Morgan Luce

Earlier in my career, I believed that if I could craft a strong QA Strategy, quality software would naturally follow. But reality taught me otherwise. Under release pressure, we were still confused. Releases still slipped. If a customer removed a feature at the last minute, we weren’t sure whether to rerun full regression — not due to effort, but lack of clarity in decision-making. That’s when I realized a hard truth: A QA Strategy does not run by itself. It sets direction — but doesn’t operate day-to-day delivery. What we were missing was the implementation layer: a Quality System. So, what exactly is a QA System, and how is it different from a QA Strategy? QA Strategy defines intent. QA Systems enforce behavior. A Quality System operationalizes QA Strategy through processes, standards, techniques, tools, and infrastructure — producing measurable quality outcomes. Here are the 6 core components I include when building a Quality System for a software organization: 1. Quality Processes (how quality flows) • SDLC checkpoints + test cycle (plan → execute → report) • Risk identification in refinement/planning • Defect lifecycle & triage • Release readiness & sign-off • Incident response + RCA 2. Standards & Regulations (what must be complied with) • Internal standards & policies • Regulatory/compliance needs (if applicable) • Coding/testing standards + DoD • Documentation & traceability • Acceptance + exit criteria 3. Techniques (how quality is applied) • Risk-based testing & allocation • Shift-left & early validation • Test design (equivalence, boundary, decision tables…) • Exploratory vs scenario-based testing • Automation strategy (what/why/when/where) 4. Tools & Best Practices (mechanisms that enable enforcement) • TMS + requirement-to-test traceability • Automation frameworks (API/UI) • CI/CD integration for fast feedback • Code quality + security scanning (SAST) • Defect tracking, reporting, test evidence 5. Infrastructure & Environments (where execution happens) • QA/staging/pre-prod environments • Branching/versioning + flags + merge/rollback • Test data management • Release pipeline/deployment 6. Measurement & Improvement (how the system learns) • Metrics & thresholds (escape rate, defect trends, confidence signals) • Technical reviews & audits • Automation signal quality (stable vs flaky) • Pattern/trend analysis • Systemic improvements (update standards/process/tools) At scale, a good QA Strategy is necessary — but not sufficient. Only a Quality System turns strategy into consistent execution. Quality is a shared responsibility across the organization.

79

9 Comments

Playwright just released Playwright Agents, and it’s a big deal for the QA community. 🚀 The new agents can plan, generate, and even fix Playwright tests, helping teams spend less time on maintenance and more time on meaningful testing. It’s great to see automation tools evolving in ways that actually make our work faster, smarter, and more reliable!

