Name Testing: How to Test a Product or Brand Name
Name testing checks whether people understand, remember and can say a product or brand name. Learn what to measure, seven questions to copy and an example.
Name testing checks what people think a product or brand name means, whether they can say and remember it and how well it fits the offer. Test the information people will actually see or hear. Use this guide’s seven questions to compare names and supporting copy for your launch.
Example: comparing names for a meal-planning service
This hypothetical naming exercise compares “Weekplate,” “Dinnerpath” and “Pantryloop” for a meal-planning service. The first encounter matters: will people see the name alone or with a category line and explanation?
Weekplate makes one buyer expect ready-made meals until the description explains planning and a grocery list. Can the launch reliably show that explanation? Keep the first meaning apart from meaning after explanation. Also complete language, availability and professional clearance checks in the intended markets.
A name that works with copy may fail in a spoken recommendation. Start with how people will discover the service, then test the information available there. Read meaning beside recall. Remembering the wrong service is still a problem.
Weekplate before and after the description
Weekplate
“Sounds like ready-made meals.”
A service that prepares a weekly dinner plan and a consolidated shopping list. You buy the groceries separately.
Save two stages of the cards. First show a name-only card for Q1–Q2, with identical type, size and layout and no description or tagline. Then reveal the same description from Q3 onward: “A service that prepares a weekly dinner plan and a consolidated shopping list. You buy the groceries separately.” Keep the description layout identical too. Do not give one name a persuasive tagline or more attractive artwork.
If people will first hear the name in a recommendation, include audio or an interviewer who says it consistently. A text-only test cannot tell you whether people understand the spoken name.
Seven questions that keep the first meaning visible
Randomly assign people to see one candidate first. Collect first impressions before showing the full list.
- What kind of product or service would you expect from this name? Open response before the description.
- What comes to mind when you see the name? Open response; ask for both favorable and unfavorable associations.
- After showing the description: how well does the name fit this service? Very poorly / somewhat poorly / neither / somewhat well / very well / cannot judge.
- How easy or difficult would you find it to say the name aloud? Very difficult / somewhat difficult / neither / somewhat easy / very easy.
- Does the name remind you of another product or organization? If so, which? Open response.
- After a defined intervening task, ask participants to recall the name without showing it again.
- In a later balanced comparison, ask which candidate they would investigate further, including “none” and “no preference,” then ask why.
After Q5, hide the name and description and show a two-minute unrelated task with no names or meal-planning cues. Ask Q6 without letting people reopen earlier cards. Save the stage, exposure ID and exact delay, and set that rule before fieldwork. This short recall task does not forecast long-term brand memory. Repeated exposure is not an unprompted first impression.
Meaning, fit and recall can disagree
Measure | Report and interpretation |
|---|---|
Category understanding | Report: Coded expectations and base Interpretation: Whether the name points toward the intended offer |
Associations | Report: Themes, examples and contradictory cases Interpretation: Possible benefits, confusion or unwanted meanings |
Fit | Report: Full response distribution Interpretation: Stated match with the supplied description |
Pronunciation | Report: Difficulty reports and any observed attempts Interpretation: Potential spoken-use barrier within the tested language |
Recall | Report: Exact/accepted matches divided by eligible exposed participants Interpretation: Recall under the specified study exposure |
Preference | Report: Choice shares with “none” retained Interpretation: Comparative stated choice under this presentation |
Agree how to code answers before ranking names. A spelling variant may count as recall if it clearly identifies the name; document that rule. Do not change it to rescue a favorite.
Keep the two stages in the record: “Weekplate sounds like ready-made meals” before explanation, and “I understand that I buy the groceries” afterward. The first shows a risk when people see only the name. The second shows why the name with its description may still be worth comparing. Keep both so a good fit score cannot hide the wrong-category interpretation.
How the description changes the meaning
Use the patterns below to decide what to change. You might need a clearer description, a more distinctive name or a check of spoken understanding. Those call for different follow-up work.
Pattern to look for | Decision and next comparison |
|---|---|
Memorable but wrong category | Decision implication: Initial encounters may create a false expectation Next comparison: Compare a clarifying description or another name under the actual first encounter |
Correct category but low recall | Decision implication: The name communicates but may be difficult to retrieve Next comparison: Inspect distinctiveness and spoken/text exposure separately |
Positive fit after explanation only | Decision implication: The description is doing important work Next comparison: Test the realistic encounter with the information actually available |
Association with another brand | Decision implication: Confusion or familiarity may affect preference Next comparison: Investigate the association and obtain separate professional clearance |
Keep the stage, candidate and exposure IDs. Name-only associations, understanding after a description and delayed recall measure different things. Report them side by side. Do not average them into one score where the weights decide the winner.
Decide what counts as a recalled name
For the two-minute delay, accept a spelling variant only when your rule clearly identifies the candidate. Record exact matches, accepted variants, wrong names, uncertain answers and missing answers separately. Seeing the card again before answering is not unaided recall.
Test the first encounter people will actually have
The two stages above show what a name means alone and after explanation. To compare full first encounters, randomly assign people who have not seen the name-only cards to one of the three names with matched copy. Show the shared category line from the start. Ask what people expect before asking about fit. For spoken referrals, test hearing, pronunciation and category meaning without hints. Save the content and stage with each answer.
Use “Weekly meal plans and one grocery list” with Weekplate, Dinnerpath and Pantryloop, with the same full service description for each name. Ask what buyers receive and what they must do themselves before asking about fit. If Weekplate still implies prepared meals, revise the name or category line. Among the versions people understand correctly, compare associations and recall using the same cards and delay.
