I was on my way to meet this very supporting client of ours. She is an expert in QA, with vast experience, certifications and accreditations. When it comes to testing, she is a purist. Her testing group has reached real maturity. Defects are found in RQ and design stage. When test execution starts, most defects are found in the very beginning - a show of process and execution maturity.
I was wondering what our discussion would be about. Would it be about what more can be done to improve further? Would it be about participation in some forum to showcase the excellence?
As I entered the meeting room, I found something amiss in her ever excited behavior. There was something that was really bothering her. What was it? Well, the best thing was to ask her. I asked her what was really bothering her.
I was surprised to hear what she had to say. Though I knew of it, I wasn't expecting to hear it from her. Her drive to perfection in testing had become her botheration. She was being asked why testing should be done when so very few defects are found. Is testing really adding any value, or is it just adding time to the release of the products? The purist in her was hurt to the extent that she was contemplating moving away from such a firm, where the top management was asking such questions. Was she wrong, perfecting testing to that extent? How do you show the business value of testing in a world where the defect count is the measure of success of testing?
Well, we had a good meeting and at least some of the lost excitement was back on her face. We both decided to think, focus and write down our thoughts that will help outline the business value of testing.
Testing is a key component of any quality software application that is oriented to ‘prevention' and aims at improving the product development process from the outset. While this facet of software testing seems obvious, it is often ignored and in fact presents significant challenges due to its intangible nature. At one extreme, companies that do not perform any kind of testing are at risk of potential financial and value consequences to the business if serious or minor defects are discovered by customers. At the other extreme, the cost of removing 100% of the defects in the software through testing are high, and can result in delays in go-to-market time.
More often than not, quality testing is compromised for a quicker market launch. Taking this chance and releasing a product in the market without testing can backfire badly if found defective, and can have a long-lasting, adverse effect on the company's reputation itself. In the long run, testing should be seen as a cost-effective opportunity to reduce risk, lower costs and enhance a company's ability to deliver a quality product to its customers.
In smaller organizations, testing is conducted by developers and business analysts that only alienate the senior management from the value of the entire process. Testing only at the final stage results in a greater number of bugs detected, which further translates to higher costs. Often these companies see testing as a success when it corroborates to a high rate of defect detection. However in reality, high defect finding rates are by far a good measure of testing. Rather, it points out to a lack of good quality assurance across product lifecycle.
Organizations with good quality assurance are those that conduct testing at the outset of the product lifecycle and thus ensure greater control over it. A defect at the requirement stage can snowball into 20-100 defects by the testing stage. Thus, the earlier a defect is found through testing, the lesser the development rework and retesting, and minimizes implementation costs. Finding defects early also increases operating efficiencies and results in a better quality product, which in turn means more satisfied customers. Ironically, it has been seen that it's a common practice to praise and accredit teams that find bugs during the testing stage than those that work throughout the product lifecycle to prevent these same defects and focus on developing a near-defect free product by the time the testing stage is reached!
A good indicator of successful testing is based on how well a company's product performs in the production environment. While IT organizations cannot predict all possible defects in a product, a comprehensive quality assurance approach minimizes the risk of developing a defective product. In a nutshell, the value of testing is essentially an investment in insurance and focuses on ensuring quality control throughout the product lifecycle.