Brand Naming · Part 1 of 5
8 min read

Brand Naming Has Two Unsolved Problems.
Almost Nobody Addresses Both.

Most founders start where we did — asking AI for names, getting a list in seconds, then realising there was no principled way to evaluate any of them. That frustration is the starting point for understanding what is genuinely broken about how brand naming works today.

Series
Origin

I'm an engineer and a founder — not a brand naming expert, not a linguist. Like most new founders, I started where everyone starts: asking AI to help name my company. It worked, sort of. I had names. What I didn't have was any way to connect them to anything. Why does this name fit my brand? What colors go with it? What should the logo feel like? Every decision felt disconnected from the name itself, and nothing looked or felt cohesive. So I built a personal naming lab to solve that problem — and in building it, I started understanding how strategic branding actually works.

The Two Problems Nobody Solves Together

Brand naming has two distinct layers. Most tools address neither of them properly. Almost none address both simultaneously.

01

Evaluation

When you have a shortlist of candidate names, how do you actually know which one fits your brand? Most tools offer aesthetics or gut feel — not a data-backed reason why one name is a stronger strategic fit than another.

02

Connection

Even when you find a name you love, there is no structured bridge between that name and the visual identity it should produce. Colors, logo geometry, tagline — none of it is derived from the name itself. It's derived from a mood board, or a brief written in adjectives.

The designer starts from scratch. The brief is written in adjectives. And the name — the thing that was supposed to anchor everything — ends up floating free of the identity it was meant to ground.

"Naming feels like guesswork. It doesn't have to."

The Foundation Already Exists in Academic Research

The starting point was an established body of academic research on sound symbolism — the principle that specific phonemes create predictable perceptual effects in the human brain. This isn't new science. It's documented in peer-reviewed work by Klink (2000), Pathak et al. (2017–2020), Nielsen & Rendall (2011), Motoki et al. (2023), and Murdock (1962).

The engineer's question was straightforward: if this research is real and measurable, can it be made computable? Can a system evaluate any candidate name against a defined brand profile — and produce a transparent, explainable score?

The answer was yes — but it required combining three methodologies simultaneously, not simply citing papers.

Research‑Backed
Directly implements specific peer-reviewed findings

Each phonetic rule in the framework traces back to a specific published study — validated against the actual phoneme patterns of real brands in each market category, not against theory alone.

Data‑Driven
Derived from empirical phoneme analysis of real market-leading brands

Not what theory suggests a Modern brand should sound like — but what Apple, Stripe, Slack, and Pivot actually have in common at the phoneme level. The profiles are grounded in what already works in market.

Semantically‑Mapped
AI-assisted synthesis connecting sound symbolism to brand strategy

Grounded in real brand phoneme analysis to ensure market relevance — not invented theory. Each mapping is validated against the empirical data before entering the framework.

The Output: Seven Computable Brand Style Profiles

Combining these three methodologies produces seven Brand Style Profiles — each one a precise, machine-readable instruction set: preferred consonants, preferred vowels, target syllable structure, phonetic flow type, accent placement strategy, and Jungian archetype mapping.

The Seven Brand Style Profiles
Luxury Modern Playful Professional Technical Rugged Discovery

Each profile is not a subjective description — it is a machine-readable rule set that any candidate name can be evaluated against, producing a score with full transparency into why each point was awarded or deducted. The Modern profile, for instance, defines 11 preferred consonants, 6 preferred vowels, a 1–2 syllable target, Harsh phonetic flow, and Front-weighted accent placement.

The naming gap — visualised

THE OLD WAY Generate names — AI / thesaurus gut feel Pick a name you like disconnected brief Design identity from scratch THE SYSTEMATIC WAY Define Brand Style Profile phonetic fit score Score names against profile Sound Print → identity brief Identity derived from the name

These profiles are not subjective descriptions. They are machine-readable rule sets that any candidate name can be evaluated against — producing a score with complete transparency into exactly which phonemes are on-brand, which are off-brand, and why every point was awarded or deducted.

The gap between a name you like and a name that fits your brand is not a creative problem. It is a measurement problem. And measurement problems have solutions.

The next post in this series covers the academic foundation in detail — what sound symbolism research actually says, which studies the framework draws from, and how phonemes create predictable perceptual effects that can be mapped to brand perception.

Try Klexaro — Free

Evaluate any brand name against 7 phonetic brand profiles. All analysis runs entirely on-device — no data leaves your phone. The resulting Sound Print can then brief any external AI tool for identity generation.

Download on App Store →
Next in series What Is Sound Symbolism in Brand Naming? →