Path and Functionality
What your users actually do. That’s the layer you need to address first, before anything else.
Not all applications have the same requirements. Depending on your specific challenges, some tests take precedence over others. The 14 types listed below are grouped by category, along with the criteria that determine your priorities.
WHERE TO START
No one automates 14 types of tests all at once. Here’s what we recommend based on your starting point.
Non-regression and functional testing, covering only 5 to 10 user flows. Those related to revenue: login, shopping cart, payment, and account creation. The rest will have to wait.
Non-regression, end-to-end, and API testing, integrated into your CI/CD pipeline. The goal is no longer just to cover everything; it’s to never block a deployment to production.
Performance, load testing, and production monitoring. The breaking point occurs before the peak, not during it.
RGAA accessibility, GDPR, and graphic design testing. The third one is surprising, but it shouldn't be: an unchecked visual change causes a mandatory legal notice to disappear without anyone noticing.
Your priorities also depend on your industry.
What your users actually do. That’s the layer you need to address first, before anything else.
Verify that each feature behaves as expected. Run this test with every release, on the workflows you can't afford to break.
They re-run the existing system to prove that an update hasn't caused any issues elsewhere. This is the first benefit of automation—and the easiest to measure.
They go through the entire process, from the first click to confirmation, including third-party services. They catch what no unit test can detect.
Replay your user journeys on real iOS and Android phones, using the screen sizes and OS versions of your device fleet.
Finding just one bug is enough to make the solution pay for itself over the course of an entire year.
The plumbing behind the interface. When it fails, the screen stays green, and no one notices until the report comes out.
Query your endpoints directly: response codes, schemas, response times. The fastest way to determine whether the outage is coming from the front end or the back end.
Make sure that every event sent to GA4 or GTM contains the correct values. Otherwise, your marketing decisions will be based on inaccurate data.
They compare the output with a validated reference. They notice the missing button, the truncated text, and the legal notice that's missing.
We have automated label verification, secured our data, and improved our reliability to make error-free strategic decisions.
An app that works every time—even on days when everyone shows up at the same time.
Measure response and rendering times on your key user journeys at regular intervals to detect performance drift before your users do.
Simulate thousands of concurrent users to identify the breaking point before the traffic peak, not during it.
They simulate your critical workflows in production around the clock and alert you as soon as a step fails.
What you need to be able to prove to an auditor, a regulator, or a client.
They verify the structure, contrast, and keyboard navigation against RGAA and WCAG standards. They generate the audit report you need.
Ensure that nothing is sent without consent: cookies, third-party scripts, network requests. The issue is rarely visible to the naked eye.
Monitor tags, redirects, canonical tags, and structured data. A site redesign could cause them to disappear without warning.
Measure page weight, the number of requests, and the energy consumed per session. Useful when digital sustainability is part of your commitments.
Mr Suricate provided Mr Suricate with a reliable, automated approach to testing and validating our accessibility requirements, ensuring a seamless customer experience that complies with regulations.
Order of magnitude figures observed among our customers. Calculate your ROI
Four factors carry more weight than the others, in this order.
The platform's self-healing capability automatically resolves some selector failures. For the rest, you decide what to handle internally and what to delegate to us.
If you're unsure about the order, our experts will review your platform and processes, and you'll walk away knowing which tests to implement first.
YOUR QUESTIONS
Test selection, priorities, maintenance: answers to the most frequently asked questions.
The most common types are functional tests, which verify that each feature behaves as expected, and non-regression tests, which confirm that a change hasn't broken anything else.
Next come performance tests, API tests, accessibility tests, UI tests, and data layer tests. The fourteen types we automate are grouped by category further up on this page, along with what each one checks and when to run it.
No, and that would be a bad idea. An initial rollout generally covers two or three types within a limited scope. You expand from there once the first scenarios are up and running and the team has gotten the hang of things.
Between 5 and 10 critical test runs are enough to ensure that the essentials are in place: login, shopping cart, payment, and account creation. This also makes it possible to quickly measure the gains achieved through a comprehensive non-regression testing campaign.
That's the real question—and the one that causes most in-house automation projects to fail.
The platform’s self-healing capabilities automatically resolve only some scenarios when a selector changes. Beyond that, you decide: either your teams handle the maintenance, or our teams do it for you. Mr Suricate replace your QA team; rather, the platform and our experts enhance it.
Details on the mechanics—including self-healing of selectors and shared blocks—are described on the test maintenance page.
Yes, on iOS and Android, through real device farms. The web and APIs are also supported. Thick clients are not supported.
No. Scenarios are built visually, without writing a single line of code. AI speeds up development and simplifies maintenance. QA, product, and business teams all work on the same platform.
The two ways to build a scenario are described in detail on the no-code scenario editor page.
Our experts analyze your platform, processes, and teams to identify the priority tests to implement.