8

"𝗝𝘂𝘀𝘁 𝗱𝗼 𝗶𝘁" 𝗶𝘀𝗻’𝘁 𝗷𝘂𝘀𝘁 𝗮 𝘀𝗹𝗼𝗴𝗮𝗻. It’s how I learned to influence without authority As a mid-level QA woman in tech. Waiting for permission wasn’t going to cut it. A colleague who was great at this stuff shared the basics: 🔸 Understand what’s really bugging them 🔸 Bring solutions that help the business 🔸 Show up prepared (always) 🔸 Do what you say you’ll do 🔸 Pick your moments to speak up But here’s what really moved the needle: 𝗜 𝘀𝘁𝗮𝗿𝘁𝗲𝗱 𝗯𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗿𝗲𝗹𝗮𝘁𝗶𝗼𝗻𝘀𝗵𝗶𝗽𝘀 𝗯𝗲𝗳𝗼𝗿𝗲 𝗜 𝗻𝗲𝗲𝗱𝗲𝗱 𝘁𝗵𝗲𝗺. Coffee chats with devs. Checking in on product managers’ side projects. Remembering what stressed people out. So when crunch time hit, I wasn’t just “that QA person” I was someone they trusted. And it worked: ✅ I heard about issues early ✅ Devs shifted priorities when I asked ✅ My meetings actually worked because I’d already invested in listening 𝗔𝗻𝗱 𝗵𝗲𝗿𝗲'𝘀 𝘄𝗵𝗲𝗻 𝗶𝘁 𝗮𝗹𝗹 𝗰𝗹𝗶𝗰𝗸𝗲𝗱... A critical customer project. I had to convince leadership to shorten a risky freeze window. They said yes, not because of my job title Because of the trust I'd earned. And because I was clear about the risk and the business impact. 𝗜𝗻𝗳𝗹𝘂𝗲𝗻𝗰𝗲 𝗶𝘀𝗻’𝘁 𝗮𝗯𝗼𝘂𝘁 𝘆𝗼𝘂𝗿 𝘁𝗶𝘁𝗹𝗲. 𝗜𝘁’𝘀 𝗮𝗯𝗼𝘂𝘁 𝘁𝗵𝗲 𝘁𝗿𝘂𝘀𝘁 𝘆𝗼𝘂’𝘃𝗲 𝗯𝘂𝗶𝗹𝘁. People who influence without authority: 1️⃣ 𝗕𝘂𝗶𝗹𝗱 𝗿𝗲𝗮𝗹 𝗿𝗲𝗹𝗮𝘁𝗶𝗼𝗻𝘀𝗵𝗶𝗽𝘀 – They invest in people, not just projects 2️⃣ 𝗟𝗶𝘀𝘁𝗲𝗻 𝗳𝗶𝗿𝘀𝘁 – They understand what matters to others 3️⃣ 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗲 𝗰𝗹𝗲𝗮𝗿𝗹𝘆 – No jargon, just clarity 4️⃣ 𝗣𝗶𝗰𝗸 𝘁𝗵𝗲𝗶𝗿 𝗯𝗮𝘁𝘁𝗹𝗲𝘀 – They know when to push and when to pause They’re not tricks They’re how real leaders show up. Keep showing up this way, and people will start listening. 𝗪𝗮𝗻𝘁 𝘁𝗼 𝗹𝗲𝗮𝗱 𝗯𝗲𝘆𝗼𝗻𝗱 𝘆𝗼𝘂𝗿 𝗷𝗼𝗯 𝗱𝗲𝘀𝗰𝗿𝗶𝗽𝘁𝗶𝗼𝗻? Start with coffee chats - not PowerPoints. Drop a 👍 if you're trying it. Go book that chat this week.

22

23 Comments

You can run Playwright locally or on your own infrastructure for free - don't buy into the scare tactics of "influencers/clickbaiters" around the new monitisation model from Microsoft - MS are actually doing good things in the tool space with Playwright. Use GitHub Actions, GitLab CI, Jenkins, or other CI tools to run tests without Azure. GitHub Actions has a free tier available, GitLab CI has free runners or use it self-hosted. Jenkins can be fully self-managed. Also look into data for yourself - a higher amount of down loads does not mean performing better as people are still doing POCs and comparisons. #Softwaretesting #softwareDevelopment #playwright

19

A flyer in the break room changed how I think about compensation forever - first as an employee, then as a leader. Early in my tech career, I was a few years into a QA automation role. Self-taught. No CS degree. I'd built the team's entire testing framework from scratch. People depended on what I'd created. Then I saw a job posting for my exact role - starting at nearly 30% more than I was making. When I confronted my manager, he didn't deny it: ➡️ Without a traditional degree, I'd been slotted into a lower band ➡️ My raises had been percentage-based, compounding on a low starting point ➡️ "That's just how the system works" They gave me a bump to appease me. Still well below the posted rate. I left within a year. That experience taught me to think differently - about both sides of the equation. As an employee: Your compensation should reflect the outcome you deliver. Your experience, your ability to lead, your technical depth - all of it contributes to that outcome. Know what that outcome is worth in the current market. If there's a gap, that's information. Act on it. As a leader: This is where it gets interesting. When companies hire for roles rather than outcomes, they end up with job descriptions that are disconnected from what they actually need. They lowball candidates because they're matching titles, not value. But when you define the outcome first, what does success actually look like in this position? - everything clarifies: ✔️ You know exactly what you need, not just a role to fill ✔️ You can justify the compensation because you've defined the value ✔️ You attract people who can deliver that outcome, not just fill a seat ✔️ You stop losing talent to competitors who understand this better Compensation isn't personal. It's business for both sides. Employees should know what their outcomes are worth. Companies should know what outcomes they're paying for. When both sides think this way, you get the right people in the right roles at the right price. When they don't, everyone loses. 😎 How do you think about the outcome vs. role distinction - either as an employee or a leader? #Leadership #Compensation #Hiring #CareerGrowth #TechLeadership

