In the previous post, we covered the five phonetic dimensions that make up a name's Sound Print — and showed how established global brands score when put through the framework. This post covers the scoring system itself: exactly how those phonetic properties translate into a single, explainable number.
The system works in four sequential steps. The first two establish what a name is acoustically and what a chosen brand profile requires. The third and fourth independently score the name on two separate axes — strategic fit and practical usability — before combining into a final result.
"Complete transparency into exactly which phonemes are on-brand, which are off-brand, and why every point was awarded or deducted."
Step 1 — The Sound Print
Every name is first decomposed into its individual phonemes — its unique phonetic fingerprint. This is the raw acoustic data the rest of the system works from. The decomposition is not an approximation: it is executed by a three-tier pronunciation engine that queries the CMU Pronouncing Dictionary (133,854 entries), then a curated database of 2,800+ real brand names, and for invented words, a 6MB on-device neural network trained specifically on brand-like neologisms.
The resulting phoneme sequence is the Sound Print. Everything downstream — composition scoring, structural modifiers, profile matching — operates on this data, not on spelling or visual form.
Step 2 — The Brand Profile
Separately, each of the seven Brand Style Profiles — Luxury, Modern, Playful, Professional, Technical, Rugged, Discovery — is defined as a precise phonetic instruction set. This is the target the name's Sound Print will be measured against.
The Modern profile, as a concrete example, defines: 11 preferred consonants (P, T, K, F, S, SH, V, Z, CH, L, R), 6 preferred vowels (IY, EY, AE, AY, IH, AH), a target syllable count of 1–2, Harsh phonetic flow, and Front-weighted accent placement. Every profile is defined with equivalent precision. These are not descriptions — they are machine-readable rule sets.
Step 3 — Strategic Fit Score (70%)
The Sound Print is compared against the Brand Profile. The system first calculates a Composition Score — the percentage of the name's phonemes that match the profile's preferred sounds. Then a Structure Modifier is applied, based on whether the name's syllable count, phonetic flow, and accent placement match the profile's structural targets.
These two components combine into the Strategic Fit Score, weighted at 70% of the overall result. The output is not just a number — it identifies precisely which phonemes are on-brand, which are off-brand, and what structural penalties or bonuses were applied.
Step 4 — Technical Quality (30%)
Independently from strategic fit, the name is evaluated across four practical usability pillars — each weighted equally at 25% of the Technical Quality score. These assess whether the name functions in real-world commercial conditions, separate from how well it fits the chosen brand profile.
Optimal Length
Does the name meet established conventions for recall? Length is evaluated against research-backed norms for brand name memorability — not arbitrary character counts, but phoneme-level assessment of cognitive load.
Pronunciation
How easily can a new customer say the name correctly on first encounter? The pronunciation engine assesses phoneme complexity, consonant cluster difficulty, and vowel ambiguity. A name that is hard to say is hard to share.
The Bar Test
How clearly is the name understood when spoken in a noisy environment — across a bar, over a phone, at a conference? This tests acoustic distinctiveness: how well the name's phonemes survive degraded listening conditions.
Safety Check
Does the name carry any unintended negative meanings? The collision detection layer runs 9,800+ checks across 27 languages — screening for profanity, sensitive terms, and negative sentiment associations. All checks run entirely on-device.
The Final Score
Strategic Fit (70%) and Technical Quality (30%) combine into a single overall score. The weighting reflects the framework's underlying position: strategic phonetic alignment is the primary determinant of brand name quality, but a name that scores perfectly on fit and fails on usability is not a viable candidate.
The framework's output is not a black-box score. Every point in the final result traces back to a specific phonetic property: which consonants matched the profile, which didn't, whether the syllable structure hit the target, whether the flow type aligned. A brand strategist presenting this to a client can explain every number — because the system produces the explanation alongside the score, not instead of it.
The next post in this series covers the pronunciation engine in detail — how a three-tier system accurately extracts phonemes from both real words and invented brand names, and why accuracy at this layer is the critical dependency the entire scoring system rests on.
See the Score for Your Name — Free
Run any brand name through the full four-step system. All analysis runs entirely on-device — phoneme decomposition, strategic fit scoring, technical quality assessment, and collision detection. No data leaves your phone.