Creative and language localisation for new markets

Translation is not localisation

Translation moves meaning between languages. Localisation changes the ad so that it works for the market, which frequently means changing what is said rather than how it is said.

The clearest symptom of mistaking one for the other is an ad that reads grammatically correctly and persuades nobody. The humour does not land, the reference is unfamiliar, or the proof point is one that market does not find convincing.

The second symptom is compliance drift. A claim that is permitted and substantiated in one market may be restricted, or simply unverifiable, in another.

Claims that travel badly

Superlatives are the most reliable failure. “Best”, “number one” and “leading” are restricted or require substantiation in a number of markets, and the evidence required differs between them.

Performance claims are next. Percentages, guaranteed outcomes and comparative results are all more heavily scrutinised in some jurisdictions than others, particularly in health, finance and anything aimed at children.

Price claims create a third category. A discount that is defined as a reduction from a previous price in one market may need a documented reference price elsewhere.

Imagery and visual conventions

Imagery carries meaning that translation cannot correct. Gestures that are friendly in one market are rude in another; family composition, dress and household settings read differently across borders; and colour association is not universal.

There is also the question of who appears in the ad. Casting that reads as everyday in one market can read as aspirational, foreign or unrepresentative in another, and that changes the response even when the message is identical.

The practical answer is to shoot or generate variants for the markets that carry the most spend, and to accept generic imagery only where the spend does not justify the production cost.

Language variants and where to spend effort

Not all language decisions are equal. Getting the headline, the call to action and the price statement right matters far more than the body copy, because those are the elements that carry the decision.

Regional variants matter too. A single language can differ enough between markets that one version reads as noticeably foreign in the other, and the fix is usually a light edit rather than a full rewrite.

Where a market is large enough, native copywriting outperforms translation consistently. Where it is a test, a well-adapted translation is usually sufficient to learn whether the market responds.

Building a review cycle that scales

The review that matters is a compliance check per market before spend scales. This is not the same as a legal review of the business; it is a check that the specific claims, offers and imagery in the creative are permitted in that market.

Build it as a checklist tied to the creative template, so it is applied every time rather than only to the first campaign. Claims that were fine in the launch creative are often the ones that cause problems in a later variant.

Keep a record of what was approved and on what basis. When a platform asks for substantiation, a record of the decision is far more useful than reconstructing the reasoning months later.

Testing structure, not just copy

Localisation is also about structure. The same offer may need a different funnel in a market where payment methods differ, where mobile share is higher, or where the dominant channel is not the one you assumed.

That argues for testing in stages: creative concept first, then the localised variant, then the funnel structure. Changing all three at once produces a result nobody can interpret.

Where a market underperforms, the honest question to ask first is whether the offer fits the market, before assuming the creative or the media buying is at fault.

Frequently asked questions