Best Practices for Video Translation and Brand Globalisation
Best practices for video translation and brand globalisation cover audience research, subtitling versus dubbing choices, ISO 17100 quality control, cultural adaptation, and measurable ROI across
Request a Free Quote
What we do
Our services
What does video translation for brand globalisation actually involve?
Video translation for brand globalisation is the governed adaptation of spoken dialogue, on-screen text, and cultural references into target languages using subtitling, dubbing, or voice-over, delivered under ISO 17100 controls so a brand keeps one voice across every market it enters.
How do you choose between subtitling, dubbing, and voice-over?
Choose subtitling for training, corporate, and social content where budget and speed matter; dubbing for entertainment, children’s content, and premium marketing where immersion matters; voice-over for e-learning, documentary, and explainer video where authority matters — each mapped against cost, turnaround, and brand-voice fidelity.
What best practices govern the video translation workflow end-to-end?
A governed workflow runs 7 stages — scoping, transcription, translation, revision, review, recording or subtitling, and QA sign-off — with a locked picture at the start and an ISO 17100 four-eyes check before delivery, so nothing ships without a second linguist’s signature.
How do you keep brand voice consistent across languages?
Brand voice stays consistent when three artefacts travel with every project: a multilingual termbase locking product and brand names, tone-of-voice tokens per locale (formality, humour, register), and a locked visual and audio style guide covering typography, colour, music, and voice-actor casting.
How do you adapt video for cultural context without breaking the brand?
Cultural adaptation covers 7 layers — humour, idioms, gestures, colour semantics, on-screen text, RTL layout, and legal disclaimers — and each layer is checked by an in-country reviewer before the master is re-conformed.
What technical standards make video translation broadcast-ready?
Broadcast-ready video meets 5 technical standards: subtitle reading rate at or below 17 CPS for English, timecode-accurate SRT or TTML, EBU-TT-D for European broadcast, -23 LUFS audio loudness for dubbed tracks, and Ofcom subtitle guidelines for UK on-demand.
Video accounts for the majority of global internet traffic and is projected to reach 82% by 2025, which turns multilingual video content into the fastest lever for brand globalisation. Under ISO 17100 quality control, professional video translation moves a UK brand into new markets without diluting its identity — provided the workflow is governed, not improvised.
What does video translation for brand globalisation actually involve?
Video translation for brand globalisation is the governed adaptation of spoken dialogue, on-screen text, and cultural references into target languages using subtitling, dubbing, or voice-over, delivered under ISO 17100 controls so a brand keeps one voice across every market it enters. It sits at the intersection of linguistic services, audio post-production, and brand governance.
The scope covers 6 asset layers a single video carries:
- Spoken dialogue and narration.
- Music and sound design cues that shift meaning by locale.
- On-screen text, lower-thirds, and title cards.
- Baked-in graphics, charts, and UI screenshots.
- Cultural references — humour, idioms, gestures, colour semantics.
- Legal disclaimers, credits, and end-cards.
How is video translation different from video localisation?
Video translation converts language; video localisation adapts the entire experience — currency, units, cultural references, on-screen graphics, and legal disclaimers — so localisation is the superset that brand globalisation actually requires. A dubbed line that says “£100” in a source video becomes “€120” in the German cut only under localisation, not translation.
Why does brand globalisation depend on video first?
Video carries 6 signal layers simultaneously — script, voice, music, on-screen text, imagery, and pacing — so translating it correctly is the fastest way to move a brand’s tone into a new market without diluting identity. A written blog can be localised; a hero brand film has to be re-performed. Brand globalisation is the programme of adapting brand identity, tone, and creative assets for multiple international markets while preserving core equity, and video is the densest asset that programme touches.
How do you choose between subtitling, dubbing, and voice-over?
Choose subtitling for training, corporate, and social content where budget and speed matter; dubbing for entertainment, children’s content, and premium marketing where immersion matters; voice-over for e-learning, documentary, and explainer video where authority matters — each mapped against cost, turnaround, and brand-voice fidelity.
| Mode | Cost per finished minute (GBP) | Turnaround | Cognitive load on viewer | Brand-voice fidelity | Regulatory fit | Best content type |
|---|---|---|---|---|---|---|
| Subtitling | £8 – £15 | Same-day to 48 h | Medium (reading required) | High (original performance preserved) | Ofcom, WCAG 2.2 AA 1.2.2 | Social, training, corporate, documentary |
| Voice-over | £25 – £60 | 2 – 5 days | Low | Medium-High | WCAG 2.2 AA (if captioned) | E-learning, explainer, documentary, news |
| Dubbing (full lip-sync) | £120 – £350 | 2 – 6 weeks | Very low | High (with cast approval) | Ofcom broadcast, EBU R128 | Feature entertainment, kids’ content, hero campaigns |
When does subtitling win?
