Web and mobile application testing services

Application testing services and mobile application testing services for web apps, iOS and Android apps, APIs and AI features. We test by hand where things are new, automate where things must not break, and keep every test in your repository so your team can run it after we leave.

Tested before launch: TradeAlert, built for 10,000 concurrent users · a Teams AI agent · a clinic website's ADA check

WHEN TO CALL US

Signs you need QA and testing services

Each of these is a ticket someone has already filed, in some form. Two or more usually means the release process needs a tester, not more developers.

SIGN-01Open

Every release breaks something that used to work

Last month's fix is this month's bug, and nobody re-checked the old screens.

Reported by: a customer
SIGN-02Open

Developers test their own code, and nothing else

Each feature works on its own; nobody checks the flow from sign-up to invoice.

Reported by: your support team
SIGN-03Open

Releases slip because regression takes days by hand

The same screens are clicked through before every release, and checks get skipped when the deadline is close.

Reported by: the release manager
SIGN-04Open

Customers find the bugs first

The first report of a broken checkout comes through a review or a support ticket.

Reported by: a customer, again
SIGN-05Open

It works on one phone and fails on another

Nobody knows which devices and OS versions your users actually have.

Reported by: an app store review
SIGN-06Open

The AI answers seem to be getting worse

A model or content change went out, and nobody can say whether answers improved or not.

Reported by: a member of staff

Application testing services we provide

Six services, each used on systems we built and now run in production. We prepare the app for your penetration-test firm too, and fix and retest what it finds.

Functional and regression testing

Every feature checked against what it should do, and a regression suite that re-checks what used to work before each release. A failing check stops the release, not a customer.

Test automation

UI and API tests written as code, run on every change and kept in your repository, so they run without us.

Mobile application testing services

Real phones and tablets chosen from your analytics, several OS versions, weak networks, and the App Store's and Google Play's pre-release tracks. Our VoIP apps were tested under varied network conditions before rollout.

Performance and load testing

Traffic at the level you expect, then above it, before your users bring it. TradeAlert was built for up to 10,000 concurrent users.

Accessibility testing

Keyboard, screen reader, contrast and zoom checks against WCAG. The urgent care clinic website had an ADA accessibility check before launch.

AI feature testing

Evaluation sets of real questions with expected answers, and a pass rate tracked on every release. Our Teams helpdesk agent was measured against real staff questions before launch.

Where testing sits in a release

Testing starts when the requirement is written, not when the code is done. Automated checks run on every merge; people test what is new on staging; your team signs off.

Where testing sits in a release: test cases are written with each requirement, developers add unit tests, automated API and UI regression runs on every merge, a person explores the build on staging, your team runs user acceptance, and the release is monitored, with a failing test sending the change back
    Automated checks catch what used to work; people test what is new. Orange boxes are the steps your team owns.

    QA, testing or test automation?

    Buyers often use these words for the same thing. They are not, and the difference changes what you buy.

    TermWhat it means
    Quality assurance (QA)The process that stops defects getting in: requirements review, test plans, release criteria
    TestingChecking the software does what it should, by hand or with code
    Test automationTests written as code that run on every change; pays off for flows that rarely change and run often
    Manual and exploratory testingA person trying to break it; still the best way to test new features and usability
    QA as a serviceA tester or team working on software someone else builds

    WHAT WE TEST

    Testing types at a glance

    Every testing type answers one question. Pay for the ones whose question matters to you now.

    Functional

    Each feature does what the requirement says.

    Worth it whenAlways, for anything new.

    Regression

    What worked last release still works.

    Worth it whenYou release often and old features break.

    Integration and API

    Systems pass the right data to each other.

    Worth it whenYour app talks to an ERP, CRM or payments.

    Performance and load

    Speed and stability at expected and peak traffic.

    Worth it whenA launch, a campaign or a busy season is coming.

    Usability

    Real people can finish tasks without help.

    Worth it whenSign-ups or bookings drop at one step.

    Accessibility

    People using keyboards, screen readers or zoom can use it.

    Worth it whenPublic-facing apps, and before an audit.

    Compatibility

    Browsers, phones, tablets and OS versions.

    Worth it whenUsers report it works on one phone, not another.

    User acceptance

    Your team confirms it fits the way they work.

    Worth it whenBefore every significant release.

    AI evaluation

    AI answers or decisions stay right as models and content change.

    Worth it whenYour app uses a language model.

    Which kind of testing fits your situation

    Start from what goes wrong today, not from a list of testing types.

    If youWhat usually fits
    Change the app weekly and the same flows keep breakingAutomated regression on those flows
    Have a new product whose features still moveManual and exploratory testing; automate later
    Release rarely, but each release is high-riskA release checklist and a regression pass before each
    Have a mobile app used on many devicesTesting on real devices, chosen from your analytics
    Ship AI answers or decisionsAn evaluation set with a pass threshold, run every release
    Have no tester and a small budgetPart-time QA, not a full team

    MANUAL OR AUTOMATED

    Manual vs automated regression testing

    Automated regression checks what used to work, in minutes, on every change. It does not replace a person testing what is new.

    Regression by handAutomated regression
    • Days of clicking through the same screens before every release.

      Stable, high-traffic flows such as login and checkout checked on every change, in minutes.

    • Releases wait for a tester to be free.

      Testers spend their time on new features and exploratory testing.

    • Checks get skipped when a deadline is close.

      The same checks every time, with a record of each run.

    • Results depend on who ran the test and how tired they were.

      A failing test blocks the release before a customer sees it.

    • Customers find what used to work and now does not.

      Tests live in your repository and run without us.

    Our ruleDo not automate everything. Automate the stable, high-traffic flows such as login, checkout and your main workflow, and test new features by hand until they settle.

    AUTOMATION PAYBACK

    When test automation pays for itself

    Automation pays when the same checks run often and the screens rarely change. Put in your own numbers: nothing here is filled in for you.

    The sum: manual hours per release × releases per month × hourly cost is what manual regression costs each month. Take away the monthly upkeep of the suite (hours × hourly cost). Divide the build cost by what is left to get the months to break even.

    Manual regression each month
    -
    Suite upkeep each month
    -
    Saving each month
    -
    Break-even
    -

    Fill in all five numbers to see the result. Amounts are in whatever currency you use.

    HOW WE TEST

    How testing was done on our live projects

    We have not published a standalone testing engagement. These are systems we built, tested and now run in production, and the method each one needed.

    1. Parallel run

      BIM to estimate integration

      Run alongside manual takeoff on live projects before it became the standard. Elements it cannot map are listed in an exception report, never dropped.

      Case study →
    2. Checks first

      Invoice processing agent

      The checks were built before the automation; anything doubtful goes to a review queue with the reason, and the agent ran alongside the manual process first.

      Case study →
    3. AI evaluation

      Copilot Studio helpdesk agent

      Measured against the questions staff really asked, with a pilot group, before anyone else used it.

      Case study →
    4. Network conditions

      Custom VoIP apps

      Call quality tested under varied network conditions, and integration tests in a controlled environment before rollout.

      Case study →
    5. Load and security

      TradeAlert trading platform

      Built for up to 10,000 concurrent users, with testing for functionality and security before deployment.

      Case study →
    6. Accessibility

      Urgent care clinic website

      Functionality and compatibility testing and an ADA accessibility check before launch.

      Case study →

    Security testing for fintech and healthcare apps

    Apps that hold money or health data need a penetration test by a specialist firm, against a published standard, before launch and after major changes. Our part is to prepare the app, fix what the testers find and retest before release.

    Standard or checkWhat it coversWho does it
    Penetration testAn attempt to break in, by a specialist firm, before launch and after major changesThe specialist firm; we prepare the app and test accounts
    OWASP ASVSThe Application Security Verification Standard: what a secure web app must doChecked by the testers; we fix what they find
    OWASP MASVSThe Mobile Application Security Verification Standard, for iOS and Android appsChecked by the testers; we fix what they find
    PCI DSSRequirements wherever card data is stored, processed or sentYour organisation, through a self-assessment or a qualified assessor
    Patient data safeguardsEncryption, access controls and audit trails, as we built into Telemeds+Built into the app; retested before release

    How a testing engagement starts

    The first result is a written audit you can use with or without us.

    1. 1

      A call about what breaks

      The app, its platforms, how often you release, and the last bugs your customers found before you did.

    2. 2

      A test audit

      Your current tests, the gaps and the riskiest flows, written up so you can act on it with or without us.

    3. 3

      A test plan and the first suite

      Automated regression on the one flow that breaks most, running in your pipeline.

    4. 4

      Testing every release

      Exploratory testing of what is new, regression on what is not, defects logged in your own tracker.

    5. 5

      Hand over or keep running

      The suite, the plan and the notes are yours. Keep us on, or run it yourselves.

