How to make your applications use around the world: 10 tips from CEO Wachanga

Meet: Maxim Kolpakov and his cool startup Wachanga – a platform for family development. The guys manage to grow rapidly in a narrow niche, attract investment, create new applications in a few weeks and endlessly experiment.

Maxim will share conclusions on how localization helps them conquer the world. This is not an exaggeration: Wachanga applications are used all over the world. In a dozen countries, they have already settled in first place in their categories.

The idea of ​​rapid growth due to additional localizations came to Maxim for a long time, and for more than three years now he has been a development model for Wachanga. Today the company’s most scalable application is Clover (female cycle tracker) – translated into 52 languages, a newborn care application Babycare available in 34 languages Pregnancy Calendar – in 26 languages, Water tracker – in 23 languages.

Eight Wachanga applications are combined into a single ecosystem and have 7 million installations, 1,400,000 active users and grow by 10-15% per month in these indicators. We are in Alconost We are glad that the Wachanga guys trust us to localize their products and that we can feel a little involved in the success of this wonderful team!

Maxim and the Wachanga team shared with us their insights on replicating applications around the world.

# 1: Scale with geography

For a b2c product, the easiest way to scale organic traffic is to expand your geography. If you have a narrow niche, then without attracting users from other countries, you will not succeed.

The first scalable product of Wachanga was an application for breastfeeding: mother had to choose which breastfeed her baby (do not laugh, nursing mothers really have a need for such a function).

While Russia was our main market, in terms of money, everything was rather disastrous. Despite the fact that at that time our application was # 1 in Russia among breastfeeding applications, in terms of numbers this meant only 7,500 active users per month.

As soon as we scaled the application to the world, the audience increased 10 times in 8-9 months:

This is completely organic traffic: we launched a poorly optimized version of the application, looked at what users of new countries liked and what needed to be improved.

At that time, the entire Wachanga team lived and worked in the Siberian province, but we made products for the whole world. Our main markets are Germany, France, Brazil, Korea, Japan. Why huddle in one country when in many countries of the world they are waiting for your quality product in their native language?

It remains to decide from which countries to begin the conquest of the world.

# 2: USA is not the best choice

If we roughly outline the main regions for scaling, then we get about the following picture:

  • The USA is a highly profitable, but extremely competitive market;
  • Europe is closer and more understandable to the Russian speakers, easier to understand than the United States;
  • Latin America is a non-competitive market with a large number of inhabitants;
  • South Korea and Japan – in terms of solvency similar to the United States, but easier.

Many developers are attracted by the potential profit from entering the US market (and other English-speaking countries at the same time), and they decide to release their application in 2 languages: Russian and English. As a result, 500 million people who speak English produce a lot of products, so the competition in English is fierce. Conclusion? Betting on English only is a bad strategy.

One of the key turning points in Wachanga’s growth occurred when we realized that the pursuit of international recognition should not begin with the United States. Instead of once again trying to jump into this highly competitive market, we focused on other countries.

There is an acute shortage of quality products in local languages ​​in the world. By localizing and adapting the product to several markets with less capitalization, we got a lot more in total than we could only take from the USA.

Another interesting conclusion about the USA: we had a hypothesis that those applications that were originally published in English and only in the USA were more successful. We tested this hypothesis when we launched Babycare on iOS, but this “trick” had no special effect on growth. Now we release applications immediately to all countries for which the product is adapted.

# 3: Market Analysis: Expectations and Reality

Choosing markets for localization, we focus on the experience of our previous applications, in this regard it is very convenient to have an entire ecosystem of products.
Be sure to analyze the main competitors, which are also scaled around the world. It’s funny, but our three main competitors in the US market are not so actively seeking to conquer other countries. This, of course, is only at hand for us.

Also, to complete the picture, we analyze economic indicators (i.e., the solvency of the population) and the development of the mobile application market. In some countries in our category – Medicine – very high competition.

Wonderful market – Germany. German localization took off on the move, and users constantly demand new paid features in the style of “Shut up and take my money!”

We are also interested in working with Brazilian users: they are both demanding and responsive, giving us many ideas for development. For some applications, we have a shaft of reviews from Brazil (mainly on Google Play).

When we translated into Spanish, we expected a big return from 100 million Mexicans – after all, the largest Hispanic nation. However, the profit came from the Spaniards and Argentines: they were more solvent and Europeanized.

