Quality Assurance is in the Middle of your Software Delivery Life Cycle

Major organizations and consulting firms make the call many times to skimp on quality assurance (“QA”) testing throughout my IT career. The executive management team always comes up with reasons for not spending the time or effort on it. Quotes like: 1) We need to hit the date at all costs; 2) We don’t have the budget for competent testers; or 3) Let’s reduce the functionality in this release to make the date; 4) We only have a handful of users in the first release so let them debug it, are some of the top ten reasons. But the poor choices and mistakes do not end there. Many organizations look at the QA role as an afterthought or at the last minute because they do not have a Quality Assurance organization.

Let face it, in today’s world releases come in days and not weeks or months. You have to be nimble and aggressive to keep up with the rest of the companies out there that have embraced Agile Development, DevOps and MicroServices. If the application you are building is customer facing, the imperative to get it right is huge, but it still matters even if it is a back-end system or non-facing customer application because it can still impact your organization in a big way.

Why does it matter? I always laugh at this question when as a consultant, we have to justify the expense. The companies asking for this don’t get it and it is one of the reasons why large consulting firms and consultants are needed. Many of the projects I have worked on have been programs that have put the company in jeopardy. They mismanage a major company transformation or ERP implementation. It then costs them tens of millions of dollars to get out of the mess. I read a good article on the subject that compares the costs of poor quality to the costs of good quality. www.isixsigma.com/ Rework, retesting and downtime are some of the internal costs of poor quality. Complaints, warranties, bad-will and sales reductions are good examples of external costs of poor quality. The author states that “The traditional view would be to conclude that if a company wants to reduce defects and by this reduce the cost of poor quality; the cost of good quality would have to be increased, meaning higher investments in any kind of checking, testing, evaluation, training of operators, etc.”

Who is accountable? It has to start at the top. The executives need to mandate it, monitor it and reward good behavior. The key is to make it a key objective of the IT organization. If you don’t, you will continue to see epic failures. I did read an interesting article about firing your QA Team and making your developers accountable. blogs.forrester.com/ I would not recommend this approach for everyone because it puts more of pressure on your development team that already has a lot of pressure on it in the first place. However, there is a lot of credence on putting your QA team in an organization outside of the program/project or IT Development groups. They can put a lot of pressure on them to “fix” reports to make it seem like everything is A-OK when the real story could be much worse. I cannot finish the count on my hands and toes the number of times I have seen this happen firsthand.

Failures can cost jobs and lives. Ironically, these types of situations have impacted my career a number of times because of the stress of the situation. One of the situations was at BearingPoint. They had an epic failure in implementing a new financial system on PeopleSoft, and it was one of the many reasons why it filed for Chapter 11 bankruptcy. Internally, we called the PeopleSoft implementation a controlled crash landing. IT had to implement by a certain date, but they were not ready. When our internal IT group did implement the ERP solution, our project financials became suspect (by the auditor) and all of the Partners/Directors had to do a complete audit of our financials. Well over $100 million dollars later in audit fees, we were able to stabilize the reporting of the financial system. BearingPoint never recovered from the poor testing of the ERP package (among other things), and in 2009 BearingPoint declared bankruptcy and the executives sold our businesses in pieces to many other consulting firms.

Is it worth it? The impact on individuals and lives do matter. Another good article rates the top five quality failures that could have or did kill people. www.processexcellencenetwork.com/ The first one they recognize was the software glitch with NORAD that could have started WW3. Another gaff was with NASA and the Mars climate orbiter using English instead of the Metric system. Flight crashes and radiation dosage mistakes took lives. Earlier this year, I wrote a LinkedIn article calling out Samsung for their lack of testing on their new Android phone. To prove it does matter, another QA article outlines that the investment in quality assurance is worth it. www.seamgen.com/blog/ The author states “A study done by the National Institute of Standards and Technology shows that the cost of fixing bugs after production increases 5 to 32 times when compared with fixing the same bug during QA. It’s obvious that the real value in quality assurance comes during development.” The author also quotes a study completed by Capgemini and HP: “as a percentage of IT budgets, QA spending will increase to 28% this year.” Ironically, the author also thinks that number is too high, which will result in budget cuts for the QA groups in major organizations.

In this day and age with social media, your brand can suffer damages it can never recover from. Jonathan Freeman, Professor of Psychology, University of London, Managing Director of i2 media research at Goldsmiths said it best: “Consumer expectations regarding apps are really high, so when people’s experience is not satisfying, they are going to go elsewhere and look for an alternative. It’s therefore very important for app developers and service providers to test and optimize.”

What is your opinion? Your thoughts are appreciated.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *