Four terms get used interchangeably in briefs, quotes, and vendor conversations: translation, content localization, transcreation, and dubbing. They describe different amounts of work, they cost different amounts of money, and they produce different outcomes.
The confusion is expensive in both directions. Brief a translation job when you needed localization and you get output that is accurate and unusable. Brief transcreation when translation would have done and you have spent a multiple of what the asset was worth.
This piece defines each precisely, shows where the boundaries actually sit, and gives you a framework for deciding which one a given piece of content needs.
Quick Answer
| What it does | Source fidelity | Typical use | Relative cost |
Translation | Transfers meaning between languages | Highest — stays close to source | Documentation, subtitles, legal, technical | Baseline |
Content localization | Adapts the full experience for a target market | High — meaning preserved, expression adapted | Video, OTT, apps, web, e-learning | 2–5× baseline |
Transcreation | Recreates intent, often departing from source wording | Lowest — intent preserved, content rebuilt | Advertising, taglines, comedy, children’s content | 5–15× baseline |
Dubbing | Replaces spoken audio with target-language performance | Varies — a delivery method, not a category | Any content with dialogue | Depends on tier |
The critical structural point: the first three sit on a spectrum of how far you are allowed to move from the source. Dubbing is not on that spectrum at all, it is a delivery format that can be executed at any of the three depths.
Translation: Meaning Transfer
Definition: rendering source-language meaning in the target language accurately, with terminology consistency and no additions or omissions.
Translation optimises for fidelity. A good translator does not improve the source, localise its references, or adjust its structure. Those would be defects.
Appropriate for: technical documentation, legal and contractual text, medical and scientific material, product specifications, and subtitle tracks where the on-screen dialogue is doing the cultural work and the subtitle only needs to convey it.
Example — Hindi source: “Woh toh gaya kaam se.”
Translation: “He’s gone from the work.”
Meaning transferred. Comprehensible. Also completely wrong in register - the idiom means roughly “he’s finished,” “he’s done for.” Translation alone gives you the words and loses the sentence.
Where translation is genuinely correct: when the reader needs to know precisely what the source said, and any adaptation would introduce risk. A dosage instruction, a contractual clause, a technical specification. In these cases adaptation is not a value-add. It is a liability.
Content Localization: Experience Adaptation
Definition: adapting content so it functions natively for the target audience - including language, cultural references, format, regulatory requirements, and, for media, timing and performance.
Content localization preserves what the source means while changing how it is expressed so that a native audience receives it the way the source audience received the original.
The same line, localized for English-language dubbing:
• Translation: “He’s gone from the work.”
• Localization: “He’s finished.” / “He’s done for.”
Now the register survives. And if this is a dub rather than a subtitle, the localized line also has to land inside the same duration as the original and avoid putting an open vowel where the actor’s lips close.
For video content, localization spans six layers: linguistic accuracy, cultural adaptation, technical compliance, regulatory clearance, performance and casting, and discovery metadata. We break each of these down in our complete guide to content localization.
Appropriate for: streaming and OTT content, YouTube libraries, micro drama, film, e-learning, corporate video, apps, and marketing sites. In practice, almost anything with an audience rather than a reader.
The distinction that matters most in practice
Inside a dubbing pipeline, translation and localization are two separate passes on the same script, and skipping the second is the single most common cause of localization failure.
Translation pass establishes what the line means.
Adaptation pass makes it performable: fits the timing envelope, matches the character’s register, sounds like speech rather than text, and works with the visible articulation.
A script that has been translated but not adapted produces recognisable symptoms in the booth: actors compressing lines to fit, retakes climbing, and delivery that sounds read rather than spoken. It is diagnosable from the session logs, and by then it is expensive. Our end-to-end dubbing workflow treats these as separate mandatory stages for exactly this reason.
Transcreation: Intent Reconstruction
Definition: recreating the effect of the source in the target language and culture, with explicit licence to depart from source wording, structure, and sometimes content.
Transcreation starts from a brief describing what the content should achieve, the emotional response, the brand association, the action it should prompt, rather than from a source text to be preserved. The source becomes reference material, not a constraint.
Where it is necessary:
Wordplay and puns. A pun cannot be translated. It can only be replaced with a different pun that does comparable work in the target language. This is writing, not translation.
Taglines and campaign lines. Advertising lines are engineered for rhythm, memorability, and connotation. All three are language-specific properties. A literally translated tagline is almost always a worse tagline.
Comedy. Comic timing, setup-punchline structure, and the specific things a culture finds funny differ enough that jokes frequently have to be rebuilt from the intent up. A joke that gets a translation and not a rebuild does not get a smaller laugh, it gets silence.
Children’s animation. Names, songs, rhymes, and culturally-specific concepts are routinely rebuilt for young audiences who have no reference framework for the source culture and no tolerance for confusion.
Micro drama hooks. The first three to five seconds of a micro drama episode carry the retention burden for the entire episode. The hook that works in Mandarin frequently is not the hook that works in Hindi, the emotional trigger is different, and the line has to be rebuilt around the trigger that works in-market rather than the one that worked at source.
The cost of getting this wrong in either direction: transcreating documentation wastes money on creative work nobody needed. Merely translating a campaign line produces something that is technically correct and does not sell.
Dubbing: A Delivery Method, Not a Depth
This is the category error that causes the most confusion in vendor conversations.
Translation, localization, and transcreation describe how far from the source you are permitted to move. Dubbing describes what format the output takes — replacing source dialogue with target-language performance recorded to picture.