In some countries, the official language is not always the preferred language for localization. For example, while our applications in India were in English, they were actively used. They made localization in Hindi – installations fell. Residents of India often use applications in English, so we rolled back the localization back to English. But the translation into Hindi came in handy: in alternative Indian stores, you need to promote applications in their native language.

English is also widely used in Israel and South Africa. So first you need to check what localization is used in applications in each specific country.

# 4: Minimum content – to translate into many languages

If with games it sometimes rolls to localize only the page in the store, then for the application it is necessary to localize both the application itself and the page. Translation of the page is inexpensive, but if the application itself has a large amount of text, then you are facing a dilemma.

In our first application, Child development, 1 million characters. Translation of this volume into 1 language costs several hundred thousand rubles. Therefore, we translated it only into English and introduced a rule: we create products with a minimum of content. You can translate content into a UGC model (User Generated Content = Content created by users themselves). Some of our applications have no content at all, it’s just a tool, and it’s much easier to scale it.

Our pregnancy application is in second place in terms of content with its 35K words, but we still localized it in 26 languages, because user involvement is maximum right at the stage of pregnancy and in the first year after the birth of the baby. Also in the next paragraph I will explain why this application certainly had to be localized and adapted to local realities.

If you want users to not abandon the product and use it for several years, take care of a quality translation. For multi-year applications, only native translators. We translate to Alconost and be sure to provide screenshots for context. For critical languages ​​we order proofreading.

To quickly translate various interfaces into many languages ​​at the same time, we use Nitro online service. This is our favorite service for translating small features, a couple of words into 52 languages.

We have many specialized concepts, and for an accurate translation it is important for us to show the translator how this screen looks and how the text should be perceived. In Nitro we write comments and attach screenshots, check the boxes in front of the necessary languages, and this information (comments and screenshots) is automatically duplicated for each of 52 translators.

Right – comments and screenshot for translator

# 5: The need to adapt content is an argument in favor of localization

Although we strive to create products with minimal dependence on national characteristics, in some of our applications we can’t get anywhere.

There are a lot of differences. In different countries, pregnancy is observed in different ways: while in Russia women undergo a lot of tests and often go for routine examinations, then in many other countries there is no need to see a doctor before 6 months of pregnancy. Diet recommendations and lists of things in the hospital differ.

How do we get information about what is accepted in each country? To compile the content, we do not hire a simple translator-copywriter, but a physician with experience in obstetrics and gynecology or in pediatrics.

In the application Babycare, where the main focus is on feeding the baby, we must take into account that foods for the introduction of complementary foods will differ in different regions. For example, in SEA, the first lure is often an avocado, and in Russia an apple. In Russia closer to a year babies are offered sour-milk products, but in Asian countries they almost do not consume dairy products.

Because of these differences, we didn’t have a question whether to localize applications for countries where English is perfectly spoken: Sweden, Denmark, Iceland, Finland, the Netherlands … Plus, even if you are fluent in English, it’s still more comfortable to perceive important information in your native language.

Details can play a very big role. In the pregnancy application in the screenshots, a beautiful, neat naked belly is a standard image for a European audience, but for the Arab countries we decided to cover the pregnant tummy, nevertheless they have rather strict looks.

Our pictures for Europeans, for Arab and Eastern countries differ in style: for Korea and Japan, for example, these are nice pictures.

It is interesting that each nation has its own requests. The Germans, for example, have the most requests for features about co-parenting. Men are actively involved in caring for a child, changing diapers, bathing, walking with a stroller. And in Muslim countries there are such requests as “prayer when laying the baby.”

# 6: Localization in unpopular languages ​​brings a good income

Many applications translate only into 5-6 languages, calculated for maximum coverage. However, as I have already noted, it is more convenient to perceive important information in the native language.

Wachanga apps are available in languages ​​such as Greek, Serbian,
Romanian, Hungarian, Croatian, Malay. We approach this issue pragmatically: each market is a certain point of growth. Given that many developers ignore small markets, we can get a significant portion of clients planning a pregnancy or already raising children.

Therefore, our investments are justified. First of all, this is organic growth, both in the number of installations and in the volume of the active user base, including paying users. Over the past year, revenues from small unpopular markets have grown significantly.