Feature labels need action and permission checks
A feature label has a different job from the brand name. Delayed recall is not the main result here. Users need to predict what the control does and whether someone else can see or change their information.
Extend the example to the shopping-list screen. A household partner should be able to view the list while its owner keeps editing control. The proposed function opens a confirmation dialog, then creates a read-only link that anyone with the link can open. It does not send email, invite an editor or order groceries. Test those actions and permissions before choosing the label.
Compare “Share list,” “Invite to shop” and “Create view-only link” in the same interface. Keep the nearby “Edit meals,” “Print list” and “Order groceries” controls visible, with the list's current saved state. A label alone on a survey card can't reveal whether someone mistakes it for the neighboring order action. Give similar groups of first-time users one label each before showing alternatives.
Ask, before allowing a click: “What do you expect this control to do?” “Who would be able to open the list?” and “What would they be able to change?” Capture the answer without supplying the intended meaning. Then let the person complete the task in the prototype. Save which control they use, whether they check the confirmation and what they think has happened afterward. A correct explanation after the dialog does not replace what they predicted from the label alone.
Candidate label | Meaning and permission evidence to retain |
|---|---|
Share list | Does the person expect a link, an email or a public post? Do they assume viewers can edit? |
Invite to shop | Does “invite” imply a named person or editing rights? Does “shop” imply an order rather than a saved list? |
Create view-only link | Does the person understand link creation, read-only access and the remaining question of who can open it? |
The final confirmation can say, “Anyone with this link can view your grocery list. They can't edit it. Creating a link doesn't order groceries,” with “Create link” and “Cancel” as the actions. Test that full treatment too. A clearer label may fix one misunderstanding while leaving people unsure who can open the link. Explain that permission where the user needs to decide.
Report one row per label, interface state and audience. Show how many people saw it. Code predicted actions as intended, email, editing invitation, order or other. Report predicted viewer access and editing rights, uncertain or missing answers and observed completion when people tried the working prototype. Keep the answers behind misunderstandings. Do not combine these into a “name appeal” score.
Use the blank export to keep those results checkable. Give each answer one main action code and keep mixed or unclear explanations in the source text. Use each question’s own total for action, viewer-access and editing-right counts. Count uncertainty as an answer and report missing answers separately. Task-completion results include only people given the working prototype task.
candidate_label,interface_state,audience,exposed_n,action_answer_base,intended_action_n,email_n,editing_invitation_n,order_n,other_action_n,uncertain_action_n,action_missing_n,viewer_access_answer_base,anyone_with_link_n,invited_people_only_n,other_access_n,uncertain_access_n,viewer_access_missing_n,editing_rights_answer_base,view_only_n,can_edit_n,uncertain_rights_n,editing_rights_missing_n,prototype_task_base,correct_task_completed_n,source_answer_ids
W3C's guidance on headings and labels calls for labels that describe their topic or purpose. That is a useful rule, but preferring a label does not show that it explains this action. Check the control's visible wording and accessible name in the working interface, along with the behavior it triggers.
Choose wording that helps users predict and finish the right task, including the permissions. If someone understands the function but rejects anyone-with-the-link access, the problem is product policy. Restricting sharing may help more than renaming the same permission.
Meaning can disqualify a memorable name
Agree which misunderstanding would rule out a name before reading scores. For spoken referrals, repeated wrong-category interpretations may disqualify a name even when recall is strong. For a name with copy, test whether the available explanation makes the service clear. Do not let an average trade away that meaning for a higher preference rating.
State where each shortlisted name can work and what still needs checking. Investigate confusing associations and obtain separate clearance for similarity to another organization. Explain those consequences; a decimal ranking can hide which launch risk its weights accept.
Choose a name and explanation the launch can deliver
If the launch will include descriptions, compare the three complete versions first. If it depends on referrals, favor a name that communicates planning without later explanation when the evidence supports it. The Weekplate answer identifies one weakness. It does not give a measured winner among all three names.
Deliver the name with its category line, exact first encounter, supporting findings and open checks. The creative team can build that naming system. The launch team can judge whether people will reliably get the information it needs to work.
Test the intended languages and markets with people who have the right expertise. A model may suggest associations to check but cannot certify that no offensive or confusing meaning exists. Get professional legal clearance and availability checks separately.
Sapien can compare simulated interpretations and objections across a defined audience and recommend a name or copy direction within the agreed study. Bring the candidate cards to Sapien to agree the audience, first encounter and measures your decision needs, including whether the proposed recall design can be supported. For the wider product promise, see our message testing guide and our positioning research guide.
Related guides: concept testing; packaging design testing; claims testing; product launch research.
Questions about name testing
How many names should you test?
There is no fixed number for every study. Start with feasible candidates and plan comparable first-exposure answers for each. The example compares three names. If people later compare several, save the presentation order. Report that comparison apart from each name’s first meaning and delayed recall.
Copy these questions
These questions come from the example above. Keep the complete offer, recruitment criteria, answer options and rules for what people see when adapting them.
- What kind of product or service would you expect from this name? Open response before the description.
- What comes to mind when you see the name? Open response; ask for both favorable and unfavorable associations.
- After showing the description: how well does the name fit this service? Very poorly / somewhat poorly / neither / somewhat well / very well / cannot judge.
- How easy or difficult would you find it to say the name aloud? Very difficult / somewhat difficult / neither / somewhat easy / very easy.
- Does the name remind you of another product or organization? If so, which? Open response.
- After a defined intervening task, ask participants to recall the name without showing it again.
- In a later balanced comparison, ask which candidate they would investigate further, including “none” and “no preference,” then ask why.
