Five Golden Rules to Achieve Agile Localization

Illustrations by Julia Feng (

Agile_Loc_LogoOver the past decade, many software development teams have switched their development methodology from a waterfall model to something much more agile such as Scrum. Through this transition, their expectations towards other teams such as Localization have changed and these teams have had to improve their agility too.

At Adobe, we have a centralized Localization group that currently supports 135 product and functional teams. Most of these teams have adopted some form of agile development methodologies and have reduced their development cycle from 18-24 months to yearly, quarterly, monthly and, these days, bi-weekly releases. A couple of Adobe product teams and other companies are even releasing updated versions of their product multiple times per day, making it imperative for Localization to keep improving its agility.

Drawn from our experience, this article presents Five Golden Rules that need to be satisfied in order to achieve optimal agile localization.

Rule 1 – “We are one Team!”

Within many companies, Adobe included, Localization is a centralized function serving all product and functional (e.g. Marketing, Sales, Legal, etc.) teams. This structure makes sense because localization is a specialized field, therefore resources (people, tools) and processes can be leveraged across the company. Nonetheless, localization should still be everyone’s concern. The Localization team can come up with many solutions, but the best ones originate when there is a true partnership between the product/functional and localization teams.

We're one team!

We’re one team!

The most agile teams treat Localization staff as if it were part of their own team.

A core aspect of Scrum is to include all skill sets, including localization, required to deliver a product to users. Therefore, Localization team members must be included in all development aspects – from backlog review to retrospective – so they could plan and address international issues early on. Strong partnerships also need to be established with localization vendors when companies, such as Adobe, engage with partners and vendors for their translation and testing activities.

Customer engagement is a key aspect of agile methodologies as it validates the quality and usefulness of the work performed thus far. We recommend engaging with international customers too, because their issues increase awareness around internationalization.

In summary, all stakeholders (development teams, functional teams including localization, vendors and customers) need to collaborate closely in order to achieve great agility.

Rule 2 – “Internationalization is King”

In our Globalization Myth Series, we defined Internationalization (commonly abbreviated as i18n) as an engineering exercise focused on generalizing a product so that it can handle multiple languages, scripts and cultural conventions (currency, sorting rules, number, date and time formats…) without the need for redesign.

Internationalization is King

In other words, the better internationalized an application is, the easier it will be to localize.

In the waterfall model, teams could possibly work around some of the internationalization deficiencies because of longer development cycles. Unfortunately, in the agile world, there is not enough time to look for work-around solutions anymore. The code needs to be internationalized from the get-go.

There are various approaches to improve internationalization in a company, which includes the following:

  • I18n_EducationEducation: Training core developers is an effective way to reduce the number of internationalization issues in a product. By exposing engineers to localization and internationalization issues, they gain a broader perspective on the impact of their code and avoid some of the classic internationalization pitfalls.


  • I18n_LibrariesInternationalization Libraries: Leveraging Open Source internationalization libraries, such as ICU or JavaScript i18n, is another good practice. Instead of re-inventing the wheel, engineers can reuse code that has already been validated by others. Also, i18n libraries usually support 100+ locales, which require a significant amount of research and development time. It will be hard for a single team or even company to support so many locales.
  • Peer_ReviewCode review: Practicing peer reviews is an effective method to reduce internationalization defects in a product. Since developers know their code will be reviewed, they pay more attention to its quality (peer pressure effect) so it also benefits internationalization. Some companies even automate this review process using tools such as Globalyzer.
  • Globalization Report Card: Benchmarking products against an ideal architecture helps to improve internationalization too. As part of our World-Readiness program, we created a Globalization Report Card system to assess the degree of world-readiness in each Adobe product. This scorecard measures products against a set of internationalization criteria (ability to input international characters, display date formats, translate the user interface, and so on…). It is an efficient way to track progress made by each team over time and can even create some healthy competition among product teams. These teams are motivated to be on top of the i18n charts!

Globalization Report Card

Rule 3 – “Integrate Localization into the Development Process”

To release a new product, development teams have many high-priority tasks and usually prefer not to have to worry about localization until necessary. As a consequence, localizability issues are often discovered too late and encounter the risk of being deferred to a future release. Product teams don’t always anticipate the impact of a particular task/feature on localization, and quite often, the Localization team isn’t able to influence design or development until the feature is already implemented.