Subtitling wins for social-first content, technical training, and low-budget rollouts because it preserves the original performance, costs roughly £8 to £15 per finished minute, and can clear same-day for clips under 1,000 source words placed before 11:00 GMT. It requires fewer production steps than dubbing — no voice actors, no studio recording, no audio post — which is why it is the least expensive delivery mode.
A brief technical note on terminology: subtitles display only dialogue and speech for viewers who can hear the audio, while closed captions add non-speech audio information — sound effects, speaker identification, music cues — for accessibility. UK regulated delivery leans on closed captions; global marketing ships subtitles.
When does dubbing win?
Dubbing wins for feature-length entertainment, kids’ content, and hero brand campaigns because full lip-synced replacement removes the reading load and lets the audience stay inside the visual world. Lip-sync dubbing specifically adapts the translation to align with syllables, pace, pauses, and visible lip movements, which is why casting, studio time, and sync editing push the finished-minute rate to £120–£350.
When does voice-over win?
Voice-over wins for e-learning, corporate explainers, and documentary because a single narrator carries authority, costs less than full dub, and tolerates minor timing drift. UN-style voice-over layers the target-language track over the original at reduced level, which suits factual and interview content where the source voice remains audible for authenticity.
Voice-over and dubbing production draw on the same casting and studio infrastructure covered by our Audio Translation Services.
What does a subtitling-vs-dubbing-vs-voice-over decision table look like?
A decision table compares the three modes across 6 axes: cost per finished minute in GBP, turnaround, brand-voice fidelity, cognitive load, regulatory fit (Ofcom / WCAG), and suitability by content type. The table above is the canonical version; use it as the first artefact in any RFP response so procurement, marketing, and legal all read the same rules.
How it works
What best practices govern the video translation workflow end-to-end?
1
How do you scope a video translation project?
Scope a project by locking picture, counting source words in the timed script, confirming target locales, and agreeing deliverable formats (SRT, TTML, mixed audio stems) before any linguist touches the file. A moving edit is the single biggest driver of cost overrun — every re-cut invalidates timing and forces re-segmentation across every target language. Deliverable format selection matters as much as linguist assignment: SRT is the universal minimum, WebVTT is required for HTML5 video players, EBU-TT-D is mandatory for broadcast distribution, and IMSC 1.1 is the standard for premium streaming platforms. Agreeing the full format list at scoping prevents expensive re-export cycles downstream.
Budget benchmarks are a core scoping input, not an afterthought. Standard business subtitling runs £8 to £15 per finished minute, narrator-style voice-over runs £25 to £60 per finished minute, and full lip-synced dubbing runs £120 to £350 per finished minute in the UK — knowing these figures before briefing allows production to choose the right modality per locale rather than defaulting to the most expensive option everywhere. For markets where dubbing culture is strong and audiences are conditioned to reject subtitled content, the higher investment in full dubbing is not optional if the brand expects meaningful engagement.
2
How does ISO 17100 quality control apply to video?
ISO 17100 applies through a mandatory four-stage chain — translation by a qualified linguist, revision by a second linguist, in-context review against the picture, and final sign-off — every step logged against the project record. The standard mandates qualified linguists, a four-stage revision chain, and a complete audit trail per project, making it the quality backbone of any professional video translation programme. On a video project, ISO 17100 compliance requires that translator qualifications are documented, revision is performed by a different linguist than the translator, and every decision — including deviations from TM suggestions — is recorded in the project log.
The named artefacts that anchor each stage on a video project are:
- Locked-picture SRT — the timecoded subtitle file cut to the frozen master, which becomes invalid the moment the edit changes.
- Pronunciation guide — proper nouns, brand names, and product terms with IPA notation and audio reference samples, essential for maintaining consistent brand sound across dubbing sessions in multiple languages.
- ADR script — the recording script with time-in, time-out, and character direction for dub sessions, structured so that voice artists maintain lip-sync discipline without repeated takes.
- Termbase and TM package — the brand’s approved terminology and previously translated segments, loaded into the CAT environment before translation begins to enforce consistency and reduce per-word cost on repeat projects.
3
The files a client hands over for video translation are the master video, timed script, style guide, and glossary.
