According to Sensor Tower's Mobile Market Forecast 2021–2025 report, the world's largest mobile app markets are Asia, Europe, North America, Latin America, the Middle East, and Australia:
App Store Consumer Spending by Region
App Store Downloads by Region
Each region differs considerably from the others in terms of traditions, the locals' mentality and outlook on life, and even the legal system that affects the specifics of doing business in those regions. As part of marketing, app store optimization (ASO) must be executed with regional differences in mind. Its role is to adapt the text and graphic metadata to make it suitable for each target country.
In this article, a RadASO specialist explains how different regions' cultural and national peculiarities affect the process of creating text metadata for the App Store (Title, Subtitle, and Description – which is not indexed but affects conversion rates) and how to overcome the difficulties you may encounter during the process of app listing page localization.
Metadata localization for the Asian region
The key countries in the region in terms of downloads and profitability are China, Japan, and Korea, so this section will primarily focus on the characteristics of these nations.
1. Features of competition in the regional market
One of the phenomena we have observed when doing ASO for regional markets is that local developers hold a strong position in the mobile app market, so focus on them when localizing app listing pages. As an example, here is a list of the top apps by the number of installs in the baby tracker niche in the UK, US, and Japan:
UK
USA
Japan
As we can see, there are significant differences in the search rankings: in Western countries, global competitors predominate, while in Japan, only local ones do.
2. Localization of brand names in the countries of the region
Users in the Asian region prefer using local variants of well-known brand names. And in China, for example, the purity of the native language is actively supported by the government, particularly through legislation. Similarly, app names are also often localized. Two approaches can be taken for this: conveying the name's sound and the content.
When conveying the sound, the translator tries to preserve the phonetic identity – the way the consumer hears the brand name. In doing so, the content of the adapted name often changes. Below is an example of an adaptation of the name of the app Le Baby – breastfeeding, sleep for China:
The name’s sound remains the same, but the meaning, although changed, still sounds good.
And here is an example of a semantic adaptation of the TinyLog name in Chinese:
In Japan, the process of adapting names is complicated by the existence of three writing systems in the Japanese language: kanji, traditional characters of Chinese origin, and the two constituent alphabets, hiragana, and katakana. The latter is often used for transliterating foreign words into Japanese, including brand names. Moreover, all three writing systems can be present in the same Japanese sentence.
As you can see, two components of the alphabet were used at once to convey the sound of the name "PiyoLog."
It should be noted that the Asian region is very culturally diverse. So, the localization of app names can vary greatly from country to country. A detailed study of the local market and the assistance of a native speaker with a good understanding of their country's cultural context will help you adapt the brand name or app name correctly.
3. Lexical diversity of Eastern languages
Asia's regional languages have a rather rich vocabulary, which also significantly impacts the semantic core: there are many spellings of different words and concepts. To give an example, here are the different spellings of the word "baby" in Chinese and Korean:
Understanding which of the possible variants appears the most organic in the text is a difficult task, and only someone deeply immersed in the respective linguistic culture will be up to the task. We recommend working with native speakers when preparing textual metadata for Asian languages. Even a professional translator's knowledge and skills may not be sufficient to correctly localize metadata for this region (to preserve slight nuances in the meaning and tone of the writing).
4. Peculiarities of punctuation in Eastern languages
An important feature of languages in the Asian region is the punctuation system, which Europeans tend to consider unusual.
One of the most significant characteristics affecting the appearance of textual metadata and application rankings for search queries included in the metadata is the special rules for using spaces in Eastern languages. In traditional Chinese and Japanese scripts, spaces are not used, and words in a sentence are written consecutively, separated by punctuation only when necessary. For example, this is what the name of one of the parcel tracking apps on the App Store page looks like:
And this is what the title would look like broken down into individual words:
Interestingly, by including each of these individual "words" in the text metadata of the App Store listing, the app will also rank for the phrase written together. For example, to get indexed for the above search term "菜鸟-快递轻松查寄取" you need to add individual keywords for this search term in the keywords field: 菜鸟,快递,轻松,查,寄,取. This feature makes text metadata and texts in Chinese and Japanese generally capacious and rich in keywords. We recommend not adding too many of them to the Title and Subtitle fields of Asian locales to avoid being flagged by the App Store for keyword stuffing.
