Three finalist names with full rationale for a financial wellness app targeting millennials.
This is a trimmed sample. The full Name Lab runs all five naming territories, generates 20-30 candidates, applies six tests to every shortlisted name, and delivers a defended recommendation with the names to cut and why.
Five moves. Three gates. No name generation before the brief is built.
The most common failure of AI naming is generating 50 names in the first turn. The skill refuses. The Naming Brief locks before any candidate appears.
Build the Naming Brief.
What's being named. The audience, the category, the values, the constraints. The criteria for "this name holds." If you came with names in your head already, they go into the Storm Pool — but the Brief still runs.
Generate across five territories.
Storms candidates across all five naming territories. Maps them by territory, by feel, by ownability. Shortlists the strongest. Halts so you can redirect.
The call. The runner-up. The cut.
Tests the shortlist against six criteria. Picks the recommendation. Names the runner-up. Tells you which to cut and why. Each one defended.
Moves 02 and 04 run inside the skill silently. Three explicit outputs and gates is enough.
Every name belongs to one of five categories.
Naming chaos comes from generating across all five at once without knowing why. The skill maps every candidate by territory, so the trade-offs are visible before you pick.
Single Real Word
A word that already exists, used as a name.
e.g., Apple, Amazon, Stripe
Word Fusion
Two real words pushed together into one name.
e.g., Facebook, Salesforce, Mailchimp
Invented
A made-up word with no prior dictionary meaning.
e.g., Spotify, Verizon, Kodak
Twisted
A real word with a deliberate spelling twist.
e.g., Lyft, Tumblr, Flickr
Initials
Initialism or acronym, sometimes derived from a longer phrase.
e.g., IBM, HSBC, BMW
A name has to survive all six.
Once the storm produces a shortlist, every candidate runs the same six tests. The recommendation is the name that passes the most, defended with reasoning.
Speak
Can a stranger say it on first read, without explanation?
Stick
Will the audience remember it 48 hours later without prompting?
Values
Does the name signal something true about the brand's values?
Own
Domain, trademark, social handles — is it ownable?
Audience
Does it land with the specific buyer, not with the founder's friends?
Visual
Will it work as a wordmark, on a tab, on a t-shirt, in a screenshot?
Three sequential outputs. Each one a gate.
Brief first. Storm and Map second. Final Recommendation third. The skill halts after each one.
The Naming Brief
After diagnosis. Before generation.
What's being named, the audience, the values, the constraints, the success criteria. The candidates you came with go in here too. Halts the work until you confirm.
The Storm and Map
Candidates across five territories.
A full storm pool, sorted and mapped by territory and feel. Shortlist of the strongest. Halts so you can redirect or narrow.
The Final Recommendation
The call, the runner-up, the cuts.
The recommended name with full defence (which tests it passed, which it didn't, why it still wins). Runner-up with same logic. Names to cut with reasons.
Brief first. Even if you came with names already.
Built for the people doing the work, not watching it.
- Founders naming a company, a product, or a service for the first time.
- Founders renaming because the current name is holding the brand back.
- Founders who have three candidates and can't decide.
- Agency owners who need a real naming methodology for client work.
- Anyone naming an internal program, a podcast, a book, a workshop, a newsletter.
- ×People who want 50 random names with no brief.
- ×Anyone who wants the skill to skip the brief because they've "already done the thinking."
- ×Pre-product founders without enough clarity to write the brief yet.
- ×People who want the AI to guarantee trademark or domain availability — those checks live outside this skill.
- ×Anyone hoping for a magic name with no inputs.