Clients need to hand over the locked master (ProRes or H.264), separated audio stems, the timed source script, brand glossary, pronunciation guide, and any on-screen text as an editable After Effects or Premiere project. Separated audio stems — music, effects, and dialogue on independent tracks — are non-negotiable for dubbing, because the dialogue stem must be removed cleanly before the target-language recording is laid in. Supplying a combined stereo mix forces the studio to perform audio surgery that adds cost and risks degrading music and effects quality. On-screen text supplied as flattened video cannot be relayered in the target language without a complete re-composite; editable project files eliminate that bottleneck entirely.
Accessibility requirements add further delivery obligations. WCAG 2.2 Success Criterion 1.2.2 requires synchronised captions for all prerecorded audio content, and Success Criterion 1.2.5 requires audio description tracks — both must be delivered as discrete, standards-compliant files rather than baked into the video. Subtitle files must also conform to Ofcom’s guidelines where UK broadcast distribution is in scope: a maximum of 180 words per minute, no more than two lines per subtitle event, 32 to 37 characters per line, and colour-coded speaker identification. Where the timed source script needs certification or archival handling, it moves through the same governed chain as our Document Translation Services in London and the UK.
How do you keep brand voice consistent across languages?
Brand voice stays consistent when three artefacts travel with every project: a multilingual termbase locking product and brand names, tone-of-voice tokens per locale (formality, humour, register), and a locked visual and audio style guide covering typography, colour, music, and voice-actor casting. These three artefacts are the brand-globalisation governance model that separates a coordinated programme from a scatter of one-off jobs.
What belongs in a multilingual termbase?
A multilingual termbase holds product names, feature names, legal disclaimers, do-not-translate lists, and forbidden terms per locale, versioned in a CAT tool so every linguist pulls from the same source. Every entry carries a definition, a context sentence, a part-of-speech tag, and an approval status.
How do you define tone-of-voice tokens per locale?
Define tokens across 5 dimensions per locale: formality (tu/vous, tu/usted), humour tolerance, sentence length, brand persona adjectives, and register — then attach concrete before/after examples to each token. A German B2B token might read “formal Sie, no contractions, sentence length ≤ 22 words, adjectives: precise, engineered, reliable”; a Mexican Spanish social token might read “informal tú, contractions allowed, humour permitted, adjectives: warm, direct, confident.”
How do you cast voice actors for brand fidelity?
Cast voice actors by shortlisting 3 to 5 native speakers per role, matching age, gender, timbre, and regional accent to the source performance, then approving through a brand guardian before recording. Signed consent, GDPR lawful basis, and buy-out terms are locked before the actor enters the booth.
How do you adapt video for cultural context without breaking the brand?
Cultural adaptation covers 7 layers — humour, idioms, gestures, colour semantics, on-screen text, RTL layout, and legal disclaimers — and each layer is checked by an in-country reviewer before the master is re-conformed. This is transcreation: modifying idioms, humour, imagery, and on-screen text to resonate with the target culture rather than translating word-for-word.
Which cultural elements most often break in video translation?
The elements that most often break are visual gestures with local meaning, colour-coded UI shown on screen, humour built on wordplay, food and gender references, and hand-drawn signage baked into the picture. A thumbs-up, a red error state, a pork-based food shot, or a joke that relies on a homophone — each is a re-shoot risk unless flagged at scoping.
How do you handle right-to-left languages like Arabic and Hebrew?
Handle RTL languages by mirroring on-screen graphics, reversing subtitle alignment, allowing wider reading time at 13 CPS, and confirming that any embedded UI screenshots have been re-rendered in the target locale. Arabic subtitle work also requires diacritic decisions (fully vocalised for children’s content, unvocalised for adult broadcast) — covered inside our Arabic Translation Services in London and the UK.
How do you re-shoot or re-render on-screen text for video translation?
Re-shoot or re-render on-screen text by requesting the editable project file, swapping text layers per locale, re-timing animations to the new string length, and re-exporting a locale-specific master rather than baking subtitles over the original. Baked subtitles over baked English titles create a stacked-text collision that a mirrored RTL layout cannot rescue.
What technical standards make video translation broadcast-ready?
Broadcast-ready video meets 5 technical standards: subtitle reading rate at or below 17 CPS for English, timecode-accurate SRT or TTML, EBU-TT-D for European broadcast, -23 LUFS audio loudness for dubbed tracks, and Ofcom subtitle guidelines for UK on-demand.
What reading rates apply per language for subtitles?
