How NameClicked scores a name
NameClicked scores every name out of 100 across three questions: is it a good name, can I get it, and can I build on it? Around twenty criteria feed those scores, and serious problems such as a matching trade mark cap the score so they can't be averaged away.
By Phil Martin at Incyworks Ltd · Published · 6 min read
Every name in NameClicked gets an overall score out of 100 and a recommendation. Behind that number are three separate questions, around twenty criteria and a few rules that stop serious problems from being averaged away. This guide explains all of it, so you can decide how much weight to give a score and when to look past it.
Three questions
Each name is scored on three dimensions, which are combined into the overall score:
| Dimension | The question | Weight |
|---|---|---|
| Desirability | Is it a good name? | 40% |
| Legal usability | Can I build on it? | 40% |
| Availability | Can I get it? | 20% |
Desirability and legal usability carry the most weight. A name that's easy to remember but can't be protected, or one that's clear legally but forgettable, isn't a strong name.
The criteria
Each dimension is made up of criteria. Every criterion is scored from 0 to 100 with a short explanation, and some count for more than others within their dimension. You can see each name's criteria by opening How this was scored on its result.
Desirability: is it a good name?
- Memorability: Is it likely to stick?
- Spellability: Could someone spell it after hearing it?
- Pronunciation: Is it obvious how to say it?
- Distinctiveness: Does it stand apart from category language?
- Searchability: Could people find it by searching?
- Competitor clearance: Is it clear of the competitors you listed?
- Relevance: Does it make sense for what you're building?
- Suggestiveness: Does it hint at something useful without being literal?
- Imagery: Does it create a useful mental picture?
- Emotional appeal: Does it feel right for the audience?
- Room to grow: Could it stretch to future products or sub-brands?
- Unwanted meanings: Are there negative or awkward readings?
- Language check: Any obvious problems in other languages?
Legal usability: can I build on it?
- Apparent protectability: Is it distinctive enough to be protectable?
- Trademark conflicts: Are there similar registered marks?
- Company-name conflicts: Are there similar companies on the register?
Availability: can I get it?
- Domains: Are suitable domains available?
- Social handles: Are matching handles available?
- App stores: Is it clear in the app stores?
For the thinking behind the desirability criteria, see what makes a good product name.
Where the evidence comes from
Each criterion shows its source, because a guess and a register search deserve different trust:
- Algorithm: fast, rule-based checks of the name itself, such as length, syllables, letter patterns, generic words and similarity to competitors you've listed.
- AI judgement: an AI model's judgement of things rules can't measure, such as relevance, imagery, emotional tone, room to grow and unwanted meanings.
- Live data: results fetched from outside services at the time of the check.
- Official register: records from authoritative sources, such as the domain registries and the trade mark register.
Preliminary scores and Full Assessments
Free results are preliminary. They score desirability in full, but they don't search any registers. Legal usability is estimated from how distinctive the name looks, and availability is left out of the overall score until it's been checked, rather than being guessed. Every free result lists the checks that haven't been run.
A Full Assessment adds live domain checks and UK trade mark screening, and runs a fresh, deeper AI assessment. Its scores can go up or down. A name that looked promising can drop sharply if a similar trade mark turns up.
Serious problems aren't averaged away
Averages can hide things. A name with brilliant scores everywhere except one fatal flaw could still average out well. To prevent that, certain problems cap the overall score and set a minimum level of concern:
| Problem | Score capped at |
|---|---|
| Contains a word or fragment that is likely to be read as offensive. | 20 |
| Looks or sounds almost identical to a competitor you listed. | 35 |
| May carry an unwanted or negative meaning. | 40 |
| An identical or near-identical trade mark appears in a relevant class. | 30 |
| Similar trade marks appear in relevant classes. | 55 |
| Pronunciation is likely to be unclear to many people. | 50 |
| Built from generic category words, which is usually hard to protect. | 55 |
These show as critical flags on the name's result, so you can see exactly why the score is held down.
What the recommendation means
| Overall score | Recommendation |
|---|---|
| 80 or more | Strong candidate |
| 65 to 79 | Promising |
| 50 to 64 | Worth investigating |
| 35 to 49 | Significant concerns |
| 0 to 34 | High apparent risk |
If a critical flag calls for a more cautious recommendation than the score alone would give, the more cautious one is shown.
What a score can't tell you
A score also can't tell you whether a name feels right to you. Use it to rule out weak and risky names and to understand the trade-offs, then trust your judgement on the rest. To get started, see how NameClicked works.