KoolavAll agenciesBy industryBy disciplineReportsHow we vet
All agencies · Localization & Transcreation

Creative Translation review

CONDITIONAL for Localization & Transcreation

Worth a conversation about Localization & Transcreation once the caveats below are settled.

A 23-year-old London transcreation specialist with genuinely attributable campaign work for PayPal, Samsung and Unilever, but no published pricing, turnaround, certification or translation-memory ownership terms, and zero readable independent client reviews.

No published price we can link to. We do not estimate one — ask on the call, and see the pricing-transparency line in the scores below.

Score 2.85/5Confidence: lowLast evaluated 2026-08-26Website

How it scored

Every dimension is scored against this discipline’s published rubric. Open one to see the claim it was scored on, what the rubric measures there and how much it weighs, and where the evidence came from.

Transcreation and marketing practice evidenceStrong

Marketing transcreation is the core business, not a service-page add-on. The site publishes nine case studies naming Canon, Samsung, Madame Tussauds, PayPal, Airbnb, Virgin Atlantic, F&F Clothing, Unilever and Monotype, with described campaign work: the PayPal case describes script translation, tagline adaptation, voice-over casting and recording supervision, subtitles, motion-graphics adaptation and legal copy across 45 markets with over 300 contributors including 'linguists, legal specialists, writers... and comedians'; the Samsung case describes building a voice-and-tone rulebook for Bixby, copywriter training workshops, three rounds of focus-group testing per market across five markets, and over 10,000 response variations. The homepage draws the transcreation distinction explicitly ('We don't slavishly translate word for word, we use cultural insight to adapt creative content in any language'), which the rubric treats as a signal the practice is real. Held at Strong rather than Excellent: the work is entirely self-attested with no client quotes, named references or independent corroboration, the case studies are short (200-250 words), scope is given as markets and regions rather than stated language pairs, and no case study reports a measurable outcome.

What this dimension measures: Is marketing work a real, named practice with attributable campaign or content work — or a service-page claim on a general translation business? Named marketing clients with described campaign work across stated language pairs scores 4-5. A marketing services page with no attributable work scores 2-3. Distinguish transcreation (copy reworked for the market, briefed like creative) from translation with a marketing label; vendors that explain the difference in their own materials tend to actually do it. Counts for 25% of the weighted score.

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

Linguist model and quality processAdequate

A model is described with named review layers rather than a bare 'native speakers' claim: the homepage lays out five phases (Concept, Campaign Adaptation, Production, Approval, Performance) with copy validation and brand guardianship sitting in Approval; the services page lists pre-launch native-speaker reviews, linguistic editing and in-market research; case studies reference in-market teams engaging writers before translation begins. Machine translation is disclosed honestly rather than hidden - the Creative AI page sells 'AI-enabled' work as neural MT combined with expert linguists under a 'Human at the Heart' framing, run on 'secure managed platforms' with a 'full audit trail' of AI usage, and warns against 'unmanaged or unsecured public AI platforms'. What is missing is the verifiable half: no ISO 17100, ISO 9001, ATC or ELIA claim appears anywhere on the site (checked homepage, about, services, agency, brand, contact), and the Community page is a recruitment form that publishes no vetting criteria, qualification requirements or credentialing standards. The About page states '3,147 of us' globally without distinguishing employees from freelance network. Capped at Adequate under the corroboration rule: the top band of this dimension requires verifiable certification and there is none to verify.

What this dimension measures: Who does the work and who checks it: in-country professional linguists versus anonymous pools, named review layers, and certifications that can be verified (ISO 17100, ISO 9001 — check the certificate claim, not the badge). A described model with named QA steps and verifiable certification scores 4-5. 'Native speakers' with no process detail scores 2-3. How the vendor positions machine translation matters: disclosed MT-plus-post-editing with stated boundaries is honest and scores on its own terms; undisclosed MT sold as human work is a red flag scored 1-2 where evidenced. Counts for 20% of the weighted score.

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

Scope and pricing transparencyWeak

No rates, no pricing model, no per-language minimums and no billing mechanics are published anywhere. Homepage, services, agency, brand and contact pages all route to 'Request a quote' or hello@creativetranslation.com. The closest thing to a stated mechanism is the Brand page promising 'real time info on budgets, language leverage, project status and costs' during delivery, which implies translation-memory leverage discounting but names no rate, unit or percentage. A buyer cannot tell whether this is billed per word, per hour, per project or on retainer, nor what a minimum engagement costs. That is bespoke-only quoting. Measured absence, not instrument failure: the soft-404 control returned a clean HTTP 404, so pages that load are the real pages.

What this dimension measures: Published per-word rates, hourly transcreation rates, or stated pricing models by service tier score 4-5. A stated model without numbers scores 2-3. Bespoke-only quoting scores 1-2. This category bills by mechanisms buyers rarely see up front (per-word, per-hour, minimums per language); a vendor that explains its own billing mechanics scores toward the top of whatever band its disclosure earns. Counts for 15% of the weighted score.

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

Stronger here: Translated scores Excellent on the same dimension.

