A collection of terms that testers might hear from devs and other engineers

One thing I want to change: Consistency This year I want to be more consistent. 2025 has been very spikey (high highs and low lows, professionally and personally). One thing I want to add: Read...

“When Amazon emails you saying your gifts won’t arrive until the 29th” - Idea from Rahul Parwal

A rollback is the re-deployment of a prior version of code. It is often performed as a fast-fix to avoid negative user impact when a regression bug or other defect is found quickly following a deployment. A rollback is not required if the new code or feature is hidden behind a feature toggle which could be turned off.While rollbacks avoid negative user impact and increased downtime, they are not always adopted as it requires increased developer effort as a rollback is not a permanent fix. Rollbacks need to be communicated effectively to ensure that future deployments don’t redeploy the buggy code back to production, which can also slow down other developers’ work. It also requires coordination with other developers to avoid introducing any breaking changes between the current commit and the rolled-back version.

Availability heuristic is a cognitive bias where people tend to rely on immediate examples or information that come to mind when making decisions or judgments. This can lead people to overestimate the likelihood of events or outcomes based on how easily they can recall similar instances, regardless of how rare or common those events might actually be.In software testing, here are three examples: A persons tendency to lean towards using a testing tool, or methodology simply because they used or learned it recently, even when a different tool or approach would be more suitable. When encountering a bug, a persons tendency to assume its cause is something they’ve seen recently while the actual root cause could be a different issue entirely, especially if that is a more obscure issue. Using the same set of functional test cases across multiple projects without adapting them to the unique requirements or challenges of the current project.  The availability heuristic should be overcome by discussing ideas with peers (who might be able to provide different perspectives or tooling choices), favouring data-driven decisions (using data to understand root causes or prioritise testing efforts) and using risk analysis to guide testing efforts i.e. Focus on high-risk areas of the software based on the business impact, user behavior, and technical complexity, rather than relying on gut feeling or past experiences alone.

The inverted test pyramid is a term used to describe when projects have too many e2e or UI tests compared to unit tests or other tests seen lower down the automation test pyramid. Many projects have too many large, slow, flaky tests and few smaller ones. Retrying failed tests and marking known failures introduce risk and waste. Examining the root causes is essential to breaking the cycle.

A matrix is an organisational structure in which two or more lines of command, responsibility, or communication run through the same individual. Matrix management is almost the opposite; an employee will report to two managers, typically a functional manager and a project manager.ExampleIn a previous role, my (functional) manager worked in another area of the business and my (project) manager was responsible for my day-to-day activities. Matrix management was adopted for all testers so that an employee (Emily) reported to a manager in the same project or product area (the tech lead, an experienced developer) and another more experienced person of the same job family within the same office location.When describing my experience in the below article, I described the more senior tester as my “main” manager, however this person didn't have all of the day-to-day context of my project.

A webhook is an automated event triggered when something happens. More specifically, a webhook is a HTTP request triggered by an event in a source system, that will be sent to a destination system, often with a payload of data. For example, when a website goes down, an automated call might send the admin an email. Webhooks are commonly used to perform small requests and tasks that don't need an entire API. They are automated; meaning they are automatically sent out when their event is fired in the source system. This focuses a testers attention on validating all the ways that the event could be triggered and ensuring that the HTTP request is sent and received as expected.

WYSIWYG is an acronym for "what you see is what you get" and is pronounced as “wizzy-wig”.A WYSIWYG editor is a system in which content (text and graphics) can be edited in a form closely resembling its appearance when printed or displayed as a finished product, such as a web page.

CTF stands for Capture the Flag. In the Cyber Security world this relates to events where people aim to use ethical hacking and web exploitation skills to find hidden flags. This may involve hacking into a machine to find the "flag" file or having a series of hidden "flags" (often strings) within a solution designed specifically for these events. Organisations may run CTFs internally or publicly prior to or alongside bug bounty programs and penetration tests. You can also find applications to practice security testing through a CTF game.