How an international approach to linguistics supports machine learning

If you read our earlier blog on how international targeting gives brands a powerful reach opportunity, then you’ll know our view that operating across markets enhances smart bidding. But there’s more – what if there was a way to structure international reach so that machine learning can spot patterns and opportunities more easily? And what if this helped advertisers retain control and predictability over their media spend? Well – there is. Read on to find out more.

Let’s start by talking about NLP

NLP stands for Natural Language Processing and is defined by the Oxford English dictionary as “the application of computational techniques to the analysis and synthesis of natural language and speech.”. This field has historically been dominated by tech giants which have developed various voice-enabled products. It has recently seen a resurgence with the release of ChatGPT. But international marketers, translators, and other language and culture geeks have been thinking about NLP for decades.

This is because launching a digital acquisition strategy in a completely new market requires in-depth international keyword and localisation research which is very different to an English-only project. Even once work is complete and your international SEO and paid search activity is live and delivering, optimising it for sustained growth and performance requires processes which are tailored to international activity.

To deliver these results, you have to understand the relationship between linguistic concepts – aka named entities – in common user syntax, and online performance or conversion rate. Common keyword research and analysis typically asks: is this a relevant keyword for one of our products? By contrast, international marketers ask: if this keyword is relevant, why exactly is that, linguistically speaking? Can we quantify just how relevant it is? Does adding or removing an entity make it more or less relevant? This in-depth analysis gives us a valuable edge when it comes to forming a strategy and we continuously develop our best practice based on this experience.

So what point are we making?

International marketers do much more than simply gather back-translated high-relevance local keywords under one umbrella and then group them into themes and products. Instead, we plan and implement tactics in a way that highlights, prioritises, and structures linguistic patterns so that they reflect natural user behaviours in any language. This specificity makes the machine’s task easier. We see it as a form of signposting that helps the algorithm match our keyword portfolios, ads, and content more efficiently to the right user, auction, or search term.

Let’s conclude with a metaphor. Imagine a competition where you had to do your grocery shopping as fast as possible, but in a supermarket you had never been to before and with no signs in any aisles. It would be much easier if there were signs, right? Our international approach provides the signs and further to that, allows us to predict more accurately how a particular keyword group is going to perform compared to another, as with a deeper understanding of user intent comes a clearer idea of what action they may take. This ability to predict the potential performance of a specific investment adds up to a much stronger digital strategy and more reliable budget planning.

. . .

If you would like to find out how Oban’s award-winning paid media team can help to accelerate your brand’s international growth, please get in touch today.

Oban International is the digital marketing agency specialising in international expansion.
Our LIME (Local In-Market Expert) Network provides up to date cultural input and insights from over 80 markets around the world, helping clients realise the best marketing opportunities and avoid the costliest mistakes.