QA InfoTech » Blogs » Why take on localization functional testing when you’ve already done pseudo localization testing?
Why take on localization functional testing when you’ve already done pseudo localization testing?
12 Jan, 2010

Why take on localization functional testing when you’ve already done pseudo localization testing?

This was an interesting discussion I had with one of my customers recently. His positioning was for the engineering team to take on pseudo localization testing and for the linguistic experts to take on linguistic testing. While there is value to this argument from an optimization and ROI grounds, I want to talk in this blog about the need for the right amount of pseudo loc, loc functional and linguistic testing, at the right stages in the product life cycle.

loc-jp1-300x227

Pseudo loc testing is a proactive step taken on before product localization primarily to:

1.        Detect under or over localization issues

2.       Find UI issues related to truncated strings, improper display of characters

3.       Perform one round of functionality testing to ensure product works as expected in a non English build

The important point to note above is the 3rd one. Why take on functionality testing post localization for every single locale if it has already been done in the pseudo localization phase? One must remember that though one goal of testing in the pseudo loc phase is to find as many issues as possible upfront, another important goal is to validate that the base English product has been designed well to support localization. This assures that the product can now be taken across multiple locales and not that the product will work fine on every locale. Think of this as prototype testing for a series of buildings to be constructed. The prototype testing is going to help you catch possible issues upfront and bring down the cost of fixing them later on. One could avoid prototype testing and actually decide to audit each individual building. If there are fundamental issues detected at that stage, it only amplifies the magnitude of the issue and fix so much more, impacting your core 3 project variables viz. cost, time, quality.

Ok, you may ask, yes, I get “why pseudo loc testing“, but now “why loc functional testing as well“?

Pseudo loc testing has only helped us identify and fix issues that existed in the product base that would have affected all locales. It has not helped us identify locale specific issues. For e.g.

1.       How does the product function when the target language is Double byte character set (DBCS) or Multi byte character set (MBCS) or bi-di?

2.       Are data inputs received and stored correctly, in the database, for the locale under consideration?

3.       Is data correctly retrieved on to the front end and being used for correct functional flow E2E?

4.       Are locale specific product customizations working fine?

Since every locale has its own character encoding complexities, doing one round of pseudo loc testing is not a replacement for loc functional testing. The place where I agree with my customer is that, one should closely look at an optimal coverage when it comes to loc functional testing.  Running all tests again is not going to give you additional bugs. It is only going to delay the loc sign off…Pick the right test cases centered around data inputs, data storage, any locale specific customizations, one E2E flow etc. to complete your pass. Discuss your test coverage with the larger product team (including program managers, developers) to ensure ROI from the loc functional testing effort. This will help them see the importance of this task and slowly start understanding the coverage and plan for these sanity checks even during the development phase. By doing this, you have right away educated the entire product team on localization engineering best practices and helped push quality up stream.

While pseudo loc and loc functional testing is normally done by the mainstream engineering team, it definitely helps to have linguistic experts for linguistic/translation testing. Translation tools will help you validate the translated text only at a high level. The contextual meaning and geopolitical issues if any, can be validated only by linguistic experts. In order to create a product with high end user acceptance and avoid any potential legal issues down the line, an upfront investment in linguistic testing by in country experts, is totally worth every $ you spend on it.

Hopefully, this has thrown light on the importance of all 3 test efforts (pseudo loc, loc functional and linguistic testing) with the right amount of optimization to ensure a smooth and successful global release of a product.

About the Author

Avatar QA InfoTech
Established in 2003, with less than five testing experts, QA InfoTech has grown leaps and bounds with three QA Centers of Excellence globally; two of which are located in the hub of IT activity in India, Noida, and the other, our affiliate QA InfoTech Inc Michigan USA. In 2010 and 2011, QA InfoTech has been ranked in the top 100 places to work for in India.