Central Translation or Local Production: Choosing Per Content Type

The two ends of the spectrum

At one end, a central team produces content once, usually in one language, and every market receives a translated or adapted version. At the other, each market plans and produces its own content, in its own language, for its own audience, with little reference to what other markets do.

Few organizations sit at either extreme for long. A company that translates everything centrally finds its campaigns falling flat in some markets; a company where every market creates its own finds itself paying for the same product video five times, with five different descriptions of the same feature. The useful question is not which model is right, but which content types belong closer to which end.

Central production with translation

In this model, content is created once by a central team and localized for each market through translation, subtitles or dubbing, with varying degrees of adaptation.

Where it tends to work well:

• Product explanations, documentation, tutorials and training, where accuracy and consistency matter more than local flavor.
• Content that must say exactly the same thing everywhere, such as safety information, policy updates or pricing.
• Organizations with few or no staff in their markets.
• Large libraries, where producing everything locally would be unaffordable.

Where it tends to struggle:

• Campaigns built on humor, local events, celebrities or cultural references.
• Content where the presenter, setting or examples feel foreign to the audience.
• Markets where buyers expect to see local customers, local prices and local proof.

How far translated content can be adapted without re-production is a spectrum of its own, from plain translation to full localization; translation versus localization explains the difference.

Local production in each market

In this model, each market creates its own content: local presenters, local customers, local scripts and local channels.

Where it tends to work well:

• Brand and campaign content where cultural fit drives results.
• Social media, where trends, formats and tone change quickly and differ by market.
• Customer stories and testimonials, which carry more weight when the customer is local.
• Markets large enough to justify their own production budget and team.

Where it tends to struggle:

• Consistency: product claims, terms and visual identity drift between markets.
• Cost: the same subject is produced repeatedly.
• Quality variation, since some markets have strong teams and others do not.
• Visibility: the central team may not know what has been published or whether it is accurate.

Hybrid models that most organizations use

Several hybrids sit between the extremes. They are not mutually exclusive.

1. Global core, local edge. The central team produces the core content (product videos, documentation, brand assets) and markets add local material around it: a local introduction, local customer examples, local social posts.
2. Central template, local execution. The central team supplies a brief, script structure, brand guidelines and assets; each market produces its own version with local presenters and examples.
3. Transcreation. Central content is rewritten for each market rather than translated, keeping the intent and message but changing jokes, references and examples.
4. Local first, then shared. A market produces something that works, and the central team translates it for other markets. This turns local experiments into global assets.
5. Tiered by market. Large markets get local production; smaller markets receive translated central content.

The second and fifth hybrids are especially common in organizations that have a few large markets and a long tail of smaller ones.

Decision factors

Weigh these factors per content type, not for the organization as a whole.

Cost

Translation spreads one production across markets; local production multiplies it

Control

Central production makes claims, terms and visuals consistent; local production needs guidelines and review to stay aligned

Relevance

Local production fits the market by default; translated content needs adaptation to feel native

Speed

Translation of existing content is fast; local teams react faster to local events and trends

Risk

Regulated or safety-critical messages are easier to control centrally; requirements vary by country and should be checked locally

Scale

Many languages favor central production; a few large markets can support local teams

A useful test for each content type: if a viewer in the market noticed the content was made elsewhere, would it matter? For a software tutorial, usually not. For a brand campaign or a customer testimonial, often yes.

Deciding by content type

Applying the factors to typical content gives a starting point, which each organization should adjust:

• Product tutorials, help videos, training: central production, translated or dubbed.
• Product launch announcements: central core message, with local introductions or local spokespeople.
• Brand campaigns: central template, local execution or transcreation.
• Social media: mostly local, with central guidelines and shared assets.
• Customer stories: local where possible; translated central stories with subtitles as a fallback.
• Legal, safety and policy content: central, with qualified translation and local legal review.

Regional differences inside one language complicate this further: Spanish for Mexico and Spain, or Portuguese for Brazil and Portugal, may need separate treatment even when the content is central. Translating video for regional audiences covers those choices.

Trade-offs and risks in each direction

• Centralizing too much: content feels foreign, local teams disengage and start making unofficial versions.
• Localizing too much: costs climb, claims and terms drift, and nobody can see the whole picture.
• Hybrids without clear rules: every piece of content becomes a negotiation between the central team and the markets.
• Treating transcreation as translation: a rewritten campaign needs creative review, not just a language check.
• Deciding once and never revisiting: markets grow, budgets change, and the right balance moves.

The balance also depends on what markets can actually deliver. A market with no production capacity cannot run local production, whatever the strategy says.

Where mydubly fits in either model

mydubly is a tool for the translation side: it takes a recorded video or audio file and produces transcripts, subtitles and, if chosen, a dubbed version in one of 21 languages, with the spoken language detected automatically. That makes it most relevant to central production and to the "local first, then shared" hybrid, where a successful local video is subtitled or dubbed for other markets. The video translator page shows the process, and the video localization page covers wider workflows.

It does not produce local content, adapt jokes or references, translate on-screen text, or replace a presenter. A dub uses one of eight stock voices for the whole video, so a testimonial with a local customer speaking loses that person's voice when dubbed; subtitles keep it. Costs are low enough to test a market cheaply: subtitling a two-minute local ad into another language costs 5 credits (0.5¢), and dubbing it costs 100 credits (10¢). The comparison of subtitles and dubbing helps decide which to test.

Next step: sort your content types

List your main content types and, for each, answer the test question: would it matter if viewers noticed it was made elsewhere? Place each type toward the central or local end accordingly, choose a hybrid where it sits in the middle, and agree the rules with your market teams. Revisit the list each year, or when a market's size or team changes. For techniques that stretch central content further across markets, see repurposing video into other languages.