Lexicon Branding review

The /brand-naming-results/ page attributes a large, industry-sorted portfolio of names that are unambiguously in market: Sonos, Azure, PowerBook, BlackBerry, Pentium, InDesign, Swiffer, Febreze, Dasani, Vercel, Impossible Foods, Lucid, Braze, CapCut, FreeForm, Oxley, Mi Campo. Individual case-study pages exist per name (e.g. /case-studies/product-names-pg-swiffer/, /case-studies/google-antigravity/). Crucially the credits are not only self-published: a third-party interview piece read first-hand (lennysnewsletter.com) independently credits Lexicon with Sonos, Azure, Windsurf, Vercel, Impossible Foods, BlackBerry, Pentium, PowerBook and Swiffer, and mainstream press (CNN, SF Chronicle) carries the same attribution in headline and search-result text. The portfolio is also not purely old glory - Pentium and PowerBook are 1990s wins, but Vercel, Windsurf, Braze, CapCut, Navan and Google Antigravity are recent, which answers the rubric's concern that decades-old credits say little about the current team.

What this dimension measures: Names the firm created that are verifiably in use by real companies — the strongest evidence this category can offer, because a shipped name is public. Attributed, checkable names for recognisable clients score 4-5. Case studies without the client's confirmation, or a portfolio dominated by decades-old wins, score 2-3: naming credits are notoriously contested and old glory says little about the current team. No attributable names scores 2. Counts for 25% of the weighted score.

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Linguistic screening is the most concretely documented part of the method. /service/linguistic-cultural-evaluations/ states a network of 'over 100 PhD linguists living in 78 countries, with an average tenure of 10+ years', evaluation of names and marketing copy 'on 7 dimensions', delivery 'in less than 7 days', and a named deliverable - 'a comprehensive report summarizing linguistic and cultural assets and liabilities'. That is a real multi-market check with countable scope, not a generic 'we check for bad meanings' line. The naming method itself is thinner on their own site: /service/strategic-brand-name-development/ describes a 'Predict' framework across four named dimensions (structural architecture, phonetic signaling, cognitive processing, competitive proximity) and an 'engineering layer' of proprietary software and cognitive science, but does not publish how territories are developed, how many candidates are generated, or how the shortlist is cut. The volume figures that do exist - roughly 1,000-1,500 names sought, 2,000-3,000 ideas generated, ~90% killed by legal and filters - appear only in a third-party founder interview, not on the firm's own site. Held below Excellent because the generation half of the method is unpublished.

What this dimension measures: A published, specific process: how territories are developed, how many candidates are generated, and — critically — linguistic and cultural screening across the markets the client sells in. A described multi-market linguistic check with named steps scores 4-5. A generic 'we brainstorm then check' description scores 2-3. No described method scores 2. Weight specificity: the firms that do this well tend to write about it in detail, so vagueness is informative. Counts for 20% of the weighted score.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

/service/trademark-clearance/ publishes an actual pre-screening step rather than silence: 'a proven, 3-level screening process that includes identical and similar searches', run by 'a fully staffed trademark team which includes an experienced trademark attorney, paralegals, and a URL search team', with deep-dive searches and detailed reporting available. The handoff boundary is drawn the right way - the page says they 'work closely with our client's counsel and their outside counsel' and names Cooley, LLP as the partner firm that 'is available to assist clients in planning and executing a trademark clearance and registration program'. So screening is positioned as feeding counsel, not replacing it, which is the behaviour the rubric rewards; nothing on the page implies their screening substitutes for legal clearance. Two things keep this off the top band: the boundary is implied by the referral structure rather than stated outright, and the page carries an uncheckable quantified claim ('our clients have an over 98% success rate in their final trademark applications') with no denominator or methodology. Class strategy is not described.

