Add Value — software, data and intelligence
IT Let’s talk

Case study  ·  Insurance

Regression testing

Every release brings the same question:
what have I broken?

Insurer · automated testing service since 2019 · first UFT, then Playwright · black-box approach

Tell us about your case The numbers

Automated runs · release after release

each row a test cycle, each mark a feature checked None of these checks is repeated by hand: they run automatically, spread over several virtual machines, with the results landing in Jira. 2019 — startUFT thenPlaywright · open source
3.700+automated tests run
4 yearsof continuous service
Black boxchecked from outside, without touching the code
Jira + Xrayresults tracked and shared

Schematic representation of the test cycles. As reported by the project: more than 3,700 automated tests run over four years.

The need

The new must not
break the old.

The insurer needed to guarantee software quality and reliable releases, by checking that new features did not break the ones already there. The answer was regression testing that simulates what users really do, against the real applications and real data.

Regression is silent

The fault does not appear where you worked: it appears in a function nobody touched and nobody re-checked.

By hand it does not scale

Repeating the same checks at every release costs time and attention: after a while what shrinks is the coverage, not the risk.

It has to be real behaviour

A useful test repeats what a user actually does, on the real applications and the real data, not on a laboratory case.

The results have to be recorded

A fault found and not recorded is a fault that comes back: the result of every run has to land where the teams work.

What we did

Simulate, automate,
scale, record.

A service, not a project with an end date: since 2019 it runs at every release cycle and grows as the coverage grows.

PHASE 01PHASE 02 PHASE 03PHASE 04 SimulationAutomation Distributed runsRecording What users really do,on real data. First UFT, then Playwright:more flexible, open source. The tests run in parallelacross virtual machines. Results and faults into Jiravia Xray, for every team. RUNNING SINCE 2019 — EVERY RELEASE GOES BACK THROUGH THE CHECKS Real applications, real data Black-box approach Scales by adding machines Automation and system test together

The project in full

A service,
not a project with an end.

Since 2019 Add Value has run an automated testing service, built first on UFT and then on Playwright, Microsoft’s open-source tool, which brought more flexibility and the ability to spread the runs across virtual machines. The tests follow a black-box approach: they check what the system does from outside, without touching the software’s internal logic.

Managing and monitoring the results is integrated with Jira and the Xray plugin, which keeps the faults tracked and makes it easier for the automation and system test teams to work together.

The approach

From outside,
like a user.

The tests follow a black-box approach: they check the system’s behaviour from outside, without touching the software’s internal logic. That is what lets you check what the application does, not how it is written.

Moving from UFT to Playwright — Microsoft’s open-source tool — brought more flexibility and the ability to spread the runs across virtual machines.

user action real data expected result result recorded application internal logic untouched The test does not go into the code: it watches what goes in and what comes out.

The result

Four years,
3,700 runs.

Over four years more than 3,700 automated tests have been run, which lets the insurer release while keeping the integrity of the existing features under control.

3.700+

automated tests run over four years

Coverage widened

across a growing number of features

Repeatable

a structured, traceable testing process

Automation also widened the feature coverage and made the testing process more structured, more traceable and repeatable.

Inside the project

Three ways
of looking at it.

Whoever automates, whoever runs system test, whoever decides to release: one service seen from three desks.

Tests you write once

Playwright made the automation more flexible and allowed the runs to be spread across virtual machines: adding coverage no longer means adding hours.

  • From UFT to Playwright, open source
  • Parallel runs across virtual machines
  • Real user behaviour simulated
Automation view

Faults that land where the work happens

Managing and monitoring the results is integrated with Jira and the Xray plugin: faults are tracked, and the automation and system test teams work through one tool.

  • Results integrated into Jira with Xray
  • Faults tracked and assignable
  • One place for two teams
System test view

You release knowing what holds

Automation made the testing process more structured, traceable and repeatable: the decision to release rests on a result, not on a feeling.

  • The integrity of the existing features under control
  • Coverage that grows over time
  • Checks repeatable at every cycle
Release view

Technologies and areas

PlaywrightUFTBlack-box testing JiraXrayVirtual machines Open sourceRegression testing

Keep reading

Other stories

Banking  ·  Software

Exchange Digital Corporate

The Corporate Banking flow backbone was rebuilt and run in parallel with the old system, handling around 1.5 million flows a day.

from 10 minutes to under 1

The full case

Insurance  ·  Software

API Management middleware

A custom .NET API Manager with a two-way orchestrator and ELK monitoring, for near real-time policy issuing, 24/7.

+20% performance

The full case

Insurance  ·  Software

Life AM

Running and maintaining the life insurance software: corrective and evolutionary work, plus the one-off requests.

97% SLAs met

The full case

Industrial  ·  Data

From on-premise to ROSA

The OpenShift environment migrated to Red Hat OpenShift Service on AWS, with automated pipelines, Infrastructure as Code and health checks on the critical components.

migrated in about 3 months

The full case

Banking  ·  Software

Corporate Remote Banking

A multi-bank, multi-company platform with a modern interface and a microservices back-end, built for a technology provider to financial institutions.

0 outages

The full case

Banking  ·  Data

Data migration machine

Automates diagnostics, normalisation and transfer of the data needed to migrate NPL portfolios and support later acquisitions.

~350 transformations  ·  10 verticals

The full case

Banking  ·  Software

NPL field agents app

The Android app and the web portal, whose features partly overlapped, were brought together into one Angular Progressive Web App for the field agents and the recovery firms.

from 2 applications to 1

The full case

What does it cost you
to find a fault
in production?