Rightsizing the Cost of Testing: Tips for Executives
By: Scott Barber
Note to readers: This chapter has been written specifically for senior man- agers and executives (subsequently, executives) . For the purposes of the discussions in this chapter, an executive is a person who has both the authority and responsibility to allocate the overall budget for software development activities, decide whether testing and testers are owned and managed by projects or are part of a centralized testing service, decide which software will be built or bought, establish or modify project time- lines and release dates, or make equivalent decisions in other areas of the corporation such as marketing, product support, or human resources (HR). If this doesn’t sound like your role and you don’t at least have signifi- cant influence over decisions such as these, you are likely to find much of what is contained in this chapter to be beyond your ability to implement. If this is the case, don’t feel discouraged or stop reading—the information contained in this chapter can still be immensely valuable to you as you find yourself interacting with executives.Maybe you can even convince your favorite executive to read this chapter and discuss its contents and implications with you as they relate to your corporation. And who knows; maybe one day in the future you’ll find yourself in an executive role, dust- ing off this chapter to help you with a challenge you never imagined you’d be facingwhen you read it the first time.
Testing is Just Another operating expense—Mostly
As much as corporate executives maywish that it were not the case, testing will simply never be anything other than an operating expense (OPEX)
unless their corporation is in the business of providing testing services as a revenue stream. No matter what fancy accounting is applied to try to make testing appear less costly or what new age calculation is used to justify the cost of testing in terms of return on investment (ROI), for most companies there is simply no direct revenue derived from this expense. For all that it can be painful to admit, testing in most cases is just asmuch an OPEX as facilities rental, human resources, and the “free” coffee pro- vided to employees in the hopes that it will make them more productive.
Naturally, everyone would notice very quickly if someone were to stop paying the rent on the office, eliminate HR, or sell the coffee pots in an attempt to make a company’s bottom line look better as a particularly dif- ficult quarter is coming to a close. In fact, doing any one of those things would probably lead to a near immediate revolt, followed by virtually every employee’s resume showing up on every job board in the area. Eliminating these OPEXs under the banner of saving the bottom line would be tanta- mount to drilling a hole in the hull of a sinking ship while proclaiming that this would help the flood waters drain
To illustrate, let’s take a look at how this worked out for Life Insurance For Testers, Inc. (LIFT). One day, after a particularly bad quarter, the entire functional testing team for LIFT’s online policy application product was simply let go . There were a few questions, and some other members of the software development team were nervous for the next few weeks, but there was no revolt among the remaining employees; there wasn’t even any whining. Business basically went on as usual. Some of the remaining members of the team spent a little more effort checking their work, but they felt like they had more time to do so, since they weren’t being bogged down in dealing with all of those bug reports, “ridiculous demands” from the testers, and incessant questions from product managers about why the testers were finding so many bugs in the first place. In fact, development seemed to be a smoother, happier, faster process—at least for a while.
But then, shortly after one of LIFT’s periodic feature upgrade releases, the number of fielded support calls went up dramatically. Before the team could get a patch fix coded and released, LIFT started receiving negative reviews in consumer reports online and in print, and both the new customer policy rate and the term policy renewal rate dropped significantly . There still wasn’t a mass revolt, but by the close of the following quarter, LIFT’s bottom line was worse than it had been before the testers were let go.
... to read more articles, visit http://sqa.fyicenter.com/art/