In this guide
Evaluate generated names by testing what people understand, say, remember and read, then compare the results using the same five criteria. Keep health promises, ingredient implications and trademark questions outside the numerical score: strong recall cannot resolve a misleading implication or a legal concern.
For a handmade skincare line, start with the products you actually intend to sell and the buyers you want to reach. Use that brief to organize candidates, test them under comparable conditions and document your decision. The method below proposes an equal-weight scorecard out of 25, with an observable exercise behind every rating. It also explains how to keep generator settings, naming judgments and preliminary availability checks distinct.
Define what the name needs to communicate
Write a short naming brief before evaluating candidates. Give the name a specific job: what should an unfamiliar buyer reasonably expect after encountering it?
Record four things:
- Intended buyers: who you want to reach and what matters to them when choosing skincare.
- Desired tone: the qualities you want the name to suggest, expressed in plain language.
- Launch products: the actual products and formulas the name will represent.
- Future possibilities: additions you would realistically consider, rather than every imaginable category.
Ingredients used in the line can provide naming material. If you take that direction, check whether the resulting name creates an accurate expectation across the products it will cover.
A brand-name or business-name generator may help you explore candidates. Before choosing one, inspect its documented inputs and decide whether they suit your brief. Keep the brief as a separate working document; do not assume a tool accepts particular instructions or provides particular controls.
Turn generated results into a manageable shortlist
Keep a batch log containing the tool used, the exact inputs submitted and any documented settings selected. Save the candidates you want to evaluate alongside that record. This makes later comparisons more useful than a disconnected collection of favorite words.
If you generate another batch, change one documented input or setting at a time. Note the change and compare the results against the same brief. Describe only what you observe in those batches; a different result does not by itself explain how the generator works.
Retain a small enough shortlist that you can test every candidate consistently. Combine duplicate entries and write one sentence explaining why each remaining name deserves consideration.
Treat each generated string as a starting point. Record any separately performed domain or trademark search in its own log, with its scope and findings. A naming result alone should not be recorded as a completed availability check.
Screen for health promises and misleading ingredient cues
Before spending time on preference tests, inspect each candidate for implications that need investigation. Ask whether its wording could suggest disease treatment, a guaranteed outcome or an ingredient that the formulas do not contain.
For each concern, write down the exact wording and the expectation it might create. Compare that expectation with the product information and supporting evidence you have. If an ingredient appears in only part of the range, consider whether buyers might reasonably assume it appears throughout the line.
Use a separate review record:
| Concern | Question to resolve | Next action |
|---|---|---|
| Health outcome | What result could buyers understand this name to promise? | Compare the implication with product evidence and seek appropriate review. |
| Ingredient cue | Which products actually contain the suggested ingredient? | Check formulas and the intended scope of the name. |
| Market-specific question | What needs assessment in the intended sales market? | Obtain advice appropriate to that market. |
Remove candidates with serious problems. Hold unresolved candidates for review before advancing them; do not convert these concerns into small deductions from an otherwise attractive score.
Build a five-part scorecard
Choose weighting according to the naming context before comparing results. For this practical starting method, give all five criteria equal weight and rate each from 1 to 5.
| Criterion | Observable test | Evidence to record |
|---|---|---|
| Audience fit | Ask what the name suggests before explaining the line. | Associations that match or conflict with the brief. |
| Pronunciation | Test reading aloud and spelling from speech separately. | Hesitation, competing pronunciations and spelling errors. |
| Memorability | Request later recall without prompts. | Accurate recall, near misses and confusion. |
| Packaging readability | Show actual-size label mockups. | Whether readers locate and read the brand name. |
| Future range fit | Pair the name with plausible additions. | Where it still fits and where explanation becomes necessary. |
Use 1 for repeated difficulty, 3 for mixed results and 5 for consistent success. Scores of 2 and 4 represent intermediate results. Support each rating with a short observation, such as “several readers needed help locating the brand name.”
Leave an untested row blank. Calculate a total out of 25 only after all five tests are complete. Keep review concerns in adjacent fields labeled with their status and next action; they do not belong in the total.
Test audience fit through associations
Show each candidate to prospective buyers before describing the business. Ask open questions: “What would you expect this name to sell?” and “What tone or positioning does it suggest?” Let participants supply their own language before offering choices.
Compare those answers with the brief. If you wanted an approachable everyday line but participants repeatedly imagine a different positioning, record the mismatch. Agreement that a name sounds pleasant is less useful for this test than a clear account of what it leads people to expect.
Keep the presentation comparable across candidates, including typography and the amount of explanation supplied. Record recurring associations rather than selecting only favorable comments.