Technology and workflow integrationAdequate

A workflow is described but no technology is named. The Agency page offers to 'meet your teams, review your technology and language workflows' and build an integration roadmap, and claims version control and revision tracking across multiple brands and regions plus feedback management to prevent duplicate requests. The Brand page describes building 'a centrally managed bank of approved work' and real-time reporting. The Creative AI page references 'secure managed platforms'. Against that: no TMS is named, no CMS or design-stack connector is named, no file-format coverage is published beyond SRT/VTT for transcription, and continuous localization for weekly-shipping content is never addressed. Two older pages that search engines still index as describing their translation memory, glossary management and a bespoke workflow system - /technology-translation-specialists/ and /business-financial-translation/ - both now return HTTP 404, so that detail is no longer published and was not scored from. The platform is proprietary and unnamed, and no export path is stated.

What this dimension measures: How work actually flows: TMS use, connectors into the client's CMS or design stack, stated file-format coverage, and whether continuous localization is supported for content that ships weekly. Named integrations and a described workflow score 4-5. Email-the-files-back workflow scores 2-3 — workable for a campaign, a ceiling for a content operation. Proprietary-platform lock-in with no export path should be named in the verdict. Counts for 15% of the weighted score.

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

Asset ownership and translation memoryWeak

Silence. Translation memory is implied by the Brand page's 'centrally managed bank of approved work' and by 'language leverage' appearing as a reporting line item, but no page states who owns the translation memory, the glossaries, the termbase or the deliverables, and no page states what happens to those assets at exit. No terms-of-service page was located; the privacy policy page loads but its body could not be read and it is in any case the wrong instrument for deliverable IP. Notably the accumulating asset is described in the vendor's own words as something they centrally manage, with no client-ownership or export commitment attached. Scored at the rubric's stated value for silence.

What this dimension measures: The translation memory and glossaries built on the client's money are the client's accumulated asset, and whether they leave with the client at exit is the switching question of this category. Explicit client ownership of TM, glossaries and deliverables scores 4-5. Silence scores 2. A stated policy that the vendor retains the TM scores 1-2 — that is the client's spend converted into the vendor's moat, and it should be named plainly. Counts for 10% of the weighted score.

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

Stronger here: Argos Multilingual scores Strong on the same dimension.

Turnaround and capacityWeak

No turnaround times appear anywhere on the site - not by volume, not by language pair, not by content type - and no rush terms or SLA language is published. Capacity is partially evidenced: the Contact page lists three regional offices with distinct working phone numbers (APAC +65 6990 9293, EMEA +44 (0)20 3882 5407, AMS +1 929 376 1397), and the Agency page claims 'client-facing specialists located globally', so time-zone coverage is at least asserted with checkable contact detail behind it. But scheduling itself is entirely undisclosed, which is what this dimension measures. To the vendor's credit, they also make no implausible same-day-everything claim, so there is nothing here to penalise for dishonesty - only for absence.

What this dimension measures: Stated turnaround by volume and language pair, rush terms, and stated capacity across time zones. Published, realistic turnaround commitments score 4-5. Undisclosed scheduling scores 2. Score honesty over speed: same-day everything at marketing quality across forty languages is a claim to be sceptical of, not to reward. Counts for 10% of the weighted score.

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

Stronger here: Translated scores Strong on the same dimension.

References and review baseWeak

A Clutch profile exists at clutch.co/profile/creative-translation and was read first-hand: it shows 0.0 out of 5 with zero verified reviews, a 50-249 employee band and London headquarters, with no minimum project size or hourly rate disclosed. No G2 presence was located. A ProZ company profile and ProZ Blue Board record exist but both returned HTTP 403 and could not be read, so no figure from either is cited here. A Sortlist listing also returned 403. Glassdoor entries exist but are employee reviews and are not client evidence. The vendor's own case studies name clients but carry no attributed client quotes and no contactable references. Effectively no independent client review base is readable. Weighted at only 5 percent by design, since enterprise localization runs on procurement references rather than public reviews.

What this dimension measures: Independent, verified reviews or named checkable references. A substantial verified base scores 4-5; a handful scores 2-3; none scores 2. Low weight deliberately — enterprise localization runs on procurement references rather than public reviews, so absence is weak evidence, and the practice evidence above matters 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.

Stronger here: Translated scores Strong on the same dimension.

Verdict

Creative Translation is what it says it is: a marketing-first language business rather than a document-translation mill wearing a marketing service page. That distinction is the first thing this category has to establish and this vendor clears it. The site publishes nine client case studies - Canon, Samsung, Madame Tussauds, PayPal, Airbnb, Virgin Atlantic, F&F Clothing, Unilever and Monotype - and the work described in them is recognisably transcreation and campaign production rather than translation with a marketing label.