The situation with Thai is slightly different. In this language, spaces are only used to separate sentences, and there are no spaces between words within a sentence. For example, the Thai page of one of the parcel tracker apps looks like this:
Let's break down the above subtitle into individual words:
But unlike hieroglyphic languages, adding all the individual words in a search term to the metadata will probably not help rank it for a phrase that is written together. That is, to rank for a search term such as "เช็คพัสดุ" ("check the parcel"), it would have to be added to the metadata in its entirety. Fortunately, most search queries are single sentences, which can be written with a space in the Title and Subtitle or a comma in the Keywords. The following application will rank for two relevant queries – "ติดตามพัสดุ" ("parcel tracking") and "เช็คเลขพัสด" (check parcel number) as independent sentences:
The differences in punctuation between European and Asian languages are not only nuanced in their use of spaces. The region’s local languages employ their own system of punctuation marks and rules for their use. As an example, here is a snippet of the description of the breastfeeding tracker app:
We see the use of specific Chinese and Japanese characters in the text: the comma "," and the Chinese dot "。". At the same time, the text also seems to contain the familiar "European" commas and dashes – but this is not quite the case. In Chinese and Japanese writing, the width of punctuation marks must match the width of the character, so punctuation marks already contain a "built-in" space, creating the necessary indentation between two characters positioned next to each other. In this case, both the punctuation mark and the "embedded" space are counted as one character. At RadASO, we call these characters "super symbols". Here is a short list of them:
Chinese dot "."
Chinese comma "，";
drop–shaped comma ",";
colon "："
dash "–";
semicolon "；"
exclamation mark "!"
question mark "?"
interjection "・".
You can use them to save space in Chinese and Japanese locales. These characters can also be used in other locales, but we have noticed that some of them "glue" the shared words together. For example, when a Chinese comma is used in this sentence structure, "baby, milestone tracker", the app listing will likely rank for the search term "babymilestone tracker". As for queries with "baby" and "milestone", the app will not be ranked. Read more about using "super characters" in RadASO's Telegram channel:
The Asian region's languages also have other punctuation marks, some of which are not mentioned in this article: specific varieties of inverted commas, brackets (in Japanese, for example, there are 14 varieties), characters that repeat hieroglyphs, etc. Only a native speaker, a person deeply immersed in the linguistic and cultural context of the respective country, can truly understand the intricacies of Asian languages.
5. Worldview distinctive features of the region
When working with textual metadata, differences between diverse cultures must be kept in mind, and the inhabitants of the Asian region are no exception in this regard. Whereas the countries of Europe, North America, and Australia are characterized by the use of clear and concise language and individualism, the written language of Asian people has a more respectful manner of communication and an emphasis on human interaction. In addition, texts are often lusher and richer, not as succinct as in the Western world. The difference between Western and Eastern approaches to textual writing (and textual metadata in particular) is most clearly demonstrated by comparing American and Japanese versions of app listing pages:
Translation of Title and Subtitle of the Japanese version of the page:
The richer style of Title and Subtitle texts in Japanese is partly due to the capacious nature of Japanese writing and the ability to fit more information into the 30-character limit. However, the different writing tone when addressing the audience is also quite evident. A comparison of the descriptions of another app, My Baby – Newborn Tracker, on the American and Japanese versions of the app listing page, demonstrates the same tendencies:
To conclude our analysis of the Asian region, let's briefly touch on the issue of regional differences in Chinese. It is well known that there are two Chinese locales in the App Store: Chinese (Simplified), valid in China, Singapore, and the US, and Chinese (Traditional) for Hong Kong, Taiwan, and Macau. All of these regions have significant political, economic, and historical differences. When localizing the app listing page to Chinese, focus on the semantic core, competitors, and cultural peculiarities of your highest-priority country. Below is an example of how optimization of the Breastfeeding Newborn trackerapp's metadata differs between China (Chinese [Simplified]) and Taiwan (Chinese [Traditional]):
Features of Western locales: European, American, and Australian regions
These regions have many historical ties and are consequently very similar culturally. Let's look at their main particularities.
