One of the challenge is how to speedup QA testing processes and increase accuracy, completeness without breaking the tight budgets. Share your thoughts

by admin on May 7, 2013

Effective Test Optimization is a key to speedup the QA testing processes with accuracy & completeness and certainly this would also help in meeting the tight budget constraints. Preparing an effective test approach is a forte of Quality/Test Manager who with his/her rich experience can plan the test distribution across the target entities using a calculative approach; that should encompass the historical data, usage statistics and an attempt to provide relevant coverage to each possible combination.

With trends such as Cloud Computing, Globalization, and IT Service Management, it seems very likely that it will be important for organizations, and their cloud providers, to have quality positions that are more than hierarchically speaking, low level testing roles. This is not meant to down play the importance of these testing roles. They are vital; their contribution essential. These (testing) roles are in fact very much needed, but ideally they are part of a quality organization, and quality management system and report up to a senior quality role, as example, a quality manager, director, vice-president or chief quality officer. Depending on the size of an organization all of these types of roles could in fact exist, along with lower level quality staff such as the ever important quality analyst, automation engineer, and tester types.

In today’s world, it seems likely that many organizations are without, but could greatly benefit from, a high level quality role (such as a chief quality officer reporting to the board at the extreme) for ensuring all roles within an enterprise adhere to (QMS) requirements (e.g., of policies, agreements, and legislation).

Ideally even Quality Control is not, or should not be, positioned or shown as limited to testing or projects, as it ideally should be inherent to a large (full enterprise) scope, and as it (quality) is or should be everyone’s responsibility, and as everyone has or should have a stake in quality. Yet even in organizations that do have a separate QC group, it is likely for most that it should be taken up a level from where it exists today (which it often seems, reports to a project manager as example) toward that which it seems would be better, and more ideal, in order for the quality organization to have a greater influence and impact on (or provide greater value to) the enterprise, such as by having the QC group shown as a CIO support function with a direct reporting relationship, and with a scope that matches the CIO’s domain or areas of responsibility and accountability.

Here is a bit of food for thought. Hope it helps. It will be presented in three parts.

Part I

Further to various prior writings (e.g.,http://icblog.firstreference.com/author/rrichard/#axzz2CtmoHZfS ), including with a key ISO connection ( http://icblog.firstreference.com/service-management-scope-definition/#axzz2MVPFqxPE ), and including on the distinction between QA and QC noted within this series (see all partshttp://www.firstreference.com/past_icl/vol4_issue7.html#5 ), it generally seems common to see QA and testing shown as basically one in the same (i.e. QA testing).

Therefore what I wonder is: How long will it take for people everywhere to better distinguish between the two?

Elaborating slightly, if one challenge is how to speedup testing and increase accuracy, completeness and adherence (e.g., to budgets, timelines, policies, standards, procedures and the like), perhaps the answer will or can be found in better application of true QA (e.g., such as by focusing on software productivity metrics, or measures/rates, to provide input to the optimization of processes, tools selection and use, and so on). Perhaps too, the answer will or can be found by ensuring the disciplines of quality are properly positioned within an organization, including having one of the related roles positioned sufficiently high enough so the quality organization overall has influence, and the highest probability that it will positively affect change, continual improvement and progressive increases to value.

In any case, it seems too often organizations do not properly distinguish between QC and other higher level quality disciplines, and in this regard, if there are Risk, Legal and other corporate level (enterprise-wide) support functions that report outside of IT, these likely should include (there likely should be one for) Quality, that addresses quality assurance, quality improvement and quality management.

In today’s world, and on route to 2020, globally organizations generally seem to need to do a better job at being proactive, engaging stakeholders ( as noted in my recent post, http://icblog.firstreference.com/change-exponential-power-enterprise-architecture-governance-and-stakeholder-engagement/ ), and ensuring all business unit policies (e.g., IT policies) align with corporate policies (e.g., ideally an organization should have an IT quality policy that aligns with a corporate quality policy). This seems to be an opportunity to improve both consistency, and optimization, and therefore should ideally be a help with addressing various challenges, including how to speedup testing and increase accuracy, completeness and adherence.

I think the problem is multi-aged sword because decrease of testing timing requires increased number of human resources. Either way, the budget couldn’t be kept tight. The solution in my view is to set up best possible test suite with n number of tuples, where the value of n won’t be more than 3. This way the suite can have definite number of test cases, thus making the model compatible with tight budget.

In my view, the other way of speeding up QA testing processes and accuracy within the tight budgets is test automation using open source automation tools. It is important to identify the repetitive test suites, regression tests and a best available open source automation tool that suites well for the product under test.

Also enforcement of SQA processes in software testing can also help in achieving the optimum speed and accuracy. The SQA ensures that QC is doing the right things in a right way and minimizes the chances of committing mistakes hence saves rework efforts. Also when you use the defined ways of doing something which are already proven on other projects, you can save on efforts and complete the project in tight budgets.

 Also see PMBOK 5, Quality Management, Chapter 8 for distinction between QA and QC.

One way is to take a closer look at the regression suite – is it really necessary to run every single regression test? Time is better spent focused on the critical areas of the application.

Here’s how I explain the difference and why it’s so important to any organization.

Quality Assurance is an investment which should be integrated into the entire process and throughout the organization. Planning, specifications, standards, and training all play a vital role in making sure quality is carried throughout the project. Testing, which is another valuable function of quality, ensures the deliverable solves the business problem. Knowledge learned and trained at this phase can be applied to all projects.

Quality Control comes after the fact and is an expense to the organization. It’s the checking of the final product after manufacturing/development process. It goes by many names, defects, rework, scrap, and regardless of what the company is manufacturing when issues are discovered here the company suffers in lost revenue and possible harm to their reputation.

Just taking the happy path through the final product is quick, but time would be better spent focusing on mission critical areas.

As example, Quality (3.1.1), degree to which a set of inherent characteristics (3.5.1) fulfills requirements (3.1.2). Note 1 to entry: The term “quality” can be used with adjectives such as poor, good or excellent. Note 2 to entry: “Inherent”, as opposed to “assigned”, means existing in something, especially as a permanent characteristic.

https://www.iso.org/obp/ui/#iso:std:iso:9000:ed-3:v1:en

Btw, while there is a time requirement to establishing and maintaining an automated regression suite, from a risk management perspective, there is likely more value in the simplicity of a nightly unattended run than in not having one, particularly given the complexity that is often inherent within our interconnected systems, access points, and secure operations?

In any case, in coming years, ideally greater distinction exist in practice, between quality control (e.g., detection), and quality assurance (e.g., prevention), so logical best practice theory and standards are more so an international or global reality.

#SQASolutionShare on FacebookShare on Google+Share on LinkedInTweet about this on TwitterEmail this to someone

Previous post:

Next post: