Not lost in translation: Freelance localization at Dataduck
Dataduck’s penchant for excellence often leads it through the paths less traveled, setting it apart in a number of areas. In the search for higher-quality, more closely translated content, the digital marketing agency opted to establish an in-house localization team — a decision that its clients benefit from every day.
Collaboration in spades
As Dataduck’s clients are primarily in fintech, its teams work with fairly similar products on a daily basis. This means that our colleagues are well-versed in modern finance, from trading to Web3, and use their know-how to produce high-quality, impactful content that achieves the goals of Dataduck clients, whether that be growing the local audience or entering a new market. In-house teams of creators, designers, editors and developers collaborate together to produce marketing content and strategies such as video ads targeted at new and existing audiences, landing pages and digital marketing campaigns — and it is all localized into the languages of the target audience by Dataduck’s in-house localization team.
The in-house localization team at Dataduck consists of four localization managers, a translator and a team lead. They work closely with the production and marketing teams to facilitate fast and high-quality localization. The team lead handles localization tasks too, as it provides a bottom-up view on the collaboration process with other teams.
These teams also analyze metrics and adjust accordingly to ensure the content’s peak performance. While most of the content is produced in English, the narrative needs to be globally consistent and locally relevant. This is where the localization team and freelance translators come into play.
Why freelancers?
- Freelance translators are already familiar with the content being worked on, which means managers don’t need to debrief translators on every related project.
- The Tone of Voice is already known to freelancers, who will not need to learn it for each project as an agency-appointed translator would.
- Collaboration with freelance experts is both cost-efficient for the company and beneficial for freelancers.
The localization team’s numbers:
- Teams: 1
- Team size: 5
- Structure: Centralized
- Vendors: 1 in-house translator, 24 freelance translators, 2 language services providers
- Core languages: 14
- Localization software: Smartcat
- Other technology: Internally developed, continuous localization integration
Getting to this point
Dataduck’s clients being in fintech means that its translation experts must be deeply immersed in the topic at hand. Team members maintain and update the term base and keep a high bar on the quality of the localized versions. However, a glossary doesn’t solve all problems.
The team used to cooperate with both freelancers and translation agencies. The freelancers performed well and the translation bureaus did their best, but there were persisting issues with deadlines and translation quality on the agencies’ side. Eventually, the team decided to do away with the hybrid mode and embrace a freelance-only method.
The set of languages being translated varies depending on each campaign’s goal and the target audience, so the resulting workflows are tailored to the individual needs of each campaign. Today, the company’s core site pages are available in 14 languages. With English being the source language of the content and Russian texts handled by the in-house translator, the team needed to appoint 24 translators (two specialists per locale) to achieve this feat. Two additional vendors are responsible for overflow work and other ad-hoc language requests.
Quality control
Alongside its other responsibilities, the localization team ensures that the localized marketing content is consistent with the client’s products, that technology and phrasing are consistent across the board, and that the user interface is effective and suitable for all locales. Freelance translation specialists review and provide feedback on each other’s work to ensure the highest-quality result and that localized content is compliant with country-specific requirements. This practice helps eliminate minor issues like typos while ensuring the messaging and content connects with the target audience. When checking each other’s work, the freelancers don’t nitpick by highlighting debatable language points or insist on preference-based changes. Everyone’s voice is respected, and only necessary edits are made.
Dataduck is a multicultural company, filled with people who speak various different languages and who are always eager to take a look at a text and speak to how well it delivers its message to the local audience. For the localization team, the consistent lack of negative reports from native speakers — be they users or colleagues — is another sign that the content is well localized.
At present, there is no formal localization quality assurance procedure at Dataduck, but the team uses Smartcat’s QA tools to ensure consistency between terms used.
Task assignment and payments
Dataduck uses Jira for its internal operations and communication. While freelance team members cannot access Jira, they can be invited to Slack, where most of the work happens. The localization team also uses Smartcat’s translation management, translation memory and glossary functions to ensure uniformity of terms used across the website’s pages and marketing materials. With all our translators in one place, assigning people to projects is easy: We simply assign them to a new task and use the “@channel” command in Slack to notify everyone that there is something new to localize. It takes three tools to get the work done: Slack, Smartcat and our internally developed continuous localization tool integrated with Smartcat that can be used not only by localization managers but also by copywriters and project managers. Smartcat’s Marketplace enables the team to make payments; we are always impressed at how convenient it is to perform so many actions through the same platform.
The results
By shifting to freelance mode, Dataduck’s localization team has not only cut down drastically on costs but also sped up the localization process — all while ensuring quality and consistency. Without needing to wait on an external organization, we can just tag the person on Slack, get the job done in a timely manner and move onto the next project. While our continued ability to stay on top of our projects has been made possible by eliminating needless bottlenecks in our workflow, the seamless inter-team cooperation that we have collectively built up is what’s really proven itself as invaluable.