    What QA and testing cost

    A test audit is a fixed price; ongoing testing is priced by the month. Three things move the price, plus whether security testing is in scope.

    high impact

    Platforms and devices

    Web only, or web plus iOS and Android on the devices your users have.

    medium impact

    Share automated

    Automation costs more up front and less on every release after it.

    medium impact

    Release frequency

    Weekly releases need a regression pass every week; quarterly ones do not.

    Ways to work with us

    Pick the model by how often you release and who builds the software.

    Test audit

    A fixed price after the first call: your current tests, the gaps and the riskiest flows, written up so you can act with or without us.

    QA engineers by the month

    For teams that release often and need testing in every sprint, through IT staff augmentation.

    Testing as you need it

    Release testing or a one-off regression pass, billed for the time used.

    Testing inside a Metwaves build

    Part of the build estimate for software we build for you, not an extra.

    CHOOSING A PARTNER

    How to choose a QA and testing company

    Signs that separate a testing partner from a test-case counter, including where we are not the right fit.

    Good signs

    • They show a sample bug report and test plan before you sign.
    • Test code lives in your repository and runs without them.
    • They can explain which flows they would automate, and which they would not.
    • They test mobile apps on real devices, not only emulators.
    • They log defects in your tracker, not a private spreadsheet.

    Warning signs

    • They promise to automate everything.
    • Reports count test cases run, not risks found.
    • Nobody can say who will test your app, or whether they are on staff.
    • AI features are tested once, by hand, and never again.
    • You need hundreds of devices and certified testers in a regulated domain: a large testing house may fit better than us.

    Related services

    Testing sits inside every build. These pages cover the builds and the people.

    Mobile app developmentiOS and Android apps, tested on real devices.
    Custom software developmentWeb and business software with testing built in.
    IT staff augmentationA QA engineer inside your team, by the month.
    Application supportFixes and regression testing after launch.
    Remote usability testingOur guide to testing with real users at a distance.

    QUESTIONS & ANSWERS

    Application testing and QA: frequently asked questions

    Straight answers on cost, automation, mobile, AI features and ownership of the tests. If you don't see an answer to your question, you can send us a message from our contact form.

    It depends on the platforms and devices, how much is automated, how often you release, and whether security testing is included. A test audit is a fixed price; ongoing QA is priced by the month. Use the payback calculator on this page before you pay for automation.

    Automate the stable, high-traffic flows such as login, checkout and your main workflow, and test new features by hand until they settle. Automating screens that change every sprint costs more than it saves.

    Tell us about it. Our published testing work is on software we built and run, so we will say on the first call whether a test audit from us fits. We need access to the code, a test environment and your tracker.

    Testing on real phones and tablets chosen from your own analytics, across OS versions and network conditions, and through the App Store's and Google Play's pre-release tracks before a public release.

    With an evaluation set: real questions or inputs, the answer each should get, and a pass rate tracked on every release. Our Teams helpdesk agent was measured this way against real staff questions before launch.

    You do. They live in your repository, run in your pipeline, and work without us.

    QA is the process that stops defects getting in, such as reviewing requirements and setting release criteria. Testing checks whether the software does what it should.

    A penetration test by a specialist firm before launch and after major changes, checked against OWASP ASVS for web or MASVS for mobile, plus PCI DSS where card data is handled. We prepare the app and fix the findings.

    INSIGHTS & NEWS

    Discover IT Industry Insights and Technology Trends

    Nikit Nandan    Nikit Nandan November 17, 2026
    Is .NET 8 Still Safe to Run After November 2026?

    Support for .NET 8 and .NET 9 ended on November 10, 2026. What that means for apps still on them, how risky it is, and how to plan the move to .NET 10.

    Learn More >
    Microsoft Technologies 5 min read
    Nikit Nandan    Nikit Nandan November 12, 2026
    Top AI agent development companies for small and mid-sized businesses (2026)

    Nine AI agent development firms a small or mid-sized business can shortlist, with Clutch figures, minimum budgets and the weaknesses their own clients name. Published by Metwaves, which is not ranked.

    Learn More >
    AI Development 9 min read
    Nikit Nandan    Nikit Nandan November 12, 2026
    Enterprise AI Agent Platforms Compared (2026)

    An architecture review of the four big enterprise AI agent platforms: where each one reasons, acts, finds its data and connects to outside agents, what it costs, and when a custom build is the better choice.

    Learn More >
    AI Development 9 min read

    Let’s build your idea!

    Let us help you with free analysis & estimate. Our product expert will contact you within 12 hours.

     All your information is kept confidential and protected by NDA.