This is What a Highly Mature Localization Program Looks Like

A thousand translators. Millions of words. Turnaround times in hours. More than 100 languages.

These are the hallmarks of a conventionally “mature” localization program: a program of continuous process improvement that bakes global awareness into all facets of the business. In an ideal world, it’s a level of localization maturity you manage proactively and repeatably as your company expands and content volumes increase—just look at how Microsoft and Oracle localize.

But let’s rewind. Maybe you’re just now needing to optimize your localization program and wondering how much work is involved. How do you manage projects of such scale and complexity? What are the implications of a continuous localization model on your supply teams? How do growing programs shift from waterfall to agile? And how did the tech giants get there?

The answers to these questions are changing as localization itself evolves, but there’s much to learn from the pros. Here are the latest trends we’ve taken away from our own clients’ experiences.

New challenges

Unstoppable business growth and customer demand have resulted in a number of challenges to localization programs, demanding a reconfiguration of the way things get done. For example:

• Scale has exploded. The numbers of languages, words and projects have multiplied with the global demand for content. To handle their programs’ increasing scope, companies are moving from in-house translators to needing multiple vendors.
• Velocity has increased. Gone are the days of monthly or quarterly projects with ample schedules. Now, project handoffs are daily and turnaround times are sometimes same-day.
• Products have changed. Digital products must now run on all devices. This impacts localization project elements such as functional requirements and testing.
• So have product delivery models. Now that everything is online, no one goes to a store and talks to a sales clerk. Products can be received and used immediately.
• Customers have different demands. They will no longer wait; they want updates tomorrow, not in six months.
• Customer focus has come to the fore. You might have heard this referred to as “customer obsession.” The customer decides quality now, whereas in the past it was based on the number of errors found and whether deadlines were met.

Language service providers (LSPs) now find themselves operating in this environment, and while it’s no small feat to meet these challenges, it is possible. Companies higher on the localization maturity ladder are almost there.

Vendor responses to business changes

When an enterprise evolves as quickly and dramatically as those undergoing agile and digital transformation do, the way vendors manage the work must change just as quickly and drastically. How? Ivan Lukavsky, Senior Program Director at RWS Moravia, spoke to this.

First, production must adapt to meet program challenges and enable continuous localization—for example, by:

• adding custom-built engineering checks to speed up work and reduce human error. They can run all the time, invisibly. These checks look for any errors that could compromise the functionality of the product or negatively impact user experience.
• using advances in recycling to speed up productivity and reduce cost. This means that a very small percentage of words requires human effort.
• developing new models for quality. There is a quality risk with moving faster that must be managed. Quality processes must work immediately—within a matter of hours.

For some companies, various elements of their business model might change substantially as well, including:

• Vendor relationships: Vendors are no longer delivering assets, but rather a service. They are accountable for the entire program instead of files, thus giving more autonomy to the vendor and less for the client to manage. Work is governed by a Service Level Agreement (SLA), and the conversation is now about the “health” of the service, not the individual transaction.
• Methods of governance: Programs are no longer measured by numbers of errors and on-time deliveries, but by KPIs such as turnaround time, volume and spend. (One interesting KPI tracks interactions: the goal is for meetings and email exchanges to reduce.) Also, programs are now best managed via dashboards, which give a view of the entire program rather than by project. The days of Excel status reports are over.
• The team’s mindset and culture: Localization PMs have stretched and grown into new roles and are now business managers. They also have to be creative and flexible when localization programs change constantly. Quick response is paramount. They are responsible for driving new, innovative processes rather than just executing what’s in place.
• Measurement of performance: When localization programs become “customer obsessed,” their success and quality are no longer based on QA scores, but on how the product is perceived by users. Any complaint is considered severe and is treated with utmost attention. And it’s collaborative: the vendor is now part of “listening loops” and helps respond to the needs of the customer.

The future of agile collaboration

What’s next? The evolution of mature localization services shows no sign of slowing down. It’s not a matter of when, but of what new complexities and challenges will show up—and all we can do is predict them. For starters, we will see more use of crowdsourcing. As an example, a crowd of resources such as university students can do bug validation and QA.

Of course, there will be more machine translation—always a viable way to increase productivity and the ability to handle high volumes and high speeds. But machine learning is also expected to play a bigger role in other aspects of programs; think content selection and quality management.

Lastly, the customer and vendor journeys will continue to intertwine. No longer are there two sides of the fence. A program increasing in complexity requires a vendor who can be creative, flexible and responsible in planning for grand scale.

Watch this space for more discussion on how client needs and localization models evolve in parallel. And you can always reach out for advice on how to take your program to the next level.