Siri can answer questions and make jokes in 17 languages. Is that a lot or little? How many languages should your startup know to reach the entire target audience? Jeni Mayorskaya, the head of product at Smartcat.pro studied the most successful companies and found interesting results.
Functionality and price is only a part of the success of many startups. The popularity of many “go-to” services is caused by its beautiful, easy-to-use mobile and web interfaces combined with an edgy brand voice. And doing so requires good writing and excellent translation.
Check out the Hotel Tonight app and the quirky-yet-fresh copy that draws you in. Or catch a ride with Uber in France, whose interface stylishly adjusts to let you order by tapping “Commander ici.” Conversely, poorly translated or bland interfaces can discourage users from purchasing.
Over the last decade, the tech industry has gotten very good at scaling technology. A startup can go from 100 users to a million without blinking, and the same applies to new markets. Startups are fighting for global domination and local market share, which compels them to constantly localize their products, increase regional availability, and set up support centers.
The Problem Is Scaling Content
To date, the content localization process has been painfully linear and incremental. Every time you add a new market with a new language, you need to bring over all of your existing content to match. So Hotel Tonight, for example, has to localize over 3,500 hotel listings for each new language. Airbnb, Uber and their ilk face enormous avalanches of content to localize, as well. So at which point do the costs of localization outweigh the gains for budding startups?
Last year, more than 50 percent of U.S. companies sold products and services outside of the U.S. In 2015, reaching 90 percent of the online audience worldwide requires localization into more than 25 languages. To reach the same proportion of web surfers in 2020, you will need 48 languages (data from SVB IEO Report 2014).
We studied the language issue empirically, looking at “unicorns” as a way to identify the languages and markets addressed by the most successful startups. These unicorns are members of the club of tech outfits and venture-backed private companies valued at USD $1 billion or more. Of the ones in our sample, 35 of them are U.S.-based (including such heavyweights as Uber and Airbnb) and 15 are located in Asia.
The unicorns are innovation leaders in software, e-commerce and more. So then we started wondering: what are the differences in internationalization between U.S. unicorns and Asian ones? Which languages do these companies “speak” with their customers?
We have been researching this language phenomenon for one month now and have identified dozens of startups. For example, Tinder speaks 33 languages and it covers the most of any unicorn, which speaks of the company’s global ambitions. Uber is next, with 32 languages (including Bokmål and Hebrew). Jawbone and Box follow, and Pinterest is surprisingly well localized, boasting around 30 languages.
You don’t need a Ph.D. to see that the commitment to internationalization is very different between the two regions. Asian startups are lagging behind their American competitors. U.S. unicorns utilize at least 5x more languages.
Which languages should a startup target first?
Based on what our unicorns did, startups should start with English, followed by French, German, and Japanese. Chinese is only 10th on the list of targeted languages (chart below).
Why is Chinese localization so unpopular, relative to the immense size of the Chinese market? One explanation is the Great Chinese Firewall and users’ preference for indigenous platforms. In China, Facebook, Twitter and other Western mainstays are either completely absent or bypassed in favor of homegrown products. These competitors include Baidu, Weibo, Sina.com, and others. So simply localizing your product may not guarantee success if it does not fit with the habits of your target audience.
Most Asian tech startups don’t localize for Europe and U.S. yet
U.S. unicorns average 9.4 languages and Asian unicorns average just 2.3 languages. Surprisingly, U.S. unicorns on average speak more Asian languages. This is because Asian unicorns average only 1.3 Asian languages.
What’s more, U.S. unicorns speak way more languages than their Asian counterparts. We believe that Asian unicorns will move to close this gap by introducing their applications and products in more languages. The Chinese government (not unlike the governments of other countries) seeks to reduce dependence on foreign platforms and has committed to supporting local coders. Combining this domestic “core” with growth on foreign markets offers a tempting opportunity.
However, localization is a necessary-but-not-sufficient condition for growing internationally. Even if Asian unicorns offer their products in more languages, will they actually gain market share? Are their products truly compelling, or just “fast followers” with little to add? Localization is only one part of the process for delivering new and unexpected value to users.