Quality Assurance
Quality is built into the work, not inspected in at the end. Renvric's quality assurance combines a test strategy set at the start, automated checks on every change, exploratory testing by people who think like users, and acceptance criteria the client can verify before release.
Start your project
Quality as a team habit
Testing at the end of a project finds problems when they are most expensive to fix. Renvric brings quality engineers into requirements discussions, so that criteria are testable before code is written, and keeps testing continuous, so that each increment is checked when it is fresh in the team's mind.
The layers of testing
Unit tests
Fast checks of individual functions, written by the engineers who build them and run on every change.
Integration tests
Checks that components, databases and external services work together, often using test doubles for unstable third parties.
End-to-end tests
A small set of automated journeys, such as sign-up and payment, that protect the paths the business depends on.
Exploratory testing
Skilled testers probing the system with curiosity, finding the problems scripts do not anticipate.
Non-functional testing
Performance, accessibility, security and compatibility checks, chosen according to risk.
Building a test strategy
- Identify the highest-risk features and data flows and concentrate effort there.
- Decide which checks to automate and which to leave manual, based on how often they repeat and how stable the feature is.
- Define test data, including anonymised or synthetic records for sensitive domains.
- Set environments and devices to cover, such as browsers and operating system versions used by your customers.
- Agree entry and exit criteria for each release.
Share the outcome you need and Renvric will recommend the smallest useful next step.
Handling defects
Defects are logged with steps to reproduce, expected and actual behaviour, severity and evidence. Severity reflects user and business impact; priority reflects when it should be fixed. Consider a typo on a rarely seen screen and a rounding error in invoices: the first is minor, the second blocks release. Recurring defect patterns are discussed in retrospectives so the cause, not just the symptom, is addressed.
Acceptance by the client
- Acceptance criteria are written with the requirement and agreed before build.
- The client reviews working software at each demonstration and in a test environment.
- User acceptance testing is planned, with named testers, scripts or scenarios, and a defined window.
- Known issues are listed in release notes and accepted explicitly.
Metrics we use sparingly
Test results, defect trends, escaped defects found in production and automation stability tell a team where quality is heading. They are not used to rank people, and coverage numbers are read with care, since high coverage with weak assertions proves little.
Related areas to explore: Quality assurance and AI Integration Services.
Useful next steps: project-based delivery, compare the engagement models and our services.
The services behind this process are described under quality engineering, including user acceptance testing support.
Common questions
Who is quality assurance for?
Product owners, technical leaders and sponsors who need confidence before release. It is especially relevant for systems handling payments, personal data or critical operations, and for products that will be changed frequently over a long life.
What do you need from us?
Acceptance criteria or access to people who can define them, details of the devices and browsers your users have, test data or rules for creating it, and testers for user acceptance. Tell us about regulatory expectations that affect how testing must be recorded.
Do you automate all testing?
No. Automation is excellent for repeatable regression checks and poor at judging usability or finding unexpected behaviour. We automate stable, high-value checks and rely on exploratory testing for the rest. The balance is set in the test strategy.
How do you test performance and accessibility?
Performance testing uses realistic load scenarios against a production-like environment, with agreed targets. Accessibility checks combine automated scans, keyboard and screen-reader review against recognised guidelines. The depth of each depends on user needs and risk, and is agreed in advance.
Who can answer a related question?
Email info@renvric.com. Test strategy questions go to a quality engineer or technical lead, and queries about acceptance and release sign-off are handled by the delivery lead. Describe what the system does and where failure would hurt most, so the discussion can start with the areas of highest risk instead of generic testing.
Keep exploring
Ask what is automated, how flaky tests are handled, who tests before the client does, how performance and accessibility are checked and what the process is for a defect found after release.