Column

Client Talk

Translation at SAS

Terena Bell

Terena Bell is an independent journalist writing for The Atlantic, Washington Post, Fast Company and others. She is former CEO of In Every Language and was on the GALA and ALC boards.

When do clients say professional translation is worth it — and why? By connecting with clients outside the traditional sales environment, we hope to discover what really drives purchasing. Every column in this series offers a different profile. Our challenge is to find patterns from one to the next: together, what do these interviews tell us about how clients see our industry as a whole?

The client

Patricia Sainz purchases translation for SAS, an analytics software and services company based in Cary, North Carolina. With the title of TMS specialist — Globalization and Digital Engagement International Support, localization is her full-time job. Sainz is originally from Buenos Aires, Argentina.

A native Spanish speaker, she’s also fluent in English, Portuguese and German. “I’d love to truly embrace more cultures by visiting new places and learning more languages,” she says, noting she’s traveled throughout Latin America, Germany and the United States.

On a scale from 1 to 5, where does SAS score professional translation?

“Definitely a 5,” Sainz says. “Many customers worldwide expect everything on their own terms and that includes a preference for communicating in their own language. An accurate translation will resonate with target audiences as if it were written in the native tongue, creating a true cultural connection with prospects, clients, users and advocates.”

Patricia Sainz

What do they buy?

With customers in 148 countries, SAS operates in 22 principle languages with two more, Thai and Indonesian Bahasa, planned for addition at the time of writing. The company also supports software in 27 languages, listing Brazilian and Iberian Portuguese separately.

“SAS marketing content is authored in multiple languages around the world to meet regional needs; but the English content that is authored at the headquarters in Cary, NC, is often translated into multiple languages,” Sainz explains. The company also buys translation for legal documents, support documentation, social media, customer case studies, and go-to-market assets and collateral — as well as software and video localization.

With some content authored in-language, multilingual communication must be well-organized to ensure continuity in customer experience. Sainz oversees localization for the site and go-to-market content, partnering with other departments to keep the company’s message focused.

She says SAS is “moving away from a decentralized model to a more centralized translation program — especially for marketing content” and that the company does have global guidelines defining what gets translated. But for that author vs. translate decision, she explains, “We give heavy consideration to local offices’ resources availability and budgets. How technical and specialized the content is also plays a predominant role in making these decisions. We’ve just started to test adaptive machine translation capabilities to explore if we can increase our linguistic reviewers’ productivity while maintaining the same level of quality and decreasing time-to-market. If the outcomes are successful we will expand this pilot for languages that lack local resources and where a basic translation will still perform better than English.”

Sounds great, but there’s a problem…

SAS would like to translate more content faster. “In an ideal world,” Sainz says, “we’d love to translate all the key content nearly in real time, but this is far from being a reality. In order to prioritize content and to choose the right translation options, we’ve started to apply a very interesting concept as part of our translation strategy — minimum viable content.”

Minimum viable content is a localization philosophy based off the idea of minimum viable product (MVP). With MVP, startups release the simplest possible software that meets user needs, adding complexity as they go. So minimum viable content, Sainz explains, is “translating the minimum amount of content with the maximum impact in our audiences. When entering new markets, the concept of minimum viable content is useful and can be thought of as the smallest catalog of content and languages that can be released while still providing a valuable user experience. Additional content units are then layered on top of this foundation. Then, when user feedback validates increased investment in our localized content, an agile translation strategy can be employed to gradually realize our full content catalog.”

But doing this requires data, which SAS gets from its translation management software (TMS), something Sainz also ranks as a 5 in importance: “To optimize quality and return on investment, a translation technology management solution needs to allow for automated and customized workflows that can include different combinations of human and machine translations based on the translation strategy.”

So what’s the pattern?

Like Sonia Zamborsky at Marriott International from our May issue, Sainz credits TMS for SAS’s translation success. The two companies definitely scale localization more effectively than clients that don’t use a tool, like Papa John’s International, profiled last April. And although they designed their own platform, last November/December’s client, Buzzfeed, also noted translation management technology’s essential role.

While we don’t want this column to become a TMS commercial, it’s hard to ignore the theme. “New advances in technology can help us globalize quickly and efficiently, decreasing the distance between content created and content deployed,” Sainz explains, making minimum viable content a reality. “Process automation is now a necessity for any translation strategy to achieve significant scale. Smart software should be used to detect new source content from wherever it’s created, instantly pull it into the translation workspace, carry the content between workflow steps and place the final product back in its associated platform for the end user.” Any language provider her company considers must use this technology.