The PayPal case covers script and tagline adaptation, voice-over casting and recording supervision, subtitling, motion-graphics adaptation and legal copy across 45 markets with more than 300 contributors. The Samsung case covers building a voice-and-tone rulebook for the Bixby assistant, running copywriter training workshops, three rounds of focus-group testing per market across five markets, and producing over 10,000 response variations. The company is real and long-established: Companies House records CREATIVE TRANSLATION LIMITED, number 04889993, as active, incorporated 5 September 2003, registered at 11 Power Road W4 5PY - the same address the site publishes - under SIC 74300, translation and interpretation activities, with accounts filed to 31 July 2025.

That independently corroborates the 'over 20 years' claim on the About page.

The process story is described but not certified. There are named review layers rather than a vague 'native speakers' assurance: a five-phase structure with copy validation and brand guardianship in the Approval phase, pre-launch native-speaker reviews, linguistic editing and in-market research, and in-market teams engaging with writers before translation starts. Machine translation is handled with unusual candour for this category - the Creative AI page states plainly that neural MT is combined with expert linguists, run on managed platforms with a full audit trail, and explicitly warns clients off unmanaged public AI tools.

That is disclosed MT with stated boundaries, which is the honest posture, and it is scored on its own terms rather than penalised. What is absent is anything a buyer could verify: no ISO 17100, no ISO 9001, no ATC or ELIA membership appears anywhere across the homepage, about, services, agency, brand or contact pages. For a 23-year-old language service company that absence is conspicuous, because certification is the standard instrument buyers in this category use to check a quality process they cannot read the output of.

The Community page, which is where a vetting story would live, is a recruitment form with no published qualification criteria. The About page's headline figure of '3,147 of us' is presented as team size, while the Clutch profile records a 50-249 employee band; the reconciliation is almost certainly a freelance network counted alongside staff, but the site does not draw that line itself.

The commercial terms are where this record thins out badly, and it thins out in exactly the three places this category's buyers get hurt. First, pricing: nothing is published - no per-word rate, no transcreation hourly rate, no per-language minimum, no stated billing model. Every path ends at 'Request a quote'.

The Brand page's promise of real-time reporting on 'budgets, language leverage, project status and costs' hints that translation-memory leverage discounting exists, but names no unit and no number. Second, turnaround: no times by volume or language pair, no rush terms, no SLA. The vendor at least avoids the implausible same-day-across-forty-languages claim, so there is nothing dishonest here, only nothing to plan against.

Third, and most consequential, asset ownership: the site never states who owns the translation memory, the glossaries or the deliverables, or what leaves with the client at exit. The Brand page describes building 'a centrally managed bank of approved work' - the client's accumulated linguistic asset, described in the vendor's own words as something the vendor centrally manages, with no ownership or export commitment attached. No terms-of-service page was located to resolve it.

This is silence rather than a stated vendor-retention policy, and it is scored as silence, but it is the single question a buyer should put in writing before signing.

Two further gaps and one limit on this evaluation. Technology is described as workflow rather than as named systems: version control, revision tracking, feedback management, an integration roadmap and 'secure managed platforms' are all asserted, but no TMS, no CMS or design-stack connector and no file-format list are named, and continuous localization for content that ships weekly is not addressed at all. Two older pages that search engines still index as describing their translation memory, glossary management and a bespoke workflow system now return 404, so that detail is no longer published and was deliberately not scored from.

On reviews: the Clutch profile was read first-hand and carries zero verified reviews and a 0.0 rating; a ProZ company profile, a ProZ Blue Board record and a Sortlist listing all exist but returned 403 and could not be read, so nothing from them is cited here in either direction - unreadable is not the same as empty. The site's own soft-404 control returned a clean HTTP 404, so the pages that did load are genuinely the pages requested and the absences above are measured, not artefacts. The shape of this vendor is a credible creative practice with an enterprise-procurement disclosure posture: the work is shown, the commercial mechanics are not.

What you can do next

Koolav can make the introduction and handle the back-and-forth, or you can go straight to the agency.

Visit their website

This agency has not published a paid trial. What a paid trial is.

What we verified

Each claim below was checked against a named source, last on 2026-08-26. Follow any of them and check for yourself — that is the point of publishing them.

What other platforms say

Clutch {'score': 0.0, 'count': 0}

No independent client review base could be read. The Clutch profile was read first-hand and shows zero verified reviews and a 0.0 rating, with a 50-249 employee band and London headquarters. No G2 presence was located. A ProZ company profile, a ProZ Blue Board record and a Sortlist listing all exist but returned HTTP 403 and could not be read first-hand, so no rating or count from any of them is cited. Glassdoor entries exist but are employee reviews, which are not client evidence. The vendor's own case studies name clients but include no attributed client quotes or contactable references.

These are other platforms' numbers, not ours. We report them because they are part of the picture, and we do not average them into our score — our score comes from the published rubric above.

What we could not verify

Sources

Others we evaluated in Localization & Transcreation

Same rubric, same evaluator, same date range — so these are directly comparable to the verdict above.

See all 17 Localization & Transcreation agencies we evaluated →

Think this is wrong?

If anything on this page is wrong or out of date, send us the correction and we will re-read the site.

Evidence moves a verdict. Money never does — no agency pays us for a listing, a placement, or a re-review. Whatever we conclude we publish, including that the verdict stands.