1. The impact of lexical differences between languages on metadata formation
In describing the characteristics of Asian countries, we mentioned that countries in this region tend to have richer writing styles. However, the cultural and linguistic differences between Eastern and Western countries are not the main reason why textual metadata in Western-localized app listing pages look dry and laconic compared to their Eastern counterparts. The main reason is the typical length of words in European languages. If you optimize your app page for European locales, you will not be able to fit a large number of search terms into the metadata (remember that the Title and Subtitle fields are limited to 30 characters each). As an example, let's look at Room Planner – Home Design 3D, optimized for Italian and Chinese (Traditional) locales:
Even though the Title and Subtitle in the Italian version are longer (27 and 28 characters, respectively) than in the Chinese version (19 and 24 characters), the latter has an advantage in terms of the information you can fit in and keyword density. Below are the Title translations of the Italian and Chinese (Traditional) versions:
The biggest problems with metadata keywords lie in optimizing the app listing page for the locales of Romance languages, including French, Spanish, Italian and Portuguese, as well as some Germanic languages, such as German and Dutch. English, on the other hand, is one of the shortest and most suitable for keyword optimization among all Western languages. The Alconost company has compiled a statistical table that shows how the text length can vary when the same information is conveyed in different languages. Here is an excerpt from the table:
Source language
Target language
How much longer (+) or shorter (–) the text in the target language is
English
French
21.18%
English
Spanish
19.52%
English
Italian
17.91%
English
Deutsch
16.67%
English
Dutch
13.80%
English
Portuguese (Portugal)
14.29%
English
Portuguese (Brazil)
12.96%
English
Polish
9.33%
English
Russian
9.11%
English
Czech
3.70%
English
Arab
–6.25%
English
Japanese
–39.68%
English
Korean
–44.04%
English
Chinese (Simplified)
–61.97%
English
Chinese (Traditional)
–63.80%
2. Using local and English semantics with a narrow semantic core
Europe comprises many countries, including small nations in terms of population size: Scandinavian, Baltic, and Benelux countries, as well as a large number of Central European ones. When collecting the semantic core for these countries, we face the problem of not having enough semantics in the local languages to populate the metadata. A possible solution is to use more semantics in English, but before opting for this approach, we should carefully consider the different penetration levels of English in each country's culture. Wikipedia provides a map of English proficiency in Europe:
As we can see, the highest number of people speaking English is recorded in Scandinavia, as well as in the countries whose languages are related to English, namely: Germany, Austria, and the Netherlands. In Eastern and Southern Europe, the permeation of English is weaker.
Let's see how the permeation of English influences the textual metadata in the example of Clockmaker: Mystery Match 3. We present excerpts of the app listing page for the English (UK) locale in all European countries, as well as for the locales of the languages of several sparsely populated countries where the semantic core is also small.
The Title field of the Danish and Dutch locales, where English is strong, is saturated with frequently used elements of English semantics: logic, puzzle, and jewel. Remember, keywords specified in this field have the highest value. Moreover, in the Dutch locale, the Subtitle field consists exclusively of English keywords:
At the same time, the Title of the Hungarian and Slovak locales consists of local low-frequency semantics. The exception is the keyword "match 3" in the Slovak locale, but this is already covered in the English (UK) locale. Meanwhile, the Subtitle of the Hungarian locale is entirely in Hungarian, and the Slovak locale comprises only English words.
In a number of apps, if there is a lack of semantics, the Title and Subtitle fields are populated with English keywords where the locally spoken language would normally be used. The text in the Description box is localized to the local language (as we mentioned earlier, in the App Store, the Description field is not indexed):
When considering this approach, remember that while we make gains in the keyword optimization department (more frequent English semantics in Title and Subtitle fields), at the same time, we risk inadvertently reducing conversion rates in the process (e.g., lower number of app installs), as most users prefer to consume content in their native language.