In an agile process, features and development tasks are tracked in a backlog and reviewed at the beginning of every sprint. To eliminate the side effects of the “throw-over-the-wall” model described above, it is critical to include Localization representatives during these sprint-planning meetings so more visibility and importance are given to the localization tasks. This also provides great educational value to all stakeholders who can then understand the impact of their decisions on the localization process. Localization or a proxy should also attend the daily sprint meetings to keep up with the development pace and decisions. By attending these meetings, Localization team members can be much more proactive and influent.

Localization Process Integration

Adobe Revel and are examples of teams that integrate localization into their development process. They also prioritize localization intensive features/tasks upfront – carving enough time for the Localization team to run its process and deliver high-quality localized releases.

In a recent Localization World event, Amrit Singh (International Program Manager for our Installation technologies) presented LocBan (Kanban applied to Localization). Just like in a Toyota factory, the Localization team maintains a board of “To Do”, “Work In Progress” and “Done” tasks which provides great visibility on the localization “conveyor belt”. Similarly, it would be beneficial to maintain Kanban boards for each translator. In the waterfall model, translators used to receive large localization kits, which they had to scramble to complete within the deadline. In the agile world, translators are now able to “pull” work as their bandwidth opens up.

Localization Kanban

By using an integrated Kanban board, everyone has a clear understanding of all the various dependencies and accountabilities, resulting in stronger collaboration and higher success rate.

Rule 4 – “Reduce, Reuse and Recycle”

Localization can generate a lot of waste if not planned properly. So, it is key to become “green” in order to become more “agile”.

Reduce, Reuse, Recycle


It is clear that reducing the localization effort will have a positive impact on a team’s agility. This could be achieved in 2 ways: by validating the localization scope and by reducing the translation waste generated during the localization process.

  • Reducing Localization Scope

The Localization Manager’s job is to ensure the company localizes the right product and content into the right language set. At Adobe, we have had situations in which we were localizing too much content. For example, using Adobe’s Digital Marketing Suite, we discovered that Russian customers prefer reading Development documentation (such as API descriptions) in English rather than in Russian. We were able to save a lot of time and cost by removing this component from our localization requirements.

Similarly, through market research, we discovered that most Middle-Eastern Creative Suite customers prefer to use an English user interface with Arabic or Hebrew documentation. This combination makes English content such as videos and tutorials more accessible to them.

In short, tracking web analytics and engaging with customers, power users, pre-release testers and geos constitute a great way to validate the localization requirements and improve agility.

  • Reducing Localization Waste

Once the localization requirements are confirmed, it is key to limit the translation waste generated during the localization process. This obviously impacts the translators’ work but also the bandwidth of the localization staff.

Sweep Away Waste!An effective way to reduce localization waste is by understanding its root cause. At Adobe, we categorize all localization defects through a common set of keywords, which provides us with a good picture of the issues faced across products. We can then develop solutions to reduce, if not eliminate, these defects.

Localization waste sometimes originates from English strings -assuming English is the source language. Indeed, translations created before English strings get finalized will need to be revisited and will likely generate some waste.

In the agile world, we can’t afford that extra time, so it is important to validate the English content before handing it off to the translators. Doing something as simple as spell checking can help to reduce a lot of localization waste. In a product such as InDesign, about 3% of the English user interface strings are updated once they get reviewed for spelling and grammatical mistakes. For a product that is localized into 25 languages, this represents a waste equivalent to 75% of a single language scope!

Also, many of the software localization testing activities are necessary because localization is happening out of context. Solving that problem can tremendously speed up the localization process. In an ideal world, localization should be a product feature that allows translators to translate the user interface in-context. Facebook did a great job in this area by enabling translators (in this case its user community) to translate and provide feedback within the application itself. Alternatively, translators should be provided context information through builds, screenshots or meta-data information (e.g. developer comments, feature name, expected delivery time, etc.).

To reduce waste, it is also recommended that localizers develop glossaries, style guides and tools that leverage previous localizations.

Ultimately, it’s critical for translators to validate their work as they translate. That way, activities down the production line can be eliminated or reduced, which makes the entire process more agile.


Reuse when it makes sense!Reusing strings can sometimes be a source of challenging defects in software localization, so it has to be handled carefully. For example, the English string “none” could be translated as “aucun” or “aucune” in French based on the gender of the noun to which it refers.