Which means dubbing can be executed at any of the three depths:
• Dubbing at translation depth — literal script, poor timing fit, high retake count. Common in low-cost work and consistently identifiable as such.
• Dubbing at localization depth — adapted script, timed and phonetically shaped, cast for character. The professional standard for scripted content.
• Dubbing at transcreation depth — script substantially rebuilt for the target market. Standard for children’s animation, comedy, and advertising.
“We need dubbing” therefore does not specify the job. Two vendors can both quote for “Hindi dubbing” and be quoting for work that differs by a factor of three in cost and considerably more in outcome. This is why vendor selection frameworks have to interrogate depth explicitly rather than accepting the format label.
The related formats voice-over, ADR, and lip-sync, are also delivery methods rather than depths, and we compare them in lip-sync vs voice-over vs ADR.
Three Places the Confusion Costs Real Money
Scenario one: the OTT catalogue drop. A platform commissions “translation” for a 40-title catalogue push into a new language market. Delivery is accurate and on time. Completion rates in the new market run well below the source-language baseline, and audio-track selection is low. Diagnosis: the scripts were never adapted, the timing never fit, and viewers switched back to subtitles within minutes. The rework costs more than the original job because the recording has to be redone, not just the script.
Scenario two: the over-specified corporate rollout. A company briefs transcreation for 200 hours of compliance and onboarding training. The material is procedural, process steps, policy definitions, system walkthroughs and has almost no culturally-bound content. Localization at standard depth would have delivered an equivalent learner outcome. The transcreation premium bought nothing measurable. (What e-learning localization actually requires.)
Scenario three: the split-depth campaign. A brand localizes its product video correctly and translates the campaign tagline literally. The video performs; the tagline is meaningless in-market and, in one language, faintly comic. Different assets in the same campaign needed different depths, and the brief applied one rule to all of them.
The Decision Framework
Five questions. Answer them per asset, not per project.
1. What is the failure mode if the audience misunderstands?
• Safety, legal, financial, or medical consequence → translation, with review. Do not adapt.
• Confusion or disengagement → localization.
• Indifference — they simply do not respond → transcreation.
2. How culturally loaded is the content?
Count the elements that require target-culture knowledge: idioms, humour, references to food, festivals, institutions, geography, historical events, celebrities.
• Near zero → translation is likely sufficient.
• Moderate → localization.
• The cultural content is the content → transcreation.
3. Is the output performed, or read?
Performed content, anything with voice, timing, or on-screen articulation, cannot be handled at translation depth regardless of how simple the language is. The physical constraints of duration and articulation force adaptation. This is a structural requirement, not a quality upgrade.
4. What is the asset’s revenue ceiling in this market?
Localization depth should track the revenue the asset can generate in the specific target market, not its importance in the home market. A flagship domestic title may justify only subtitles in a small territory. A mid-catalogue title may justify full lip-sync dubbing in a large one.
5. Is this a first entry or an established presence?
First entry into a market rewards higher depth. The initial impression sets audience expectations for everything you release afterwards, and a poorly localized launch title suppresses uptake of the catalogue behind it. Established presence with proven demand can operate at lower depth on long-tail catalogue without the same risk.
Matching Depth to Content Type
Content type | Recommended depth | Reason |
Technical documentation | Translation | Precision over resonance |
Legal and compliance | Translation, reviewed | Adaptation introduces liability |
Corporate training | Localization | Procedural, low cultural load |
Product and explainer video | Localization | Performed; timing constraints apply |
Scripted drama and film | Localization | Performance and register are the product |
Documentary and factual | Localization (voice-over tier) | Sync fidelity not the quality signal |
Micro drama | Localization, hooks transcreated | Retention depends on in-market emotional triggers |
Children’s animation | Transcreation | No source-culture reference framework in the audience |
Comedy | Transcreation | Jokes do not survive translation |
Advertising and taglines | Transcreation | Effect is the deliverable, not wording |
Song lyrics | Transcreation | Meaning, metre, and rhyme cannot all survive |
How to Brief This Correctly
Most localization briefs specify a language pair, a volume, and a deadline. That is not enough information to quote against, and it is how two vendors end up quoting the same job at very different depths.
A brief that produces comparable quotes and predictable output specifies:
• Depth per asset class — using the vocabulary above, explicitly
• Adaptation licence — what the vendor may change without approval, and what requires sign-off. Names, brands, cultural references, humour, and structure should each have an explicit rule.
• Timing constraints — whether output must fit source duration, and the tolerance
• Register decisions — formality rules per character pair or per audience segment, decided up front rather than left to the translator
• Glossary — locked terminology, names, and brand handling
• Delivery specifications — technical format requirements per platform
• Review structure — who approves adaptation decisions, and how quickly
Depth ambiguity is the most common defect in localization briefs, and the most expensive one, because it is only discovered after delivery.
The Short Version
Translation when precision matters more than resonance.
Localization when the audience needs to receive it the way the source audience did, which is nearly always, for anything performed.
Transcreation when the effect is the deliverable and the wording is just the vehicle.
Dubbing is how it is delivered, not how deep it goes, and it needs a depth specified alongside it.
Most media projects are localization projects with transcreation applied selectively to the highest-leverage moments: the hook, the tagline, the punchline, the title. Getting the base depth right and spending the premium precisely, rather than uniformly, is where localization budgets produce their best returns.
Scoping a project and not sure which depth applies? Send a sample asset, your target languages, and your delivery specs. We will come back with a depth recommendation per asset class and the cost implication of each. Talk to Sukudo Studios.