The same applies to traditional Chinese. On the one hand, it is more logical to enter the Chinese market with simplified Chinese: this is the language option used in mainland China, i.e. the vast majority of Chinese. And traditional Chinese is popular in Taiwan, Hong Kong and Macau, and more than 20 million people live there. This, for a moment, is the population of all of Scandinavia, and the people of Hong Kong and Taiwan are more solvent than the people of mainland China. We have already localized our applications into traditional Chinese and Hong Kong dialect using Alconost.

# 7 Screenshots in stores: take into account national characteristics

Our designers draw people in the screenshots for a certain nationality: we try to adapt and like more users of a particular country and as a result – increase the conversion. In the product description, we also rely on national traditions.

Screenshots of the Babycare app in China and Brazil

We tried to take screenshots of different babies, adapted to the country: in Brazil – curly babies, a football field and a ball in the screenshot, in Asian countries – babies with the appropriate appearance, and also added one of the popular names of children for the last few years to such a screenshot. Agree, it’s nice to see in the application children with your usual appearance, and not typically European golden-haired and blue-eyed babies.

Emoji screenshots in Korea boosted conversion. We also concluded that Koreans need to make fonts fatter and call-to-action frames (call to action) – thicker.

Our results can be used as a guide, but you should definitely conduct A / B testing in the Google Play console and check how these ideas work for your audience.

# 8: Check CTA with Carrier

After you translated Call To Action (native speaker, I hope) and placed this text in the screenshots, again send the screenshots to the carriers for verification. The goal is to understand how a native speaker perceives a screenshot of a given CTA: does the call to action sound exactly there? The translator may accidentally translate this not “Download”, but “Download”, for example.

Especially often, such inaccuracies occur when the translation is from English, and you give the translator just a set of single words without context. In any case, the right call to action means a higher conversion in the installation, so I am in favor of double-checking these important marketing materials.

# 9: Localization Surprises

Not all developers are aware of local marketplaces: in Iran, for example, it is Bazaar, in India and China there are their own alternative sites. It is important to understand this so that it doesn’t turn out that you have released the application on Google Play, but no one downloads it, because this country has its own content.

Chinese and Japanese characters are insidious: there are no spaces between words, so if you wrap lines as you like, the meaning of the phrase may be distorted.

Did you know that the Arabs of Asia and Egypt do not use Arabic numbers (like us), but Indo-Arabic ones? They look like this:
٠ ١ ٢ ٣ ٤ ٥ ٦ ٧ ٨ ٩

When you first encounter localization in Arabic and Hebrew, you are shocked by the fact that the entire interface needs to be redone, because the text in RTL languages ​​is written from right to left. Imagine what happens when in one sentence you need to use both RTL and LTR languages ​​(i.e., text from right to left and left to right) ?!

We have a meme about the Arabic language: “But the Arabs have all the interfaces rotated, do they have the thumb on their right hand too RTL, or are they all left-handed?”

# 10: Work with reviews in all languages ​​of the world

Every day we receive hundreds of reviews in all languages ​​of the world. We translated the template answers into the main important languages, in critical cases we answer in English. And all the specialists of our Customer Success Department speak several languages.

In other cases, we resort to Google Translate, but be sure to run the translation from English to the desired language and back to English. If the meaning remains, the answer can be sent. In terms of reviews, it is more important for people to quickly solve their problem, so no one complains about the quality of the translation.

True, Koreans and Japanese respond in English: Google Translate does too poorly with such languages. And Indonesian turned out to be the most difficult language for Google Translate: sometimes machine translation from Indonesian does not even give an approximate idea of ​​what the user wrote to us.

Thanks to Maxim and the team Wachanga for valuable observations and advice and wish you success in further global expansion! And we at Alconost and Nitro are always happy to help you 🙂

Do you localize applications in several languages? If you want to translate short texts quickly like Wachanga, try Nitro. Create an account by this link, and your account will have $ 20 as a welcome bonus for the first transfer.

about the author

Article written in Alconost.

Alconost is engaged localization of games, applications and sites in 70+ languages. Native translators, linguistic testing, cloud platform with API, continuous localization, project managers 24/7, any format of string resources.

Nitro – Professional translation service online in 35+ languages. Created by Alconost specifically for the translation of small texts.

Native translators, order from 1 word, online chat with customer support. 50% of orders are ready within 2 hours, 96% – in less than 24 hours.

Similar Posts

Leave a Reply