Contact us
Contact us

The argument

Stop asking for translation. Start talking about applied linguistics.

Ask an AI chatbot what an advertiser needs to run campaigns in another language and the answer starts and ends with translation. Hannes Ben argues that this framing is the industry’s core problem. Translation is one skill among many: localisation, transcreation, origination, copywriting, machine translation with human post-editing and creative adaptation all serve different purposes.

Because machine translation made language work look simple, fast and free, it has been commoditised. In many marketing budgets, linguistics is still an add-on or an afterthought rather than a strategic element from start to finish.

Three insights from the essay

01

No economies of scale across languages

More volume across more languages brings no efficiencies. A German writer can’t start producing Chinese content because they have spare capacity; each language still needs its own team.

02

Cheaper locations don’t mean cheaper content

An office in a lower-cost location doesn’t make Swedish, Danish, Dutch, Korean or Japanese content any cheaper to create well.

03

AI changes the job, not the need

As AI takes on basic text and predictable tasks, linguists spend more of their time on research, cultural adaptation, fine-tuning and measuring what resonates.

Make AI your friend, not the thing you fear

Written in 2023, as ChatGPT reached the mainstream, the essay closes with a practical stance: leave basic text creation to AI, automate anything predictable, and put human expertise where it changes outcomes. An incrementality-testing framework then shows where content resonated, to what extent and which KPIs it moved.

It remains a useful brief for any marketer deciding where people add the most value in an AI-assisted content workflow.

Read the essay

Explore our AI localisation and post-editing and transcreation services.

Related research

Cover artwork for "Data Science and its Practical Applications in Multilingual Content", showing a spiral tunnel of words in many languages

Data Science and its Practical Applications in Multilingual Content

How natural language processing, sentiment analysis and incrementality testing make multilingual content more effective, and why human linguists remain essential.

Cover artwork for "The AAA of Multilingualism" by Hannes Ben, showing "hello" written in different languages on a chalkboard

The AAA of Multilingualism

How multilingual competence is acquired, assessed and applied, and why native-speaker status alone doesn’t guarantee excellent language production. By Hannes Ben.

Get Started

Every beautiful relationship begins with a simple hello.

So let’s chat. It might be the start of something.

Contact us