A short name can look settled long before anyone has tried to use it. It fits neatly in a logo, and the team gets used to saying it during meetings. Then a prospective customer hears it on a call, types something unexpected and lands somewhere else. A small naming test is useful because it gives those ordinary moments a place in the decision before the name becomes expensive to change.
The aim is to learn how the name behaves in context. You are not asking a handful of people to predict the future of a business. You are looking for recurring confusion, workable pronunciations, unwanted associations and a name people can recover after their attention has moved elsewhere. Treat the following approach as a practical exercise to adapt to your audience, not a validated scoring model or a substitute for professional clearance.
Write down the decision before choosing the questions
Start with the business use. A spoken referral to a local service creates different demands from an app discovered through a visual store listing. List the places where the name must work during the first year: a phone introduction, an email address, a search result, a small product label or a presentation to a purchasing team. Choose two or three situations that would be costly to get wrong.
Then write a question for each situation. Can a likely customer spell the name after hearing it once? Can someone recognize the written name later? Does the name suggest a category so strongly that it interferes with the actual offer? These questions are more useful than asking whether the name feels premium. A vague preference gives the team room to reinterpret the answer after the fact.
Keep a separate list for legal clearance, domain ownership, social handles and commercial terms. Those are different investigations. A name that people remember can still be unsuitable for a business, and a positive research session does not establish trademark rights. Ask an appropriately qualified adviser to assess that area for the intended markets and services.
Recruit people who resemble the first customers
Choose participants who can understand the proposed offer because it relates to a job they already do or a purchase they might reasonably consider. Colleagues and friends may be useful for checking whether the instructions make sense, but their familiarity with the team can influence the conversation. If the audience includes people using different languages, devices or assistive technologies, include those needs in the research plan.
There is no universal participant count that makes a name safe. A small round can reveal problems worth investigating, while a broader study may be appropriate for a consequential launch. Record who participated and what the group does not represent. Avoid turning a few favorable responses into a percentage that appears to describe the whole market.
Test spoken spelling before showing the logo
Read the same short sentence to each participant at a natural pace. For example: “The service is called [candidate]. Could you write the website address you would try?” Do not spell the name first or give a hint about its length. Record the first response, any requested repetition and the participant's own explanation afterward. A corrected answer should not erase the initial difficulty.
Test the word and the complete domain as separate moments if both matter. Someone may remember the name but assume a different extension. Another person may hear letters as an acronym when the team expects a spoken word. Neither observation automatically disqualifies a candidate; each tells you what the introduction, advertising or customer support would need to address.
Use consistent recordings if several people are conducting sessions. Otherwise the most enthusiastic interviewer may accidentally give one name clearer pronunciation or more context. Rotate candidate order between sessions so the first or last position does not belong to the same name every time. Keep the exercise short enough that participants are not simply practicing a list.
Give recall a little distance
Show the name in a plain, comparable treatment alongside a brief description of the offer. Then move to another task for a few minutes. Ask what the service was called without showing the candidate list again. Note exact recall, approximate recall and no recall separately. Recognition from a list can be a later question, but it is a different task from producing the name unaided.
A useful distraction might be asking participants how they currently solve the underlying problem. That conversation can also expose vocabulary the team has overlooked. Keep notes on the naming result separate from comments about the offer. A participant may remember a weak product description and forget a perfectly usable name, or be excited about the product while misremembering its identity.
Put the name inside a real task
Build a simple page or message that resembles the intended context. Ask the participant to find the business in a small group of fictional search results, identify which message belongs to a service they just saw or read an address aloud. Use invented surrounding examples and avoid collecting personal account details for a naming exercise.
The GOV.UK guidance on moderated usability testing recommends tasks with clear goals and neutral instructions that do not reveal the answer. Applied here, that means asking someone to find the service rather than asking them to click the unusually short name. Observe first, then ask what led to the choice. The explanation often reveals whether the participant used the name, the description or an unrelated visual cue.
Keep presentation comparable across candidates. If one option appears in a polished logo and another in a rough text box, you are testing a mixture of name and design. Later, a separate round can test the preferred name with its actual visual identity. That sequence lets you see what the design adds or obscures.
Ask about associations without planting them
After the practical tasks, ask what the name brings to mind and what sort of organization might use it. Let the person answer before naming your intended category. Follow up on specific words: “What made you think of travel?” produces more information than “Does this sound trustworthy?” Write down the participant's language rather than translating every response into the team's preferred brand vocabulary.
Be careful with a familiar short word. It may carry an existing meaning, be read as initials or remind someone of another organization. Investigate repeated associations in the markets that matter. One surprising response can prompt a follow-up question; it does not need to become the entire strategy. Likewise, silence about a problematic association is not evidence that no such association exists.
Compare observations, then make the decision
For each candidate, assemble a short record: spoken spelling errors, unaided recall, recognition, task completion and the associations that came up. Add context about the participant and the order in which candidates appeared. Resist compressing everything into a single score unless the team has agreed what the score means. A serious ambiguity in an email address may matter more than several mild statements of preference.
An illustrative result might show that Candidate A is recalled often but regularly spelled in two ways, while Candidate B is less exciting to participants but written correctly after a spoken introduction. The next decision depends on the distribution plan. A referral-led business may investigate the spelling issue more deeply. A visually discovered product may first test the written presentation at the size customers will encounter.
End with a documented choice and its conditions. State what the round answered, what remains uncertain and which checks must happen before launch. Keep a few examples of actual participant behavior so future discussions do not collapse into “people liked it.” Choose the name whose strengths fit the audience, offer and situations you tested, with a clear plan for any remaining concerns.