The problem: “it’s broken” is not a bug report
A useful bug report says what is wrong, where, and in which browser at which size. The people best placed to find problems on a website before launch are often the least likely to write that down: a client, a colleague from another team, a tester borrowed for an afternoon. So the report reads “the menu is broken on my phone”, and the first hour goes on finding out which menu, which phone and which page.
Most of a good report is not opinion. It is facts the browser already has. ClickNote collects those so the person reporting only has to point and describe.
How a QA round runs in ClickNote
- Add the snippet to the staging site. One script tag, under 1 KB, idle for anyone who has not opened a review link. The tester sees the real site in their own browser: nothing is loaded through a proxy, which is where staging sites tend to break.
- Invite the testers. Send the review link, or invite people by email with one-tap sign-in. Limit the project to the addresses you list if the site is not public yet.
- Testers click and describe. They choose Something’s broken, Wording, Design or Question, and write what they expected and what happened. Browse mode lets them use the site normally (open the menu, add to the basket, submit the form) until something goes wrong.
- They finish the round. One button says there is nothing more on its way, and you are told.
- You fix and they check. Set a note to In progress, then Done. The tester sees the status on the site and can check the element it is pinned to.
Testers on a project can see each other’s notes, so the same broken link is reported once.
What every report contains
- The page address and a pin on the element.
- The element: CSS selector, XPath, text, a snippet of its HTML, its attributes and styles.
- The environment: browser, operating system, device, viewport and whether the layout was mobile, tablet or desktop.
- A screenshot of the element and what is around it.
- The tester’s words, the type of note and who left it.
None of that is typed by the tester. What anyone has typed into a form on the site is never captured.
Working through the list
The dashboard lists a project’s notes with filters for status, page, reviewer and type, so “everything broken on the checkout that is still open” is one view. Each note has a thread for questions, with @mentions, and an internal note that testers and clients never see. The whole list exports as JSON or Markdown.
New notes can post to Slack, Microsoft Teams, Discord or Google Chat on every plan, and to a webhook of your own on Agency and Scale, which is how they reach an issue tracker. A developer who works with an AI coding agent can hand it the notes directly: see feedback for AI coding agents.
Where it fits in QA and UAT
ClickNote covers the part of testing where a person looks at a site and reports what they find: exploratory checks before launch, a client’s acceptance round, a content check after a migration. It sits beside a checklist, not in place of one. Our website QA checklist lists what to look at, and the UAT template gives the scenarios, the severity levels and the sign-off block that ClickNote does not have.
For testing the signed-in screens of a product, read feedback on web apps, dashboards and admin panels.
What ClickNote does not do
- No console log, network log or error tracking. It records the element and the device, not what the browser logged.
- No video and no session replay. Steps to reproduce are the tester’s own words.
- No automated testing. No scripted tests, no visual regression, no link checker.
- No test cases or test plans, and no severity, priority or assignee on a note. A note has a type and one of three statuses.
- No built-in integration with Jira, Linear, Asana or Trello. Notes go there by webhook, through a tool such as Zapier or Make, or as an export.
- Websites only. Not native mobile apps, not APIs.