Thousands of practitioners rely on these apps for diagnosis and monitoring in wound care, where an inaccurate release is not a defect but a treatment risk.
“In wound care, an inaccurate app is not a defect. It is a treatment risk.”
The client’s apps support medical practitioners with diagnosis, insights, and behavioural analytics in wound care. With hundreds of use cases per application, manual testing created significant effort and delayed releases, and any inaccuracy risked patient treatment, dissatisfaction, and regulatory fines. Previous vendor experiences had left the client skeptical that automation could handle their complexity.
ML arteka streamlined end-to-end automation with a CI/CD pipeline per app, built on a custom Jenkins plugin. The team deliberately chose the most challenging use case to build a proof of concept, then fetched iOS builds from TestFlight, ran end-to-end testing through UXPLORE, triaged bug reporting to TestRail, and sent automated notifications via Slack.
The result: 60% savings in labour costs, 50% fewer production defects, release cycles cut from one week to three days, 4x more DevOps releases, and 3x faster execution.
Business Outcomes
Automation replaced skepticism with stability, running more tests and finding more defects before release.
Compared to manual testing.
Compared to manual testing.
Down from one week.
The Transformation
From manual, skeptical, and slow to continuous and trusted.
- 1
ProveStart with the Hardest CaseBuilt a proof of concept on the most challenging use case to demonstrate that complex automation was possible. - 2
PipelineCI/CD per AppImplemented a CI/CD pipeline for each app using a custom Jenkins plugin that fetches iOS builds from TestFlight. - 3
TestEnd-to-End with UXPLORERan robotic end-to-end testing across hundreds of use cases, catching significant defects before production. - 4
ReportTriage and NotifyRouted bug reporting to TestRail and sent automated status notifications via Slack.
The Business Challenge
Complexity, Manual Effort, and Doubt
Multiple apps with hundreds of use cases, and a client who had been burned by automation before.
Multiple apps
Different functionality and hundreds of use cases per application.
Human-intensive effort
Manual testing strained resources, budget, and on-time delivery.
Automation skepticism
Prior vendor experiences left a lack of trust in automation.
Business Outcomes in Detail
What the Numbers Mean
Compared to manual testing.
Compared to manual testing.
Down from one week.
Technology Snapshot
The Continuous Testing Stack
Trust in automation is earned, not argued. Proving the approach on the hardest use case first turned a skeptical client into a continuous-testing organization, and safety followed the speed.
Executive Questions and Answers
The questions leadership tends to ask when evaluating an approach like this.
Trust
How do you win over a team that does not trust automation?
By proving it on their hardest problem first. A proof of concept on the most challenging use case demonstrated the capability before asking for broader commitment.
Speed
How did release cycles get cut from a week to three days?
A CI/CD pipeline per app, with end-to-end automation and automated reporting, removed the manual bottlenecks that stretched each release.
Safety
Why does testing accuracy matter so much here?
These apps inform diagnosis and monitoring in wound care. Inaccuracy can affect treatment and trigger regulatory fines, so more tests and fewer escaped defects directly reduce patient risk.
