Goodbye, Source Text A Different Way to Conceive Localization
Events dedicated to localization, such as the 40th edition of Localization World held in Estoril, are a good way to take the pulse of our industry. While most of the discussions inevitably center around the usual suspects – machine translation or globalization, to name just a couple – every once in a full moon, a hidden, wholly unexpected gem makes an appearance on center stage.
Such was the case for Giulia Tarditi, Head of Localization at Monese, who took the spoils during the Process Innovation Challenge (PIC) at LocWorld. Giulia presented the different approach they took to localizing the Monese app, which does away with source text altogether. She aptly named it “Goodbye, Source Text!”
You might rightfully wonder, what this is about. We did too and decided to follow up with Giulia. However, we ended up not only discussing the project itself and how it came about, but we also found ourselves interviewing one of the brightest minds in the language industry. Giulia, as it turns out, is a woman on a mission. The “Goodbye, source text!” innovation is really just its latest manifestation.
Goodbye, source text: boosting user conversion & localization ROI
The “Goodbye, Source Text!” innovation contradicts the traditional way you imagine translation. Under regular circumstances, you would start with existing source content, which a linguist then translates into the desired target language. Giulia and her team ended up scrapping the whole idea of having source text to begin with.
Instead, the content design team at Monese creates a brief, which describes the functionality of a button or a screen within the app. From there, it is the linguists’ responsibility to imagine what the best translation would be. The critical point behind this idea is that the linguists’ imagination is not hindered by the original source text. Essentially, they are the end-users of the app.
Giulia and her team have leveraged negative user feedback, filtering in through customer services to pilot this new method for two languages, Romanian and Bulgarian. And they have done it to great success. They concentrated their efforts on the first few screens a potential end-user has to go through in the Monese app. The resulting translation has led to a sizeable increase in user conversion. At the same time, the series of improvements to the workflow has the team on track to reduce localization costs by 80 percent.
“I’d rather spend 90 percent of our localization budget on the 15 percent of content which drives conversion.”
Where localization meets user experience
The driving force behind the whole idea of “Goodbye, Source Text!” has been Giulia’s vision to “make localization a growth driver.”
If we take a step back, it’s not necessarily about a localized end product. The effort to localize should go hand in hand with improving user experience (UX). Positive UX then drives conversion. Imagine a localized app which has better conversion rates in English than in German. On the one hand, a German user base exists precisely because there is a localized version of the app. On the other hand, it’s not growing quite as fast as the English one. In Giulia’s own words, “Bad UX often comes from poorly translated content.”
It’s this idea of integrating localization upstream and influencing product development that Giulia was able to act upon in one of her earlier stints as the Head of Localization at Badoo. “At Badoo,” she says, “I wanted to move localization from customer services into product [development]. What I did was I looked at big companies such as Google or TripAdvisor, to see where localization was integrated there and whether it was successful.” The idea was to work on localization becoming a factor for growth. During her time at Badoo, she helped grow the number of supported languages of the platform up to 43, while doubling conversion and boosting revenue by up to 10 times in mid-tier markets.
Challenge? What challenge?
The terrain to apply novel ideas is difficult to come by. As we wrote in a recent piece on data services, only IT giants such as Amazon or Google really have the resources for constant experimentation. Opportunities to try something new don’t come around often. Giulia agrees, saying that, “It takes a specific leader to convince the team to try it.” A track record of developing localization programs at different companies helps to get buy-in, too.
“Everything needed to change [at Monese],” Giulia says. “Even the way we do English. If all languages are created from a brief, English has to be produced at the same time as the other languages. We had to design a custom workflow and a custom issue type in JIRA to keep track of this process.”
Resourcing and subsequent pricing of localization efforts have been a challenge unto themselves. With “Goodbye, Source Text!” there is no price per word because there is no text to translate. Giulia works with freelancers at Monese. They are paid on an hourly basis for the work they put in. When they are on-boarded, they need to perform a test to see how they perform without the crutch of a text.
The next frontier at Monese
Talking to Giulia about the new process she’s been able to implement at Monese, you’d almost think that “Goodbye, Source Text!” is a lightning-in-a-bottle kind of effort – one which only happens if certain conditions are met or that is only applicable in a specific situation. Giulia acknowledges that there is still work to do. The new approach is yet to be live-tested on all of the 12 languages Monese currently supports.
“What’s next for us is identifying all key conversion screens in the funnel and re-do them with the new approach.”
In effect, this approach lends itself well to a certain content type. Marketing content, for example, works well with this model. Things like push notifications, billboards, or ad campaigns are well suited to this approach. Any other type of content where the text is sufficiently short, meaning it can fit in a single paragraph, will work well too. “I don’t see this approach viable for all content. Certainly not for something like translating a novel,” Giulia opines.
It is Giulia’s hope to influence localization budget owners, whether in marketing or product development, so that they try out this approach. “I want to show them that it is a viable way of doing things,” she says.
As for Giulia herself, her mission doesn’t end here. Besides her current efforts, she plans to work on a Ph.D. in neurolinguistics, comparing creative writing with translation.
The language industry needs more thought leaders who are pushing us to try new things. Watch out for what’s on the horizon!
Stay up to date as Nimdzi publishes new insights. We will keep you posted as each new report is published so that you are sure not to miss anything.