The Method: Name Readiness Score

A score is worth as much as the method behind it. The Name Readiness Score (NRS) is the 0-to-100 number NameSapiens assigns to every name it evaluates, and this page explains where it comes from: the seven dimensions, the weights, the thresholds, the rules that can block a name regardless of the total, the sources consulted and the limits we prefer to state rather than hide. The method is versioned: the current version is nrs-1.0.0 and every report states the version it was generated with.

The seven dimensions and their weights

The total score is the weighted sum of seven dimensions. The weights reflect a deliberate choice: trademark conflict risk weighs more than how pretty the name sounds, because an unusable name has no need to be beautiful.

Dimension	Weight	What it measures
Strategy	15	Fit with the brief and the positioning
Phonetics	10	Pronounceability, rhythm, memorability
Trademark	25	Risk of conflict with existing trademarks
Domain	15	Availability and quality of relevant extensions
Linguistics	10	Meanings, ambiguity and portability across languages
AI Visibility	15	How AI models and search engines interpret the name
Defensibility	10	Distinctiveness and ability to be protected over time
Total	100

Strategy (15)

Measures how well the name answers what you asked for. The strategist and the critic assess consistency with the chosen semantic territory, the register (descriptive, evocative, coined), the target length and the distance from the competitors you named. A beautiful name that ignores the brief loses points here, nowhere else.

Phonetics (10)

We analyse syllable structure, difficult consonant clusters, natural stress and the risk of divergent pronunciations in the brief's languages. A name that every speaker pronounces differently costs you on every phone call and every word-of-mouth recommendation.

Trademark (25)

The dimension with the greatest weight. We query public trademark databases for identical or similar signs, compare the Nice classes stated in the brief with those of the marks found, and estimate visual and phonetic similarity. The score drops in proportion to closeness and to overlap in class and territory. A serious conflict does not merely lower the score: it triggers a hard gate.

Domain (15)

We check via RDAP the status of the extensions you flagged as relevant plus a standard set (.com, .it, .eu and the main generic TLDs). A free domain scores full marks; one that is taken but parked or for sale scores less; one with an active operating business scores lower still. If the TLD you declared mandatory is taken, a cap applies to the total.

Linguistics (10)

The linguist looks for unwanted meanings, embarrassing homophones, negative associations and transliteration problems in the requested languages. We also check written legibility (double letters, ambiguous sequences) and how easily the name can be spelled from dictation.

AI Visibility (15)

A dimension traditional methods do not have. We ask several AI models what they think the name means, which entities they associate it with, and whether they confuse it with something that already exists. We check for the term in open knowledge bases and search results. A name that assistants systematically attribute to another company starts at a disadvantage, because more and more searches go through them.

Defensibility (10)

Estimates how distinctive, and therefore protectable, the name is: a term that describes the product rarely becomes a strong trademark; a well-built coinage can. We also count how many similar trademarks and domains already exist, which measures how crowded the territory is.

The thresholds

Score	Verdict	Meaning
≥ 75	Go	No obvious obstacles in the sources consulted; a candidate to take to a professional clearance search
50 – 74	Caution	One or more significant risks; usable with awareness or with changes
< 50	Stop	Serious problems; proceeding is not recommended

Thresholds are fixed in the current version and shown in the report next to the score.

Hard gates

Some conditions are too serious to be diluted in a weighted average. When they occur, they override the total:

• Serious trademark conflict (identical or near-identical sign in an overlapping class and relevant territory): the verdict is Stop, whatever the sum of the other dimensions.
• Mandatory TLD taken (the extension you declared indispensable is registered and active): the total score is capped at 69, so the name cannot reach Go.
• Offensive meaning in one of the brief's languages: the verdict is Stop.

Hard gates are flagged explicitly in the report together with the source that triggered them, so you can verify and, if it turns out to be a false positive, request a re-evaluation.

The sources

Source	Use
EUIPO	European Union trademarks
TMview	Aggregated national and EU trademarks, including Italian ones (UIBM)
WIPO Global Brand Database	International and participating national trademarks
RDAP	Domain status, dates and registrar
Wayback Machine	Historical use of a domain
SearXNG	Privacy-respecting web metasearch, for presence and context of the term
Wikidata	Known entities, meanings and homonyms
Multiple AI models	Interpretation of the name, associations, AI Visibility

All sources are queried at evaluation time; the report records the date and time of each query and flags any source that was unavailable.

Declared limits

• The sources are not exhaustive. We do not cover every trademark office in the world, and not all are updated at the same frequency.
• Italian trademarks are consulted only through TMview, not directly in the UIBM database. Recent filings may not yet be visible.
• This is not a professional clearance search. We do not consider unregistered trademarks, company names, trade names or copyright, and we do not apply the judgement of a trademark attorney.
• The subjective dimensions (strategy, phonetics, linguistics) are assessed by AI models against explicit criteria, but remain estimates.
• The score is a snapshot. Domains and trademarks change every day; a Go today is no guarantee for tomorrow.

Pipeline version

The current version of the method is nrs-1.0.0. The version changes when we modify weights, thresholds, hard gates, sources or the calculation logic of a dimension; it does not change for updates to the underlying AI models that leave the criteria unchanged. Every report states the version it was produced with and is never recalculated retroactively.

Changelog

nrs-1.0.0 — 2026-10-24

• First public version of the Name Readiness Score.
• Seven dimensions with weights: Strategy 15, Phonetics 10, Trademark 25, Domain 15, Linguistics 10, AI Visibility 15, Defensibility 10.
• Thresholds: Go ≥ 75, Caution 50–74, Stop < 50.
• Hard gates: serious trademark conflict (Stop), mandatory TLD taken (cap 69), offensive meaning (Stop).
• Sources: EUIPO, TMview, WIPO Global Brand Database, RDAP, Wayback Machine, SearXNG, Wikidata, multiple AI models.
• Report with per-dimension evidence, query timestamps and flagging of unavailable sources.