Each country has a variety of national and niche specifics regarding the use of English-language vocabulary. Only a detailed analysis of the semantic core and competitors will help determine those. For example, in Germany, the national variant for the Match-3 gameplay game "3 gewinnt" is also widely used at the same time as the search term "match 3":
Meanwhile, in France, where English proficiency is lower than in Germany and where the population is often reluctant to speak English (influenced by the age-old rivalry between the two states), the English-language term "match 3" is used as the main term. In the example below, the national variant of the gameplay name "séries de 3" is used in the Description field, which is not indexed in the App Store, while the English variant is used in the Title:
3. Regional differences between English, French, Portuguese and Spanish
Another feature of the European, American and Australian regions is the wide geographical distribution of the main languages: English, French, Portuguese and Spanish. The App Store has even created separate locales for different variants of these popular languages, such as English (UK), English (US), English (Canada), and English (Australia). See our table for more details on country coverage. As a result, regional versions of these languages can differ significantly in terms of vocabulary, speech patterns, and even grammar rules. For example, despite the shared historical past of the US and the UK, the national versions of English in these countries differ significantly from a grammatical point of view:
US words ending in "-or" become "-our" in the UK: "colour," "favour," "flavour" or "behavior";
Certain US words ending in "-er" become "-re": "centre," "theatre" in the UK;
In American English, notice the doubling of the letter "l" in many words: "skillful," "fulfill," "enroll" ("skilful," "fulfil," "enrol" in British English).
In addition, the vocabulary in the two speech variants is also different. Below is a comparison of the popularity of the search terms (SAP) "truck games"/"lorry games" and "jail games"/"prison games" in the US and the UK:
Search term
Translation
SAP* English (U.S.)
SAP* English (U.K.)
truck games
games with trucks
62
39
lorry games
games with trucks
–
32
jail games
prison games
29
8
prison games
prison games
32
24
*Search Ads Popularity (SAP) – shows the popularity of the search term from 5 to 99.
As you can see, the term "lorry games" is used exclusively in British English, while "jail games" is typical in American English. Sometimes the same expression can mean different things in different regional variants of the language. Let's look at the search results for the search term "coin price" in the US and the UK. Note that the search term's popularity value (SAP) in the two countries is not much different: 8 in the US and 5 in the UK.
As we can see, in the US, the search term is primarily used to describe numismatics, and only one app in the top 10 is related to the cryptocurrency niche – CoinMarketCap: Crypto Tracker. In the UK, the opposite is true: the search term results almost entirely consist of cryptocurrency apps. The only exception is PCGS CoinFacts Coin Collecting, which is ranked number one.
Significant regional differences are also characteristic of other common world languages. Here is a comparison of the popularity of the search terms (SAP) "soldes" and "aubainerie" in France and Canada, which illustrates the lexical dissimilarity of the Canadian and French versions of the French language:
Search term
Translation
SAP* French
SAP* French (Canada)
soldes
sale
25
5
aubainerie
sale
–
14
*Search Ads Popularity (SAP) – shows the popularity of the search term from 5 to 99.
You can read more about the regional characteristics of French and English in Canada and the peculiarities of app localization for the Canadian market in this Alconost article.
Let's now demonstrate the impact of lexical differences on app listing pages in the store. As an example, let's compare the text metadata of American and Australian English locales for the top three apps in the Australian market in the social casino niche (simulated games you can find in casino halls, but where you can't win real money):
As we can see, in Australia, the term "pokies" is more commonly used to refer to slot machines than its American counterpart "slots." The conclusion here is simple: a semantic core collected for one regional version of a language may be completely irrelevant for another. Always carefully collect and analyze the semantics for each of the countries you are localizing for, no matter how similar the language and culture may seem at first glance.
Localization patterns in the Middle East
The Middle East is an interesting region from a business perspective, with one of the highest average revenue per user (ARPU) rates in the world. In addition, according to the already mentioned Mobile Market Forecast 2021–2025 report by Sensor Tower, the region is growing rapidly; by 2025, it will have reached a growth rate of 208% in terms of revenue compared to 2020:
App Store Consumer Spending by Region
The Middle East lies between and is culturally similar to the European and Asian regions. Let's focus on the cultural characteristics of this region that affect App Store metadata.
1. Regional differences in the Arabic language
Similar to the Western world, the Middle East region has a dominant language: Arabic. It is also widely spoken outside the Middle East, in North Africa, and has many regional differences:
The App Store's Arabic locale also applies to roughly the same countries (see our table for more details on localization coverage). The good news is that when working with textual information, including metadata, for any of the countries with a valid Arabic locale, we use Modern Standard Arabic (MSA) rather than local Arabic dialects. Consequently, the App Store only supports a single locale for Arabic as well. At the same time, the process of collecting semantics in Arabic is similar to that of English, Spanish, French, and other languages with a wide geographical spread; the semantic core is collected for the most important country where you are looking to promote the app. For example, a semantic core collected for Egypt would be ineffective in optimizing an app listing page if you are targeting a country such as Saudi Arabia to do business in.
