appraised ranges: buying as an end user, expect to land around the widely published ×2–3 retail multiple over wholesale (DNAcademy, DomainDetails). Our asking-price guide walks that negotiation in full, including the marketplace tactics worth discounting.

The whole procedure, in order

1. Run the shortlist through the comparison — ranges, classes, and grades for every candidate at once.
2. Read the verdict. A separation is a real asset-quality gap (weigh it, noting the grades); an overlap hands the decision to judgment with the data’s blessing.
3. Apply the judgment tests — radio test, audience trust, collision search, room to grow — to the candidates still standing.
4. Price the finalists: registration or asking price for each, read against its appraised range. For a single finalist’s full itemized record, run it through the appraisal tool; to weigh one name across extensions, use the TLD comparison.
5. Trademark-check the winner before money moves — last, always.

What no comparison can tell you

Which name is right for your company. The numbers rank candidates as resale assets — honestly, with their uncertainty shown — and most of the time they’ll tell you the shortlist is a tie, because it usually is. What they buy you isn’t the decision; it’s a floor under it: no candidate quietly worthless, no four-figure asset dismissed as equivalent to a $12 hand-registration, and no appraisal number wielded in the naming argument with more confidence than the data behind it deserves. The choice, as with every tool on this site, stays yours.

This guide is for informational purposes only. It is not financial, legal, or investment advice, and it is not a certified appraisal. A domain’s real price is set by what a specific buyer actually pays — no article or model can know that in advance, and we say so instead of pretending otherwise.

Last reviewed: July 2026 · Against the NameBio sales anchors behind our appraisal engine and the industry sources linked in the body.