Linguistic Testing – Setting up globalisation in a quality way
In today’s digital world, everyday products are rarely confined to just one geographical market and the software industry is no exception. Making efforts to “globalise” a product in the engineering stages, so that it is suitable for worldwide markets, is essential.
This is not a recent trend: organisations have long factored internationalisation and localisation of software products as important considerations in the development stages. After all, for businesses to succeed internationally, a product must have international appeal.
A huge part of this is ensuring the language elements of the product are appropriate for the region it is being launched in. And this is where linguistic testing comes into the picture. But what is it, and why is it so vital to a development process?
Why is linguistic testing so important?
Linguistic testing, like quality assurance (QA), is an essential process in localisation projects, especially for websites, software and applications where any errors or ambiguities can cause issues for users. It can broadly be defined as language-specific testing and a language quality assurance step.
Because software and website localisation generally use tools and technologies outside of the actual running application, linguistic testing is a process that typically occurs directly in the final product. And it can be invaluable in pinpointing potentially damaging mistakes like untranslated text, inconsistencies in terminology throughout the user interface (UI) and inappropriate cultural references, among many other things.
The usability of a webpage or UI can be heavily impacted by translation errors. A localised product full of errors requires updates to rectify the faults, which costs the business valuable time, resources and money. Naturally, rolling out a product across different geographic locations is bound to have a few linguistic issues after localisation, but fixing those issues should be quick and inexpensive.
Against this backdrop, linguistic testing aims to solve these issues. Such processes enable the tester to review the translation and correct any contextual or truncation issues to ensure that the final content is correct and flows smoothly.
Linguistic testers will look at the built product, running it exactly as local users in a specific country will be using it. In doing so, they will see the problems that the out-of-context translation process may have caused and rectify them before the ‘real’ users are live on the product.
The difference between ‘overt’ and ‘covert’
Test coverage considers two categories of linguistic testing, namely the “overt” and ”covert”. Overt issues are the more obvious ones, such as grammatical errors, typos, translation issues, missing voices, confusing instructions and complex animations.
On the other hand, covert issues are more subtle cultural aspects, such as geographically-powered usability or associations with colour. For example, in the United States and much of the Western world, red is considered a colour of danger or warning, whereas in China red is a colour that connotes wealth and power. Taking such aspects into account is vital for the comprehension of the reader.
Other factors include Lunar calendars, holidays and other nuanced cultural considerations. The engineering team, including the stakeholders, need to understand that, alongside the usual ‘spell check’, there are societal and cultural attributes that need to be engineered and tested for as part of the localisation process.
Ensuring that both covert and overt categories are accounted and tested for leads to a smoother roll out of the product and avoids a cultural faux pas or miscommunication that can ultimately damage the brand.
The consequences of not testing
While it might be tempting to cut costs or rush a product to market, the linguistic testing phase is essential and not something that should be skipped. Often, a company that does not engage in linguistic testing runs the risk of having to subsequently make changes at a design level, which is more disruptive and far costlier.
Another risk is that any localised product becomes filled with textual as well as functional or display errors. As well as potentially interfering with the reputation of the brand, linguistic, formatting and technical errors can severely damage the end user experience, making a product frustrating or impossible to use.
Ultimately, linguistic testing is critical after translation and prior to the delivery of the final product. Any discrepancy or shoddy error in a product will hurt the reputation of the business in that region if it is incomprehensible or does not conform to the cultural and linguistic requirements of the region.
For the end user, an emphasis on quality assurance at all stages of product development demonstrates the company’s commitment to understanding the needs and requirements of its customers and users.
Pooja Tyagi, Testing Engineering Specialist, NTT DATA