Construct names that feel real
Made-up words that still read as real, and are usually free. Scored, not picked on vibes.
How the engine chooses them is in the doctrine.
Availability is not checked here — a generated name is scored for how naturally it reads, nothing else. Every candidate must still be verified at the registry before you buy.
Why we make up words
Real words
Maximum phonological density, and all taken — dictionary words are exhausted across .com and .ai. Buyability tends toward zero.
Free but forgettable
Free, but there is nothing to hold on to. A word like tidezo or dawnva sits near no word you already know, so memory and trust have nothing to attach to.
Made up, but familiar
Constructed words that feel real — planity reads as plan / sanity / plain — while remaining legally empty. This is where we generate.
How each name is scored
score = how much it reads like English + 0.35 · log(how many real words it sits near)
The first term measures how much the name reads like ordinary English. The second counts real dictionary words within two edits of it — a name that sits near many real words is remembered and trusted like one.
The evidence
Vitevitch & Donoso, 2012
Words that read more like real English get a higher purchase intent.
Vitevitch–Luce
High neighbourhood-density nonwords are processed and trusted like real words.
BrandScout, 2026
Made-up words that follow English patterns scored twice as trademarkable and were recalled 31% more after 48 hours than real words.
The construction doctrine — added 2026-09-16
The construction method is published in full, with the research it came from.
Read the construction method →