Translation at any level can be daunting, but it’s increasingly becoming business critical as our digital-first world demands great user experiences across all the markets you sell in.
It can be confusing trying to understand where to invest and what good really looks like. In this blog, Jeremy Clutton explains 5 key things you should consider for a global translation strategy.
Despite the huge efforts that businesses pour into getting their global messages right, often the translation of this content for local consumption isn’t given the time, attention or resources it requires.
This is unfortunate, but not surprising given the language industry struggles with how to communicate the choices and routes available to customers. This includes communicating the core elements of a translation strategy (costs, onboarding, the risks and rewards around where you invest time and resources), or explaining seemingly mixed messages about getting high-quality translations cheap and fast.
And it’s never been more important to communicate effectively with your colleagues and customers, wherever they are.
So if you’re looking to get your multilingual content right, these are the 5 key things you need to think about:
1) Not all content is created equal
As you create your content, think about the complexity of your source language and talk to your language partner about what translation level is needed to support it. Most businesses have lots of different types of content and using the same translation service level for all of them doesn’t make sense.
Consider doing a content audit and then asking about different service levels as appropriate for the various content strands you have.
A simple example would be the difference between highly creative advertising copy versus simple product data. The former will use carefully constructed branded messaging, which will need more time and resources to get your messages translated exactly right, while the latter could achieve the right result using AI-powered neural machine translation supporting human teams at reduced turnaround times and spend.
2) View your approach to language as an investment, not a cost
Link an ROI to your translation to properly measure your return. Remember: content has a purpose and the outcome should outweigh the resources you invest.
ROI for your business could mean:
- increased conversion or reduced bounce rates for your ecommerce website
- better employee engagement from your global workforce
- better UX in your app
- fewer resources required to support your language workflows.
3) Plan against business goals
Think about what is most clearly going to align against your company’s goals.
Is it speed, is it differentiation? Is it accuracy? What is the cost of getting it wrong?
Remember to be clear upfront on your goals and the potential compromises they may demand. For example, there is always a balance between the speed to market of your content and achieving the appropriate quality level.
By using technology you can support these goals: creating and embedding terminology specific to your company, using advanced file filtering, having internal reviewers working in the same translation tools and with the same language resources, or leveraging AI alongside human translation teams.
Making sure you employ the right approach and technology is the best way to align what you have to achieve with the time and money available.
4) Use automation to reduce strain and increase speed
Automating your workflows is a key way to reduce strain around managing language. You should evaluate the platforms and tools you use and explore ways that you can reduce touchpoints where possible.
APIs, for example, can connect your content platforms directly to your language partner’s translation management system (TMS), decreasing friction in the process and reducing translation turnaround times.
Speak with your language services partner about your workflows, platforms and their capabilities early in the process to solve challenges sooner rather than later.
5) Discover how technology supports your translation
In the language world, technology moves fast and it can be hard to understand how it all fits together.
Put simply, translation takes place in a translation management system, which controls all necessary elements of the process. The TMS might connect to your content platform via API as mentioned above, or allow for automation in task allocation. It also houses the CAT (computer-assisted translation) tool where translation and in-country review happen and it’s where your translation memory and terminology live, as well as where highly-productive AI-powered neural MT engines are plugged in.
There’s a lot of different ways to use a TMS depending on your preferences: you can buy the platform and manage a multi-vendor solution, you can have a hybrid approach where you use the TMS to outsource some work and do the rest in-house, or you can utilise the benefits of the technology via your language services partner.
It’s really important to understand the options here. Your language services partner should ensure you choose the right solution that also has the technology you need to make your content scalable.
To take one option as an example, AI-powered neural machine translation offers a hybrid approach to traditional translation, and is capable of translating your content faster, at a reduced cost, and to the same quality expectation.
AI-powered NMT works by “cascading” changes across all your content live, “like magic” as one of our translators said. What does this mean? For our translators it means improved efficiency as they can translate more per hour, and for you it means cost savings and reduced turnaround times. AI-powered NMT works best when applied to large sets of structured content, such as product descriptions.
When used appropriately alongside our translation teams we’ve seen productivity boosts of 40%.
Obviously, what you decide to do should be aligned with your challenge and the purpose of your content. There are some amazing solutions out there. The key takeaway for me is to plan and scope your solution, give time to the process, leverage the technology and remember that while it isn’t easy, it doesn’t have to be hard.