For intended languages and markets beyond your own, ask suitable speakers about unwanted meanings or associations. Leave that part of the review unresolved until it has actually been checked. Do not infer suitability from your personal reaction alone.
Test pronunciation and spelling separately
Run two exercises with people unfamiliar with the candidates. First, show a written name and ask the reader to say it aloud. Avoid demonstrating your preferred pronunciation beforehand. Note hesitation, requests for help and different readings.
In a separate exercise, say the candidate once and ask a listener to type or write it. Use people who have not already seen that spelling where practical. Record the result exactly, including errors, rather than correcting it before you take notes.
Keep the observations separate even though they inform one pronunciation score. A candidate may be straightforward to read but difficult to spell after hearing it. Another may produce several spoken versions despite being remembered visually.
Base the rating on both exercises and explain the tradeoff in the notes. If results are mixed, identify which part caused difficulty instead of compressing everything into a vague judgment that the name is “easy.”
Check what people remember
Give candidates comparable exposure, then ask participants to recall the names later without prompts. One practical approach is a brief initial showing followed by a recall request the next day. Keep the exposure and recall interval consistent across the candidates being compared.
Ask for remembered names before displaying the shortlist again. Record accurate recall, approximate versions and confusion between candidates or with other names. Keep later recognition separate from unaided recall: selecting a familiar-looking option answers a different question from producing the name independently.
Use the results to support the memorability rating, retaining the actual responses alongside the score. If someone remembers a candidate because it created an unwanted association, carry that concern back to the audience-fit review.
Treat a small informal exercise as directional evidence. Describe what these participants remembered under these conditions, without presenting the result as a conclusion about every prospective buyer.
Try the name on real packaging dimensions
Create label mockups at the actual dimensions intended for your containers. Include the smallest planned label and enough surrounding wording to show the brand name beside the product name. Use provisional copy where necessary, clearly marked in your working files.
For the first comparison, hold typography, placement and other presentation choices as consistent as practical. Show the mockups at actual size and ask readers to identify the brand, identify the product and read the name without explanation.
Record what happens: readers may locate the wrong text, hesitate over letter shapes or need help distinguishing the brand from the product. Use those observations for the readability score.
If a candidate struggles, try a reasonable layout adjustment and repeat the relevant check. Record which version was tested. A result from one mockup supports a decision about that presentation; it does not establish that every possible design will behave the same way.
Check room for future products
Place each candidate beside the future additions in your brief. Depending on your plans, these might include a cleanser, a body product or a formula using different ingredients. Treat them as planning scenarios, not announced products.
Ask whether the name still gives an accurate impression without a corrective explanation. If it emphasizes one ingredient, consider a future formula without that ingredient. If it closely describes one launch product, try it beside another product you could realistically introduce.
Record where the candidate fits comfortably and where participants need clarification. These observations support the future range fit score.
Keep the role of the name clear. A narrowly descriptive candidate might suit an individual product even if you decide it cannot represent the whole line. Do not give that candidate a strong range score merely because it works well for its original product.
Run preliminary trademark and domain checks
Search for identical names and for similar-looking or similar-sounding variants in the markets and product categories relevant to your plans. For a UK launch, include relevant trademark classes and spelling or sound variations. If you are unsure which classes or records to examine, make that a question for a qualified adviser.
Keep a search log with the date, search terms, markets checked and results needing investigation. Record potentially relevant existing brands and business names as well as trademark findings. An absence of obvious results in a limited search should remain a limited finding.
Check domains and intended social platforms separately. Write down workable domain and handle arrangements rather than treating one preferred spelling as the entire decision.
An available domain does not establish legal permission to use the name. Likewise, preliminary searching is distinct from professional trademark clearance. Keep unresolved similarities and clearance questions visible beside the scorecard, and investigate them before committing to the name or printing packaging.
Choose finalists with a written rationale
Compare completed scorecards alongside the observations and review records. First remove candidates with serious problems and hold those with unresolved concerns. Then compare the remaining tradeoffs rather than automatically choosing the highest total.
Use a compact decision record for each finalist:
- Observed strengths: which tests went well, with examples.
- Observed difficulties: what participants struggled with and under which conditions.
- Completed score: all five ratings and the total out of 25.
- Outstanding review: claims, ingredient implications, language associations or trademark questions still requiring attention.
- Next action: the specific check needed before commitment.
A useful rationale might explain that one candidate produced more accurate associations and stronger recall, while another was easier to spell after hearing it. Keep the explanation tied to your observations.
If a close decision rests on one uncertain criterion, repeat or improve that test rather than adding arbitrary points. Select a finalist whose strengths and limitations you can explain, and resolve outstanding questions before approving the name for packaging.