Reading rates cluster at 17 CPS for English, 15 CPS for German, 13 CPS for Arabic, 12 CPS for Chinese, and 21 CPS for vertically-set Japanese — each cap driving subtitle segmentation and minimum on-screen duration.
| Language / script | Reading rate (CPS) | Minimum on-screen | Segmentation implication |
|---|---|---|---|
| English | 17 | 1.0 s | 2 lines × 37 chars caps most dialogue |
| German | 15 | 1.2 s | Compound nouns force line breaks earlier |
| French | 16 | 1.0 s | Expansion 15–20% vs English |
| Spanish (LatAm) | 17 | 1.0 s | Expansion 20–25% vs English |
| Arabic (RTL) | 13 | 1.5 s | Wider timing; alignment reversed |
| Chinese (Simplified) | 12 | 1.5 s | Fewer characters, higher density |
| Japanese (vertical) | 21 | 1.0 s | Vertical-set caption strip on right |
Which subtitle file formats should you use?
Use SRT for social and web, TTML or EBU-TT-D for UK broadcast and BBC iPlayer, WebVTT for HTML5 players, and IMSC 1.1 for premium OTT distribution. Timed-text standards including TTML, WebVTT, SCC, and SRT cover broadcast and web delivery across every major UK platform.
What audio loudness targets apply to dubbed tracks?
Dubbed tracks target -23 LUFS integrated loudness for EBU R128 broadcast, -16 LUFS for streaming, and -14 LUFS for YouTube — with true peak held below -1 dBTP in all cases. Loudness compliance is measured across the entire programme, not the loudest segment.
How do UK regulations shape video translation and accessibility?
UK regulations shape delivery through 4 frameworks: Ofcom subtitle guidelines for on-demand services, WCAG 2.2 AA for public sector video, GDPR for handling voice-actor and speaker personal data, and Home Office evidentiary rules for translated video exhibits. Translated video used as legal evidence — witness interviews, CCTV, body-worn footage — falls under our Legal Translation Services.
What do Ofcom subtitle guidelines require?
Ofcom guidelines require 180 words per minute maximum, 2-line subtitles with 32 to 37 characters each, colour-coded speaker identification, and hard-of-hearing sound effect descriptions on qualifying UK on-demand services. Speaker colours follow a fixed hierarchy — white, yellow, cyan, green, magenta — assigned by prominence, not by character order.
How does WCAG 2.2 AA apply to captions?
WCAG 2.2 AA requires synchronised captions for all prerecorded audio (1.2.2) and audio description for prerecorded video (1.2.5), which extends to translated versions used on any UK public-sector or regulated site. See the W3C guidance on captions (prerecorded) for the normative text.
How does GDPR affect voice actor and speaker data?
GDPR affects the project because a person’s voice is biometric data — consent, lawful basis, retention limits, and processor agreements have to sit on file before recordings leave the studio. Voice cloning and AI dubbing raise the bar: explicit consent for the specific downstream use is required, not a generic release.
How do you measure the ROI of video translation and brand globalisation?
Measure ROI across 5 KPIs: watch-through rate per locale, CTA conversion lift versus source, share of voice in target markets, cost per qualified lead in GBP, and brand consistency score from a quarterly linguistic audit. Completion rate, watch time, and click-through rate are the industry-standard engagement metrics feeding these KPIs.
Which engagement metrics matter per locale?
The engagement metrics that matter per locale are watch-through rate, average view duration, subtitle-on percentage, and re-watch rate — benchmarked against the source-language baseline before any campaign spend is unlocked. A locale that trails the source by more than 15 points on watch-through triggers a linguistic and creative review.
How do you run a brand consistency audit across languages?
Run a brand consistency audit by pulling 10 to 20 sample assets per locale each quarter, scoring them against the tone-of-voice tokens and termbase, and issuing corrective actions to any vendor scoring below 90%. The scorecard covers 6 dimensions: terminology adherence, tone match, register, cultural fit, technical compliance, and legal accuracy.
When should you use AI translation, and when should a human linguist lead?
AI translation leads on speed and scale for internal, low-risk video such as knowledge-base clips and social micro-content; human linguists lead on brand, legal, medical, and marketing video where cultural nuance, liability, and tone-of-voice fidelity make a mistranslation expensive. Machine translation alone is insufficient for marketing, legal, or culturally nuanced content without human post-editing, because tone, nuance, and cultural context require human judgement.
Where does AI dubbing and voice cloning fit?
AI dubbing fits internal training, rapid social iteration, and A/B tests where cost and turnaround dominate — but hero marketing, regulated healthcare, and legal video still require a cast human voice under GDPR-compliant consent. Lip-sync remains the boundary: AI dubbing excludes true lip-sync or charges a premium, which pushes premium content back to human dub.
What does a human-in-the-loop workflow look like?
A human-in-the-loop workflow drafts a machine translation, routes it to a qualified linguist for post-editing under ISO 18587, then feeds the ISO 17100 four-eyes revision and review cycle before recording or subtitling. The linguist edits for accuracy, terminology, tone, and cultural fit — not just grammar — so the machine-drafted output is materially rewritten, not lightly polished.
