Selecting the right testing metrics for specific projects

Insight Main Image

Each application has a different set of features and is created with specific user activities in mind, meaning that no two pieces of software are going to be exactly alike. When this is the case, how are quality assurance teams supposed to chose the best set of metrics to track over the project's lifetime? While there is no list of specific areas that should be monitored, there are a few steps QA can take to determine which software testing metrics will be the most advantageous. Here are some things to note when selecting the right metrics for specific projects:

1) Make sure they define and enhance goals
Companies can choose from a vast number of metrics to track, but the main point of this data analysis will be to further their goals and success. If the statistic will not yield the information to help achieve these benchmarks, it's likely not worth monitoring. Six Sigma contributor Michael Ohler noted that QA should focus on one primary metric while still factoring in the impact of other available information. This type of mindset will help choose the best set of metrics to analyze and will enable business development.

"A sound project statement is phrased using the primary metric," Ohler wrote. "For example, "The project goal is to improve (primary metric) from (current performance) to (desired future performance) by (desired date of completion)."

From establishing the primary metric, QA teams can then select other areas that will bolster the main goal. This will help create a strong set of metrics to track that will provide insight related to what developers need to do to maximize benefits.

2) Address every aspect of the project
While apps may not be all the same, they tend to undergo the same creation processes. This could make it easy to create a set of universal questions that will help ensure that each program is on track and will live up to expectations. CVR/IT Consulting's Gary Evans suggested building categories of metrics based on the cost, time, scope, risk, quality and other aspects of the project. By focusing on these areas, organizations can build out the metrics they wish to track, ensuring that they will only be looking at information that is important to their needs.

For the time category, for example, developers may want to have metrics pertaining to if project goals are meeting their deadlines. This information could foster decisions for how to fix issues and how much is being impacted if teams are not using time management effectively.

3) Ask questions from the start
Using metrics just because others are leveraging them may not yield the same results. Organizations have to approach projects with a fresh set of eyes, and understand what each area will mean to the overall quality of the application. Six Sigma contributor Niraj Goyal noted that there are a few questions to ask to verify the QA test metrics are appropriate for business needs: What will be measured and why is it required? How precise does the measurement need to be? How will it be measured and what will it be used for? Addressing these queries will be imperative to ensuring that the metrics will be valuable to the project.

"One of the reasons that a measurement might go unused is if the metric is inappropriate for the user," Goyal wrote. "Among other reasons, this may be because it is not understood, does not serve a purpose or is too cumbersome or complex to use."

Organizations must ensure that all of their metrics have a purpose and are valuable to their projects as well as the bottom line, and they can all be stored in an enterprise test management solution. By following these considerations, QA can evaluate the possible categories and better leverage information that will help further business goals and development.

Related Articles: