A one-afternoon protocol for testing a name before you commit

Naming processes fail in two ways: they produce a shortlist nobody tests, or they produce a testing regime so elaborate that it is skipped. What follows is short enough to actually happen and ordered so that the cheapest tests eliminate candidates first.

Run it on the exact string that will appear in the address bar: lower case, no marks, no spaces.

Test 1: the cold reading (10 minutes)

Show the bare string to three native speakers separately. Ask: what does this say? Write the first answer verbatim before responding.

Pass if all three match your intended reading. Fail if any produces something unfortunate. Partial if they split — which is real information, not a tie.

Test 2: is it a thing people say? (5 minutes)

Same people, second question. Is this an expression in actual use, or two words next to each other?

This catches the compound that looks plausible to a foreign ear and reads as a typing error to a native one. It is the fastest test in the set and eliminates more candidates than any other.

Test 3: the telephone test (10 minutes)

Say the name aloud to someone who has not seen it and ask them to write it down. Then ask them to read it back.

Every failure here becomes a permanent tax: support calls, lost traffic, spelled-out addresses. Hyphens, doubled letters and near-homophones fail this test and are why they are worth avoiding.

Test 4: what is already there (15 minutes)

Search the term. Are the results about your intended meaning, or something else entirely? Is there an established business using it? Is the search volume plausible for a category term?

A term whose results are dominated by an unrelated subject is a term you will fight for years.

Test 5: the trademark register (20 minutes)

Search the Vietnamese register and the registers of any market you intend to enter. A generic category word is usually clear; anything that reads as a brand needs checking.

This is the one test where a professional search is worth paying for if the name matters. A clear-looking register does not mean no rights exist.

Test 6: the extension sweep (10 minutes)

Check what is available and what is held: the .com, the local extensions, the obvious near-variants. Who holds what you cannot get, and are they a business or an investor?

This tells you both the cost of the full set and whether somebody is already trading under a confusable address — a conversation better had before launch than after.

Reading the results

A candidate that passes all six is rare and should be secured immediately. A candidate that fails test one or two is finished, regardless of how much the team likes it. A candidate that fails three, four or six is a judgement call about how much friction the business can carry.

Failing test five is a stop, not a discount. Trademark problems do not improve with time or traffic.

The discipline that makes it work

Run the tests before anyone falls in love with a candidate, and write the results down where the team can see them. A name defended on the basis that “we already made the logo” is how a business ends up spelling its address out loud for the next decade.