Automation CoE Lead
Building enterprise automation capability, reusable frameworks and engineering standards across business units.
Automation CoE leadership
The scope expanded from solving one project’s problems to building capability for many.
I moved into the Automation Center of Excellence, where I led automation for one business unit and worked across multiple projects and QA teams. My responsibilities covered automation strategy, framework architecture, tool selection, standards, team leadership, training, adoption, execution, ROI analysis, vendor evaluation, resource planning and governance.
I had four resources reporting directly to me and partnered with project QA teams to create a shared automation capability that they could eventually own and maintain.
Enterprise automation strategy
Automate the right things, using an architecture that can be reused.
- Defined standard automation tools and framework patterns.
- Evaluated programming languages, commercial products and open-source technologies.
- Established reusable components and shared execution utilities.
- Helped projects move away from fragmented automation approaches.
- Researched emerging automation technologies.
- Evaluated build-vs-buy decisions.
- Defined where automation would provide meaningful ROI.
- Connected standards to portfolio priorities and long-term ownership.
Reusable framework architecture
Project teams focused on what to test instead of rebuilding the infrastructure underneath.
I designed frameworks that individual projects could adapt without requiring every QA team to build its own automation foundation. The architecture combined:
Vendor replacement and cost optimisation
An in-house solution built for speed, reuse and scale.
Parallel execution
Executed multiple UI automation tests in parallel on the same machine, increasing execution capacity without a proportional increase in physical or virtual machines.
Faster test development
Combined an Excel-driven data model with keyword-based components so engineers could create or modify many scenarios through structured test data.
Faster feedback
Designed the execution engine to reduce overall runtime and provide faster feedback than the existing commercial solution.
Cloud execution
Distributed tests across cloud machines so execution capacity could scale beyond a single workstation.
Framework capabilities that reduced repetition
Common problems became platform-level capabilities.
Intelligent popup management
A central mechanism detected and managed different popup scenarios so individual test cases did not each need their own popup-specific logic. This turned a common source of UI instability into a reusable framework capability.
Risk-based test execution
Database-backed reporting and risk information helped determine what should be executed, moving the team toward risk → priority → test selection → execution → result instead of treating every test case equally.
Centralised cloud execution
A central automation hub recognised available execution capacity and pushed work toward cloud machines with the required services installed and running. Results were centralised through database reporting.
Automation onboarding and training
A framework only creates value when teams can adopt and own it.
Project onboarding model
- Introduce automation concepts.
- Demonstrate the framework.
- Train engineers.
- Convert manual scenarios into automation.
- Support initial development and execution.
- Transition ownership back to project teams.
- Continue maintaining the shared framework.
Enterprise QA training
I organised training across JPMorgan business units covering what should and should not be automated, framework concepts, test design, data-driven and keyword-driven testing, reusable components, execution strategies, maintenance and automation ROI.
The goal was to make automation a team capability rather than a specialist skill held by only a few engineers.
Tool evaluation, leadership and portfolio planning
Engineering capability and business value had to be considered together.
Technology evaluation
Compared functional capability, execution speed, parallelism, maintainability, licensing cost, scalability, integration capability, learning curve, extensibility and long-term ownership.
Team leadership
Handled work allocation, technical direction, project prioritisation, mentoring, adoption, delivery tracking, training, resource planning, progress reporting and stakeholder communication.
Automation as an investment
Worked with effort estimation, resource planning, prioritisation, vendor and licensing comparisons, reuse potential, expected regression savings and portfolio-level opportunities. The question became: Where should the organisation invest automation effort to get the greatest engineering and business return?
Measured impact
The strongest outcomes from the shared automation capability.
Automation reuse
Increase in automation reuse through shared frameworks and accelerators.
Regression effort
Reduction in regression effort through faster, reusable execution.
Commercial licensing
Annual licensing cost avoided through the in-house automation solution.
What this chapter added
From solving individual problems to building organisational leverage.
The Automation CoE changed my scope again. Instead of solving a problem for one client or one project, I focused on building reusable engineering capability across projects: frameworks, standards, tooling, training, distributed execution and strategy.
Automation is an engineering discipline that can reduce operational risk, improve delivery speed and make quality more scalable.