4

I started in QA feeling completely lost — to be honest I just jumped straight into Automation so I did not have a clue about what Quality actually meant beyond writing code. Not the "imposter syndrome" kind of lost that everyone talks about. Actually lost. Like, "what do I even study first?" lost. Every roadmap I found was either too basic or assumed I already knew things I didn't. So I wrote the one I wish existed back then. It covers the fundamentals, the different paths you can take (web, API, mobile, performance, AI testing, leadership), the soft skills side of things, and some harsh truths I had to learn the hard way. It's not a course, not a "30 days to QA" thing. Just an honest brain dump of what I think actually matters in this field. Even with AI nowadays, solid fundamentals shape how we think — AI can assist that journey, not drive it. https://lnkd.in/eAjWxv54 If you want more QA stuff, I also post on TikTok: @angry_tester

40

2 Comments

A Senior QA shouldn't just point out the holes in the system. They should be the person capable of closing them—even if it’s temporary. I’ve always struggled with staying inside the "tester" box. To me, the best part of this job is the overlap. I love that my day can start with Business Analysis, move into UX design, and end with writing the actual logic for an internal tool. Missing a task description? I’ll write it. Requirements are fuzzy? I’ll architect the flow. No dashboard for the metrics? I’ll code it myself. I honestly enjoy the "puzzle-solving" aspect of planning a feature from scratch. It’s not just about finding bugs anymore; it’s about making sure the user never encounters a problem in the first place because the logic was solid from day one. Being a QA is a great job. But being a "Troubleshooter" is a mindset. 🛠️ #QualityAssurance #ProductOwnership #Troubleshooting #StartupLife #Fintech #Generalist

8

1 Comment

BDD and AI assisted coded is an awesome combo. Have your user stories include acceptance criteria written in Given/When/Then (aka Gherkin) whenever the change impacts users. You can use an agent, don't write it yourself This stops the uneasy feeling when the work is completed where you think to yourself "really? This does not feel right" (even with excellent unit test coverage) I force my agents to get a pass 100% across lint, unit test, integration, BDD (end to end across components). I means I know it works from a user perspective. Across systems.

9

1 Comment

On Wednesday July 23rd, I’ll be speaking at the National Software Testing Conference about Testing Beyond the Browser: QA in the CTV Landscape Connected TV testing isn’t just a resized browser window. It means underpowered hardware, fragmented platforms, remote controls and sometimes no dev tools in sight. At FX Digital, we run over 250,000 automated tests/month across real devices like Sky, Fire TV, and Samsung. I’ll be sharing: - Challenges faced with TV testing compared to browser-based testing - Balance hands-on testing with smart infrastructure - A glimpse into some of our tools and approaches on how FX Digital scaled automation in a CTV-first world If you work in QA or test automation and want to see what life looks like beyond keyboard and mouse, come say hi! #CTV #QA #TestAutomation #OTT #NationalTestingConference #CTVApps

44

1 Comment

For my QA peeps: They say not to take your work home with you, but that’s easier said than done when you work from home. I find myself constantly asking the hard questions: What happens when I stress test the air fryer? How about load testing the dishwasher? Or the ever-popular task: debugging interpersonal relationships? (Let me tell you, “good catch” is not a commonly heard phrase at home.)

215

38 Comments