Pricing
How much does professional video translation cost in the UK?
Professional video translation in the UK ranges from £8 to £15 per finished minute for subtitling, £25 to £60 per finished minute for voice-over, and £120 to £350 per finished minute for full lip-synced dubbing, with document-adjacent transcripts from £30 per page under ISO 17100.
| Deliverable | Rate (GBP) | Unit | Included |
|---|---|---|---|
| Subtitling (single locale) | £8 – £15 | Finished minute | Transcription, translation, timing, SRT export, QA |
| Closed captions (SDH) | £10 – £18 | Finished minute | Speaker ID, sound effects, WCAG 2.2 AA |
| UN-style voice-over | £25 – £45 | Finished minute | Single narrator, studio, mix at -16 LUFS |
| Character voice-over | £40 – £60 | Finished minute | Cast narrator, direction, revisions |
| Dubbing (lip-sync) | £120 – £350 | Finished minute | Cast, ADR script, studio, mix, EBU R128 |
| Transcript / timed script | from £30 | Page | ISO 17100 revision chain |
What drives video translation cost up or down?
Cost is driven by 6 variables: source duration, target language count, subject-matter complexity, voice-talent tier, turnaround (same-day versus milestone), and whether on-screen text has to be re-rendered from editable project files. Regulated content (medical, legal, financial) adds a specialist reviewer and lifts the finished-minute rate by 15–30%.
What turnaround should you expect for video translation?
Expect same-day turnaround for civil clips under 1,000 source words placed London before 11:00 GMT, 24 to 48 hours for legal and academic packs, and agreed milestones for projects above 10,000 words — all under ISO 17100 quality control. Feature-length localisation into 6+ locales runs on published milestones with sign-off gates at transcription, translation, revision, and mix.
How do you choose a professional video translation partner?
Choose a partner by verifying 6 things: ISO 17100 certification, in-country native linguists, a documented four-eyes revision chain, UK studio capacity for voice recording, transparent per-minute pricing in GBP, and case studies in your regulated sector. A partner that cannot show the revision log for a past project cannot demonstrate ISO 17100 compliance.
What questions should you ask on a video translation RFP?
Ask 7 questions on the RFP:
- Which ISO standards do you hold, and can you show the current certificate?
- How are your linguists qualified, and how many years’ subject-matter experience per locale?
- How are termbases and translation memories managed, versioned, and returned to the client?
- How is confidential footage stored, transmitted, and destroyed?
- How are voice actors cast, paid, and released under GDPR?
- What does your revision workflow look like end-to-end, with named roles?
- How do you measure post-delivery quality, and what remediation applies if scores drop below 90%?
The most-asked questions cover the difference between subtitling and dubbing, the role of ISO 17100, the cost per minute in GBP, turnaround for London orders, and how AI translation fits alongside human linguists.
Frequently asked questions
Sources
The standards, regulations, and technical specifications referenced across this guide underpin every professional video translation and brand globalisation engagement we undertake. Each one carries binding or best-practice authority in its respective domain, and our workflows are designed to satisfy all of them simultaneously rather than treating compliance as a checklist applied at the end.
ISO 17100 is the translation-services quality standard that mandates qualified linguists, a four-stage revision chain, and a complete audit trail for every project — it is the quality framework against which our linguist vetting, revision workflow, and project logging are structured. WCAG 2.2 is the accessibility standard that requires synchronised captions under Success Criterion 1.2.2 and audio description under Success Criterion 1.2.5 for all prerecorded video content published on the web. Ofcom’s subtitle guidelines set the broadcast-grade ceiling for subtitle speed (180 words per minute maximum), line count (two lines), line length (32 to 37 characters), and speaker identification (colour-coded) — standards we apply to online content by default because they represent the safest reading-rate discipline for the widest audience. EBU R128 is the loudness standard governing dubbed and mixed audio delivered to broadcast platforms, specifying −23 LUFS integrated loudness and a true peak ceiling of −1 dBTP.
GDPR is the data-protection regulation that classifies human voice recordings as biometric personal data, requiring a lawful basis, explicit consent where applicable, and signed data-processor agreements before any voice talent recording is stored, transferred, or processed — an obligation that applies equally to source-language recordings shared with our studio partners and to target-language recordings created during dubbing sessions.
All technical specifications, reading-rate tables, file-format definitions, and cost benchmarks cited in this guide reflect the current professional consensus across the UK localisation and broadcast industries and are applied directly in our production and quality-assurance processes.