Triage is a medical term used to assign degrees of urgency to wounds or illnesses, in order to decide on the order (priority) of treatment across a large number of patients.
Defect triage (or
bug triage) is a term used in modern software development and is the practice of reviewing, discussing, and categorizing software defects based on their negative severity.
Unfortunately, performing defect triage can be a real pain (pun intended), since the process requires intimate knowledge of the entire software and its intended functionality, the capabilities of all relevant systems and components, and the status of all current or incoming defects in the system.
To help you ease the pain of defect triage, in today's article we'll be exploring exactly what defect triage is, and we'll look at a few useful tips aimed at helping you implement a robust and powerful triage process into your own software development life cycles. Let's get to it!
The entire act of performing defect triage revolves around appropriately defining and categorizing defects into relevant
severity levels. Thus, it is critical that your organization has a clear understanding of what both priority and severity mean in this context. The goal is to reduce potential ambiguity as much as possible so that everyone on the team is clearly aware of the differences between each level of priority and severity that your team will be employing.
Priority is defined as "the fact or condition of being regarded or treated as more important." Thus, applying this to defect triage is merely the act of treating or regarding one defect as more important than another. As dozens, if not hundreds, of defects are discovered and processed by you and your team throughout the entire development life cycle of a given project, it should become instinctual for those involved in the quality assurance process to naturally assign a similar priority level to defects of a certain type. This will come naturally with practice, but can also be supported by well-defined documentation shared across the team, which aims to outline and detail the priority (and severity) levels that should be used.
Perhaps the most important practical result of prioritizing defects is the actual turnaround time necessary to resolve a defect. As multiple defects are assigned various priority levels, a natural hierarchy of importance is generated, which ensures that those defects higher on the list are fixed before those further down the list.
You and your team can decide on how many priority levels are required, but it is common practice to use three or four different levels:
urgentpriority indicates that the defect should be resolved immediately, usually taking precedence over every other defect in the system.
highpriority is for everything that should be just below
urgent-- that is, if a defect should be prioritized, but is not mission-critical to the stability of the software (or to the ability to test said software), a high priority will usually suffice.
Severity is defined as "the fact or condition of being severe," which is one of those frustrating definitions that essentially uses the original term within the definition.
Severe is defined as "(of something bad or undesirable) very great; intense," so we can extrapolate that severity essentially means "the fact or condition of being intensely undesirable." In the realm of bug triage, most people think of severity as the seriousness or potential negative impact of a defect, if left unresolved. Measured severity levels are often directly correlated with the client's (or organization's) risk assessment -- that is, the potential negative impact brought about by the given defect.
Just as with priorities, it is up to your organization to decide on the appropriate number of severity levels. That said, most teams find parity between the number of priority levels and severity levels to be useful and natural, so three or four is often the standard:
criticaldefect is one that completely hinders the software in some way, by either halting normal execution, or by preventing normal testing.
There are many possible priority and severity combinations in which a defect can be categorized, so we'll briefly go over a few common examples here to illustrate how you might consider categorizing your own bugs.
criticalseverity: Defects that prevent normal execution or testing should be given the highest priority and severity categorizations.
highpriority and a
majorseverity: This categorization might be assigned to a defect that has a high impact on software functionality, but is only experienced in an unlikely/extreme use- or test-case.
minorseverity defects are reserved for issues that have no tangible, functional impact, but still fail to meet quality assurance standards. These often include things like minor cosmetic issues.
The act of performing defect triage can be draining and challenging, but ensuring that the process is carried out efficiently and regularly throughout the development life cycle will generate more robust, stable software upon release. There are a handful of major stages of the process, each of which consists of a number of minor steps that will force your team to evaluate any and all defects in the system, whether new or existing and ensure every one of them is properly categorized.
The first stage in the defect triage process is initial
screening, which simply asks a handful of questions aimed at determining which incoming defects can be ignored, and which should be investigated further:
The next stage is
confirmation, in which one or more team members attempt to replicate new defects, or to confirm the functionality of submitted fixes for (or possible regression of) existing bugs.
This stage is also when your team will decide whether a particular bug should be recorded and marked for resolution, or whether it should be ignored or put off until a later date. As such, it is vital that at least one member of the team is assigned as a facilitator of the group meetings aimed at tackling defect triage. This individual should be in charge of conducting such meetings and coordinating discussions between other stakeholders, each of which can provide input into the severity and priority levels that should be assigned to each defect.
Categorization is the real meat and potatoes of the defect triage process. It is within this stage that the team will actually categorize each and every defect into the pre-defined
severity levels. As previously mentioned, multiple team members should be capable of completing this categorization process on their own, assuming each individual priority and severity level is well-defined and documented. That said, it often helps to have multiple people discuss defects during this stage, since members of different teams may have varying levels of insight as to how severe or impactful an issue may be.
A sub-stage of
revision stage, allows existing bugs to be reviewed, removed, or recategorized, depending on the status of the defect. An obvious example is a defect that has since been fixed in the latest version -- if QA testing has confirmed the fix, it can be marked as such and set to low priority for follow-up regression testing in the future.
Automated defect tracking and reporting tools, like Airbrake, ensure that your team is immediately aware of exceptions the moment they occur. Airbrake's powerful error monitoring software guarantees that your team won't need to worry about losing track of that rare production defect that slips through the cracks. Airbrake provides real-time error monitoring and automatic exception reporting for all your development projects. Airbrake's state of the art web dashboard ensures you receive round-the-clock status updates on your application's health and error rates. No matter what you're working on, Airbrake easily integrates with all the most popular languages and frameworks. Plus, Airbrake makes it easy to customize defect parameters, while giving you complete control of the active error filter system, so you only gather the errors that matter most.
Check out Airbrake's defect monitoring software today and see for yourself why so many of the world's best engineering teams use Airbrake to revolutionize their defect handling practices!