That said, reusing strings – in the same context – could also help to improve agility, since these strings won’t need to be translated multiple times.

An area where Adobe has experienced positive results with reducing and reusing English content is in our instructional content. In documentation, Adobe relies on Acrolinx to control the quality of the English (source) content. Authors need to use a certain authoring style (e.g. shorter sentences) and are encouraged to leverage existing paragraphs (e.g. legal disclaimers). This improves consistency in the English documentation and has the great benefit of reducing the localization workload too.

Similarly, DITA (read Reduce, Reuse and Recycle: Developing a Reuse Strategy for DITA) and Content Management Systems such as Adobe Experience Manager (formerly known as Day CQ) are designed to reuse/share content across multiple channels and publications.


Recycling is the process of transforming existing materials (or waste) such that they could be reused again – sometimes for a totally different purpose. Creating polar fleeces from used plastic bottles or isolating walls using old denim jeans are classic examples of recycling.

Such transformations can apply to translations too. Translators don’t need to translate every sentence from scratch. Translation technologies such as Translation Memories and Machine Translation engines can help translators recycle previous translations and speed up the translation process. At Adobe, we have experienced dramatic productivity gains when we used these technologies. In general, a translator supported by these technologies will deliver in an hour what other translators would deliver in a day. These are impressive gains that contribute to localization agility too.

Rule 5 – Automate, Automate, Automate

The last requirement to achieve agile localization is automation. With agile, you can’t afford to send translation requests through e-mails or cut and paste strings from a spreadsheet to a source file. All translation hand-offs should be automated and managed through a centralized system. Over the years, Adobe’s Globalization team has developed such a platform. This system is able to connect with various source control systems, manage translation jobs, leverage existing translations across projects and content types and provide machine translation engines. In the Globalization Myth 4 article, Guta Ribeiro introduced Airport, our new system to automatically connect with our vendors and help us march towards our lofty one-hour translation goal.


Beyond translation, it’s also important to automate other aspects of the localization process, such as build, quality assurance, bug fixing, screenshots and distribution of the localized releases. We can only go as fast as our slowest component, which is why it’s critical to automate all aspects of the localization process.


Localization agility can be achieved as long as all stakeholders work as a unified team. It is critical for core engineers to develop well-internationalized code from the get-go. This can be achieved via training, code reviews, usage of (Open Source) internationalization libraries and globalization report cards.

The localization process should be fully integrated within the overall development process so that all dependencies and accountabilities are clear. We recommend using Kanban boards (via tools such as Trello) to raise this visibility. To become agile, it’s also important to act “green” (i.e. reduce, reuse and recycle). Doing so represents an effective way to control waste generation before and during the localization process. Finally, all efforts should be made to automate all parts of the streamlined localization process.

At Adobe, not all localization projects are handled with great agility – yet! Some projects are more agile than others. However, based on our experience, we believe agility can be achieved by adopting these Five Golden Rules:

  1. “We are One Team!”
  2. “Internationalization is King”
  3. “Integrate Localization into the Development Process”
  4. “Reduce, Reuse and Recycle”
  5. “Automate, Automate, Automate”


Special thanks to Rob Jaworski, Amrit Pal Singh, Ashish Saxena, Janice Campbell, Leandro Reis, Peter Green, Julia Feng and Quynn Le for their invaluable feedback on this article.

Globalization Myth Series – Myth 3: Language support should be limited to the official language(s) of a country

This article was originally written in English. Text in other languages is provided via machine translation.

Problem statement

In the globalization field, we often observe cases where companies mix up the usage of countries and languages when delivering products to their global customers. Assumptions are often made that people living in a country should receive services in that country’s most common language(s) – such as – Japanese in Japan, English in the United States or Russian in Russia. Good companies will even support multilingual situations like Belgium and Switzerland, which support multiple official languages – but is it really enough?

In this blog article, we argue that the concepts of “Country” and “Language” should be treated apart – even though they are related.

Languages have no borders

Historically, languages have spread around the world through immigration, invasions and colonization. This is why Spanish is spoken in South America, English in the United States, French in Senegal and Canada – just to name a few examples.

Languages haven’t been limited to their homelands in a long time. For example, Spanish and Portuguese have spread to the point where most of their native speakers are found outside the Iberian Peninsula. Only 8.95% of Spanish speakers are actually Spaniards [1]. We could argue that there are different flavors of Spanish, Portuguese or French, but, in essence, these languages are the same and they are not confined by country boundaries.

Twitter is a great tool to visualize languages spoken around the world. Eric Fisher created a cool language map using Twitter data [2], which clearly shows borders being redefined. Catalan appears on the map and multilingual countries like Belgium disappear under France and The Netherlands.

Map of languages as used with Twitter

Figure 1: European Language Map – Author: Eric Fischer

Official languages leave some major gaps

Many companies launch products/services in a country and limit localization and features to the official language(s) of that country. In software, a typical example is to create a new product for the American market and limit functionalities to English-speaking users’ needs. But this is usually not enough – even if the market is limited to the United States.

Recent U.S. Census data shows that 60 million (or 21%) of Americans aged 5+ years old speak another language than English at home. Spanish takes a big part of the pie with 25 million, but languages like Chinese, Tagalog, French, Vietnamese, German and Korean are each well above the 1 million speakers mark [5]. This is very understandable considering the immigration history of the country.

In addition, the U.S. welcomed 59.7 million international visitors in 2010 [6]. Of those, 33.4 million were from Canada and Mexico and 26.4 million from outside North America. Knowing that these foreigners spend on average $4,000 during their stay, this market segment shouldn’t be ignored.

Using the Greenberg’s index, which measures the probability that 2 persons in a country would speak the same language, The Economist shows that language diversity is happening in all countries around the world and, especially, in nations other than the United States, Russia and China [7]. Also, is an excellent source of data around language adoptions.

This diversity is also visible in Adobe’s analytics. Thanks to the Adobe Digital Marketing Suite and our Dynamic Language Delivery technology, we are able to capture the languages in which our worldwide customers would like to receive products and/or documentation. Figure 2 shows the language requests for Adobe Nav in the United States. Data collected with this small application basically confirms the diversity of the population in the U.S since languages such as Spanish, Arabic, British English, Korean, Simplified Chinese, Traditional Chinese, Dutch, Russian and Turkish are all requested by users located in the United States.

Languages expected by DLD users

Figure 2 – Unsatisified language requests for Adobe Nav in the United States

Language diversity in any country is the chief reason why software products should be developed with a global audience in mind. Even if the user interface or documentation is left in English, desktop, web and mobile applications should at least be able to input, display and print characters for most languages.

Similarly, we have been able to capture data about the preferred languages on This data confirms that countries and languages are two separated concepts – and while most customers from a country read the content in an official language, a big chunk of visitors still prefer to read the content in a different language. For example, with the Action Script Reference (ASR) guide, we noticed that, in Russia, only 68% of the Flash developers prefer reading the documentation in Russian versus 31% in English (a non-official language) and 1% in other languages. Similarly, numerous Flash developers outside of Russia choose to read the ASR documentation in Russian (See Figure 3).

Distribution of Russian Action Script Readers around the world

Figure 3 – Country distribution of Action Script Reference read in Russian language

This is another example why assumptions between countries and languages shouldn’t be made.

As people continue to become more mobile, it is critical to separate the concepts of country and language. This mobility is what spread languages beyond borders and creates the need to have languages supported everywhere.

A language is a customer preference, which can’t be constrained by country borders.

Consequently, if languages don’t have borders, flags shouldn’t be used to represent them as often seen on web-sites. Flags are meant to represent political entities [3]. A common example is to represent the English language with the British “Union Jack”. But it’s not culturally sensitive for English speakers from Australia, India or United States. Some have addressed the problem by mixing multiple countries’ flags into one although it’s making the situation even more confusing. The flag below (Figure 4) represents the German language (spoken in Germany, Austria and Switzerland) and we can imagine how convoluted the Spanish language flag would look like since Spanish is the official language in 14 countries [4] and spoken in many more!

Using flags to represent languages can become complex

Figure 4 – Flags and Languages (WikiMedia)


In summary, we offer 3 take-aways to consider when launching a product globally:

  1. Differentiate the concepts of “country” and “language”. As such, don’t use flags to represent languages.
  2. Don’t limit support to the official language(s) of a country/region. Some languages may need to be supported even though they don’t have official status.
  3. Languages don’t have borders and need to be supported globally. Capture users’ preferred language(s) so you can serve them in their language independently of their location.