Enterprise Test Automation · Since 2010

Right Tool.
Right Domain.

Nobody hands an electrician a table saw and expects him to wire a building. The domain defines the tool requirements. Not the other way around.

Tools in service of craft. Craft in service of the domain.

People
RJCadvisors
The practitioner. QA advisory and strategic guidance for enterprise application teams.
rjcadvisors.com
Process
QA Meditations
The craft. Practitioner writings on testing, quality, and the knowledge that doesn't run itself.
qameditations.com
Tools
The Automation Expert
The technology. A practitioner's annotated guide to what works, what doesn't, and why.
theautomation.expert
Philosophy

Domain knowledge is the appreciating asset.
Everything else is infrastructure in service of that.

The case against vendor tools is familiar: lock-in, limited flexibility, maintenance cost. But code has lock-in too. Not vendor lock-in. People and pattern lock-in. Poorly documented frameworks. Copy-paste code no one owns. A developer who understands the tool... but not the business.

When the tooling drives the thinking, the domain becomes a constraint. Testing becomes a development effort. At some point, you're no longer validating the business. You're maintaining an ecosystem.

"That works when the problem is technical. It breaks when the risk is domain-driven."

— Richard Cavallaro, The Automation Expert

The tools listed here are not recommendations in the abstract. They are a practitioner's curated assessment — each one evaluated against the domains where it actually performs, with honest commentary on where it doesn't.

The Stack

Vetted. Practiced. Opinionated.

Each tool evaluated from the practitioner's seat — not the vendor's slide deck.

AI-Native Platform · Class 1
Functionize
Intelligent Test Automation · Self-Healing Scripts

Functionize uses a proprietary AI model — not a wrapper around a general-purpose LLM — to build, execute, and self-heal tests. The architectural distinction matters: the model was trained on testing data, not general text. For teams with high UI churn or large regression suites, this is where the ROI conversation starts.

AI-Native Platform · Class 1
testRigor
Plain English Test Automation · Cross-Platform

testRigor lets practitioners write tests in plain English — no XPath, no selectors, no code. The business analyst can read the test. The QA lead can write it. The result is a suite that domain experts can own and maintain without becoming developers. For Oracle and ERP environments where business rules are complex and turnover is real, this changes the sustainability equation.

Finished COTS · Class 2
SmartBear
TestComplete · ReadyAPI · Zephyr

SmartBear's portfolio covers the breadth of enterprise testing — from UI automation with TestComplete to API testing with ReadyAPI to test management with Zephyr. TestComplete in particular has a long track record in enterprise Windows and web applications where stability and IDE-like tooling matter more than AI novelty. Deep practitioner experience with the full suite.

Finished COTS · Class 2
Tricentis
Tosca · NeoLoad · Testim

Tricentis anchors enterprise-scale QA programs where governance, traceability, and SAP/Oracle integration are non-negotiable. Tosca's model-based approach reduces script maintenance at scale. NeoLoad handles serious performance engineering. For organizations running regulated workloads or complex ERP migrations, Tricentis is in the short conversation.

Deep Specialist
Oracle-Native · Legacy Expertise
Oracle OATS
Oracle Application Testing Suite · EBS · Fusion

Oracle Application Testing Suite is a shrinking category with a very short list of practitioners who actually know it. Built specifically for Oracle E-Business Suite and Fusion environments, OATS captures and validates Oracle Forms, HTML, and Flex interfaces at a depth no generic tool matches. If your organization is running OATS — or evaluating whether to migrate off it — this is a practitioner conversation, not a vendor one.

Current Research

The New Beautiful Testing
White Paper

Craft knowledge from fifteen years of practitioner testing is encoded in the weights of every large language model. The question isn't whether AI changes QA. It's whether the humans directing it know enough to make the difference. The New Beautiful Testing examines what that looks like in practice.