What this dimension measures: What the firm does before candidates reach the client: knockout searches, class strategy, and how it hands off to trademark counsel. A stated pre-screening step with honest boundaries — screening is not legal clearance, and firms that say so score better than firms that imply it is — scores 4-5. Silence on trademark process scores 2, because presenting unscreened names wastes the one thing this purchase buys. Any implication that the firm's screening substitutes for counsel is a red flag, scored 1-2. Counts for 15% of the weighted score.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

Nothing resembling a fee, range or starting point appears anywhere on the site. The page sitemap contains only seven public pages (home, explore, brand-naming-results, our-services, contact-us, privacy, a test page) and none is a pricing page; the seven service pages carry no numbers either. /contact-us/ asks for name, company, country, reason for contact and how you heard about them - it does not ask for a budget or present tiers, so a buyer gets no signal at all before a sales conversation. Scope boundaries are also unstated: the services menu spans naming, brand architecture, linguistics, trademark screening, consumer research and even a venture arm, without saying what a naming engagement itself includes or excludes. The only checkable pricing signal is third-party: the Clutch profile, read first-hand, lists a minimum project size of $10,000-$49,999, a most-common band of $50,000-$199,999 and a maximum band of $1,000,000+. That is useful but it is a directory's disclosure, not the firm's.

What this dimension measures: Naming fees are famously opaque. Published fees, ranges or stated starting points score 4-5. A stated engagement structure without numbers scores 2-3. Nothing at all scores 1-2. Score scope honesty alongside: a firm that states what a naming engagement excludes (identity, strategy, legal filing) is more useful than one implying a full rebrand. Counts for 15% of the weighted score.

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Stronger here: Tanj scores Strong on the same dimension.

For the naming engagement the site never states how many candidate names a client receives, how many rounds are run, what accompanies the shortlist, or what happens if every candidate dies in trademark screening - the contingency the rubric specifically asks about, and the one most likely to matter given the firm's own statement that legal filters eliminate the large majority of candidates. The Swiffer case study is outcome-focused and contains no counts, rounds or process detail. Partial credit only because one adjacent service does define its output: the linguistic evaluation promises 'a comprehensive report summarizing linguistic and cultural assets and liabilities' assessed on 7 dimensions, and trademark work can 'prepare detailed reporting'. A buyer still cannot know the size of the shortlist before contracting.

What this dimension measures: What the client actually receives: how many candidate names per round, how many rounds, what accompanies them (rationale, screening results, territory map), and what happens if every candidate dies in legal. A stated structure with a stated contingency for clearance failure scores 4-5. Undefined deliverables score 2 — in a category where the visible output is a shortlist, not knowing its size before contracting is buying blind. Counts for 10% of the weighted score.

Evidence: inferred — our reading of indirect evidence, not a documented fact.

Stronger here: Tanj scores Strong on the same dimension.

One published timeline exists and it is specific and plausible - the Language Suite returns results 'in less than 7 days'. But that covers the linguistic evaluation only. No kickoff-to-final-candidates timeline is published for the naming engagement itself on any of the pages read, and the contact page sets no expectation. Indirect signal that projects run long: several Clutch reviewers, read first-hand, cite a wish for faster turnaround while saying quality took priority - which implies multi-week engagements but is not a disclosure by the firm. Scored between the rubric's bands because scheduling is partly, not wholly, undisclosed.

What this dimension measures: Stated engagement timelines from kickoff to final candidates. Published, realistic timelines score 4-5 — naming projects run weeks, not days, and honesty beats speed. Undisclosed scheduling scores 2. Counts for 10% of the weighted score.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

The Clutch profile, read first-hand, carries a 5.0 out of 5 rating across 20 reviews that Clutch marks as verified through its human-led verification process, with project-size bands attached. For a boutique naming consultancy running small annual project counts, 20 verified client reviews is a substantial base rather than a thin one. Reviewers speak to specifics a buyer can use - linguistic sophistication across cultures, thorough trademark and legal clearance, project management - and the criticisms are visible too (speed, scalability, limited late-stage support such as URL acquisition, premium cost rated 4.0-4.5). Named client references also appear on the firm's own pages: Sonos founder John MacFarlane on the homepage and Harry West of Columbia University on the Swiffer case study.

What this dimension measures: Independent, verified reviews or checkable client references. A substantial verified base scores 4-5; a handful scores 2-3; none scores 2. Low weight deliberately: boutique naming firms run on referral and small annual project counts, so a thin review footprint is normal and the in-market names matter more. Counts for 5% of the weighted score.

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.