How do we measure the Costs of Software Testing?
Measuring the cost of testing is an important step toward justifying any software testing initiative. All software testing expenditures are justified by comparing the benefits accrued with the cost. The benefits could be in terms of quality measurement and failure prevention or earlier detection etc.
Before testing managers start thinking on the testing cost measurements, following questions essentially come to their mind.
Q. 1: How much should be spent on software testing?
Q. 2: How much should be set aside for dedicated testing resources?
Q. 3: What levels of software testing are reasonable?
Q. 4: When are the software testing costs too high?
Q. 5: When are the software testing costs too low?
The answer to the above questions lies in determining as to how much the organization is spending on software testing now. The fact is that majority of the managers don’t have this information. They think they know because the project – control system reports expenses for each project phase. Such systems detail expenses, often by individual task, for each phase of the development life cycle. However, the cost of the test phase is not the cost of testing for the project.
Some testing work is carried out in other phases (design testing, unit testing, etc.) and a lot of work is performed during the systems test that is not testing (for example, documentation, debugging, deficiency analysis and removal, conversion, and training).Careful analysis usually reveals that actual software testing costs usually lies between 15 to 25 percent of the total project cost. Lot many testing managers feel that the software testing costs are much higher to the tune of 50 percent of the cost of a project. They mistakenly treat all project expenditure after programming as a testing cost. While it is true, for many projects, that a good way to estimate final total cost is to take actual expense through the programming phase and double it, it is not true that most of the cost after programming is the result of software testing. In fact, conversion, training, and debugging are at least as costly in total as is the true testing!
Determining software testing costs is a crucial first step to planning any improvement initiative and justifying the investment. An estimate of the amount being spent to test and measure quality, as well as the cost of rework or corrections, is fundamental. Most organizations spend too much time worrying about how to calculate these costs in detail. Crude estimates are fine at the beginning. Estimate conservatively so that the “actual” cost is a larger figure than the one we are working with. Our numbers will be big enough to justify the steps we want to take. There is no need to inflate them and run the risk of having someone challenge our argument.
As a percentage of total development, direct software testing costs will approach 25 percent. Indirect testing costs, or the costs of poor testing, are usually at least twice the direct costs and may be spectacularly higher. The total cost of software testing in most organizations is sufficiently large to catch the attention of almost any manager.
Once the costs of poor testing have been estimated, justification problems seem to vanish. The management should expect to see a return on the testing improvement investment. The benefits can be tracked and the results become visible.
Measuring the cost of testing:
The cost of software testing includes direct measurement costs plus indirect failure and correction costs.
|Sr.||Direct Costs of Testing (Indispensable Costs)||Indirect Costs of Testing (Due to Poor Testing)|
|3||Systems testing||Corrective action costs|
|4||Acceptance testing||Re-keying data|
|5||Test planning and design||Failures|
|6||Computer time||Analysis meetings|
|8||Cost of Terminals, staff etc.||Re-testing|
1) There is no hard & fast rules for prescribing how much software testing cost “should” be.
2) As long as the investment in software testing shows a positive return – in reduced costs of correction and failures – it is justified, and, in fact, adds to the bottom line.
3) When failure costs have been reduced so that they are no longer significant, then we might say we are spending too much on software testing. Short of that, no expense is too much.
4) The justification for more money spent on direct testing is the reduction of money spent on indirect testing.
|If you want to keep track of further articles on Software Testing,
I suggest you to subscribe my RSS feed.You can also Subscribe by E-mail and get All New articles delivered directly to your Inbox.
Get your Absolutely Free Copy of Several MS PowerPoint Presentations & E-Books related to ISTQB, HP Load Runner, IBM RFT, HP QTP & QC Certification Exams, prepared by Popular Writers & Trainers, by writing to: Software.firstname.lastname@example.org
Full Study Material for Popular Certification Exams:
An expert on R&D, Online Training and Publishing. He is M.Tech. (Honours) and is a part of the STG team since inception.