Blog
Practical guides on writing briefs, reviewing submissions, and doing QA that ships.
Practical guides for developers and testers. Step-by-step product how-tos live in Help. Product overviews: QA Testing and Usage Learnings.
Latest
Screen recordings that actually prove a bug
Timestamp + marked moment + form field — not a 20-minute tour. Reviewable evidence beats narration every time.
wp-admin UX failures that look like “the plugin works”
It saved; nobody could find the control. That’s still a fail. Runtime success without findability is not a ship gate you can defend.
UserTesting vs pay-per-test QA for indie developers
Different jobs: moderated research vs reviewable ship evidence. Pick the instrument that matches the decision — not the brand with the bigger panel.
How to price a first paid QA run without overbuying
Start with a small smoke path; buy another round after you fix. The first purchase is a learning loop, not a one-shot certificate.
Uninstall leftovers the WordPress.org review still misses
Deactivation is not uninstall. If your plugin leaves data behind without an explicit keep-data choice, users discover it the hard way.
Why testers bounce: logins, 2FA, and missing credentials
The mission never starts. Login is broken, invite never arrived, or 2FA expects your personal device. Access friction is the silent killer of structured QA.
Custom report forms beat “tell us what you think”
Same questions → comparable submissions. That is what lets you approve, reject, or see a pattern — not a pile of unrelated opinions.
How to brief testers to find plugin conflicts
Name the other plugins and the failure you care about — don’t say “try stuff.” Conflict QA only works when every tester attempts the same stack and the same path.
Playwright is green. Why you still need a human path
CI proves contracts. A cold human proves finishability. You need both — not because automation failed, but because it answers a different question.
User testing and ship QA are different jobs
Neither is “better.” They answer different questions. Pick the one that matches what you need to decide — then write the brief that way.
How many testers you need before you can ship
Pattern over sample size theater. Two or three cold runs on the same brief usually beat a dozen vague sessions — then stop, fix, or ship.
PHP version fatals Plugin Check will never catch
Your zip passes Plugin Check on PHP 8.2. A host still runs 8.3 with different deprecations — or 7.4 without the extension you assumed. Lint cannot simulate that matrix.
Get paid to test software without wasting the developer’s time
Follow the brief, fill the form, show the stall. That is the whole job — not a tour, not a redesign pitch, not twenty minutes of “I clicked around.”
A ship checklist for a one-person team
Scripts, then one human path you can defend — not “we clicked around.” That is enough process for a solo ship without pretending you have a QA department.
Isolated WordPress sandboxes vs handing out your staging login
Shared staging leaks state. A fresh install per tester is the point — not a nicer way to paste the same password into Slack.
When session replay is enough (and when it isn’t)
Replay explains what happened. A structured test asks someone to finish a path. Those are different jobs — and mixing them is how teams watch hours of video and still cannot ship.
The testing gap between big companies and everyone else
Enterprises often treat testing as a department. Smaller teams treat it as a luxury. The difference shows up in confidence, first-run friction, and how fast you can iterate without learning from production.
Structured QA vs an open beta
Open beta is a megaphone. A structured test is a ship question: one mission, a form, and evidence you can act on.
How to test a WordPress plugin with real users (no staging)
Sharing your staging site with testers leaks state, credentials, and other people’s work. A fresh install per person is the cleaner gate.
What a good tester submission actually looks like
A strong submission is evidence a developer can act on in minutes — not a personal essay about the product.
What Plugin Check can’t see in a real wp-admin
A green Plugin Check run is necessary. It is not evidence that a human can use your plugin. wp-admin is where that gap shows up.
How indie developers should run a paid QA test
You do not need a research lab. You need one named risk, a few paid humans, and evidence you can ship or stop on.
Why user research costs so much — and what lean looks like
The sticker price is rarely “insights.” It is recruiting, skilled time, synthesis, and process built for big studies — even when you only needed a ship decision.
The confidence human testing builds — and why AI can’t replace it
AI can draft, summarize, and suggest. It cannot stand in for a person who tried the path and left a trail you can defend.
Brief, high-signal tests beat open-ended ones
Most shipping teams do not need a sprawling study. They need a clear mission, a short form, and enough humans to see the pattern.
What human testing still reveals
Scripts catch regressions. People catch confusion, judgment calls, and “this works but nobody would finish it.”