Getting Support for the Tests You Need Done
About literature, Mark Twain said, “A classic is something that everybody wants to have read and nobody wants to read.”
The same can be true for software testing. It’s often hard for teams to get sufficient time and resources for the amount and quality of tests they believe are needed. It’s like management wants testing done but at the same time doesn’t want to commit what’s needed to do it.
In these situations, I usually look at the business side, because that is what testing is for. Testing costs time and money, which are business investments. Not testing also can lead to problems later on, becoming a business impact. There are myriad stories showing how the cost of testing late (or not at all) can cost companies hundreds of thousands of dollars in potential lost revenue.
The relationship between not testing and the business impact of issues is often a less direct correlation. Compare it to having a bad-health habit, like smoking. If you don’t test enough, the chances of issues in the product later on increases, just like a young smoker usually won’t encounter serious health issues until later in life.
If means are limited, negotiate priorities. Make a list of possible tests and discuss their priorities—which tests are must-haves and which can wait. Take a business approach by determining what the test costs and the benefit in risk reduction. A decision-maker who does not want to sign off on a test may hesitate to go on record saying “No” or “It can wait.”
Another question is whether to automate tests. It is common to analyze the ROI on automation. If investments in technology, knowledge, and test case conversion have a positive payoff because tests are executed repeatedly, the decision is to automate. This makes sense, but my approach is different.
Instead of ROI on automation, I like to look at the ROI on the tests. Once a decision is made to create a test, I practically always automate it. In the Action-Based Testing methodology, the tester writes the tests directly in a keyword-based format. Once those reusable actions have been automated, the test is also automated.
On a final note, Mark Twain’s quote says “nobody.” For testing, I would change that to “not everybody.” There are many professional testers who have made testing their living, and they can contribute their knowledge. However, they must get the room to do so, and in sprints that can be hard. This is where commitment from management is the main factor. Managers need to provide resources and make testing and test automation priorities for the whole team.
If you want to test but are not getting the support you need, find out if the decision-maker is one of those people Mark Twain talks about, who wants testing to happen but is concerned about time and cost. If that is the case, work out a solution: get more resources, set priorities for what can be done, or consider automation or outsourcing to save costs and speed things up.
