Affirm offers both merchants and consumers financial products that are designed to make their lives easier. They offer a buy-now-pay-later service for shoppers, as well as a high-yield bank account. For merchants, Affirm offers several solutions aimed at helping businesses attract new potential customers and, ultimately, convert browsers into customers.
Darci Stacy, a knowledge manager at Affirm, and her team have taken on the challenging work of centralizing and scaling their company’s localization program. At Smartling's Global Ready Conference, Darci discussed the challenges her team has faced around budgeting and bandwidth and describes the strategies and tools they’ve leveraged to scale while maintaining high standards for quality.
Expansion Plans and Centralization
In 2020, Affirm announced its acquisition of PayBright, a leading provider of buy-now-pay-later services in Canada. This move would pave the way for Affirm to expand its reach and enter the Canadian market.
Prior to the acquisition, PayBright had taken a decentralized, ad hoc approach to localization — where internal employees localized and translated content as needed in addition to performing their other day-to-day duties. Affirm, however, recognized how this approach could hinder scalability and quality, and opted to centralize its localization efforts within the knowledge management team.
With all translations flowing through one team, others in the company were free to focus on their primary responsibilities. The knowledge management team then centralized all translations within Smartling. To ensure quality and consistency and confirm the application of the company’s preferred terminology across content types, they leveraged the translation review step and began building out their glossary.
Addressing Localization Budget Challenges
Expanding into new markets and working on projects with many stakeholders across different teams can be challenging — particularly when navigating budget allocation dilemmas. Companies often have to be very intentional about how they deploy localization dollars.
For Affirm, working with budget constraints meant the knowledge management team had to find ways to build efficiency in the translation process. To do so, they leveraged the following tools and strategies:
- They became the internal experts on Smartling. The team centralized the task of troubleshooting and helping stakeholders get set up and accustomed to Smartling. That ensured they were aware of all the issues people were running into and ultimately helped them avoid the same issues occurring in future projects.
- They set up a dedicated channel in Slack. Any team within Affirm could use that channel to ask questions and troubleshoot issues related to localization or their use of Smartling.
- The team documented their processes. That way, all information was available on demand. Other stakeholders could consult the documentation and mostly self-serve, getting help from the team only when they had questions or ran into issues.
- The team embraced the value of translation memory. The team prioritized leveraging Smartling to build up Affirm’s translation memory. That, in turn, has ensured that they don’t have to pay to translate the same string twice (or three or four times) — a fact that, according to Darci Stacy, has helped everyone embrace the idea of centralizing translations within one team and solution.
- They are automating as much as possible. The team made sure they had projects set up in Smartling for each of their different content types. These include Affirm’s app and website, customer communications, help center and support content for both the company’s merchants and consumers, and internal content for agents (talking points, email templates, etc.). That way, they could automate as much as possible, spend as little time as possible doing manual work, and scale quickly.
Affirm’s Translation Mix
For Affirm, high-quality translations are paramount. The specialized and highly regulated nature of the financial services industry means that accuracy is nonnegotiable and that the company has specific terminology requirements. As a result, Affirm has currently opted for a human-first translation approach: human translation and editing, with strings being reviewed by the knowledge management team before they are published.
They are also building out their glossary. As the team reviews translated strings in Smartling, they identify terms that need to be validated with other stakeholders within the company. Each quarter, Darci then brings that list of terms to the Affirm legal and compliance teams to confirm that these are the best and most appropriate terms to use. Any necessary changes are made and provided to Smartling to ensure consistency across the board.
That said, Affirm, like many other companies, is also interested in using the latest AI-powered solutions to do more with less. Darci mentioned that once both the translation memory and the glossary are built out for the Canadian French market, her team may start experimenting with AI-powered machine translation for some of the lower-profile content types.
—
Want to learn more about scaling localization through budgeting challenges? Watch this year’s Global Ready Conference in its entirety. All sessions are available on demand here.