2. Localization of app names and brand names in Arabic
The need to localize app names and brands in Arabic depends on the niche. For example, the consensus is that game titles are usually left written in English. We checked this with the Match-3 gameplay niche in Saudi Arabia. Of the top 10 apps by the number of installs, only four have a localized app listing page in Arabic. Below are excerpts from these apps' pages with the corresponding Title and Subtitle:
Indeed, in all the above examples, the apps' names were left written in English. But the situation was different when we analyzed the child activity tracker niche. Of the top three apps by the number of downloads, two had a name completely localized in Arabic and one had a name partially localized:
Long story short, before you localize the app listing into Arabic, you need to conduct extensive research on the niche and your competitors. There is no single rule for localizing app names in Arabic.
This section concludes with another important takeaway: in Arabic and Hebrew, the writing direction is from right to left. On the other hand, numbers and parts written in Latin are read from left to right. In one of the examples above, the text in the Title reads like this:
However, titles of this mixed type may not display correctly in the App Store search: the part written in Latin will be displayed in its entirety regardless of its location in the Title, while the same might not be the case for the Arabic part. The text will either be displayed entirely in Latin or Arabic.
3. Understanding the rich vocabulary of Arabic
Arabic is a lexically rich language. There are different spellings of the same terms and names:
We encountered a similar situation when analyzing the languages of the Asian region.
In Arabic, the problem is solved in the same way – by working with a native speaker when localizing the text. Only a native speaker can help you choose the words that most seamlessly match the textual metadata.
4. Wording to be avoided in Arabic text metadata
There are many topics to avoid in Arabic culture: overly explicit depictions of human relationships, and topics related to alcohol, drugs, violence, mysticism, and gambling. When localizing app listing pages for the Arabic market, a native speaker will help determine the appropriateness of using certain wording in the metadata. Previously, we mentioned that when localizing games, the names are usually left in English. Below are some examples of games whose titles have been localized in Arabic to avoid ambiguous wording in the Title:
Sometimes the topics to be avoided are not obvious to people with a Western outlook. For example, completely harmless topics such as psychology, psychotherapy, and mental health practices are still taboo in Arab culture. Mental health is not spoken about, and the psychologist's role is assigned to a religious figure or a doctor. Below are examples of how localization in Arabic has replaced taboos in Arab culture with "mental health", "coaching", "cognitive behavioral therapy (CBT)" and even "meditation":
The world is gradually becoming more globalized and the number of taboo topics in the Arab world has decreased over time. But adapting the textual and graphic metadata to the Arab countries' cultural sensitivities is still the key to successfully promoting apps in their markets.
The takeaways
Some important points to keep in mind when working on textual metadata (recommendations are equally valid for the App Store and Google Play):
When starting out, look at how competing apps have structured their metadata. Even something as quick as a mini-study like this will help reveal important cultural nuances of localization in the target regions.
The differences in culture, mentality, and even the legal system of different countries are enormous. Only a native speaker immersed in the cultural environment of their own country can take all these peculiarities into account.
Be careful when using English semantics in visible text metadata in non-English-speaking countries. Competitor research will tell you how suitable and appropriate your approach is.
The regional differences between the world's most commonly spoken languages – English, French, Spanish, Portuguese, Chinese, and Arabic – are quite substantial. When compiling metadata, use the semantic core gathered for the country that is your target language. Keep in mind that the App Store has multiple locales for many of these languages (see our table where we have compiled data on country locale coverage).
When collecting metadata for different countries, use popular local search terms in local languages: the lower competition in the rankings increases the chances of getting good positions for these search terms.
And, of course, remember that a well-localized app listing page is not just about well-compiled text metadata. Only comprehensive effort on textual and graphical metadata will help your app listing secure a good position in search results, have a high conversion rate, and have an excellent financial outcome.
15
6
5
4