According to DEV Community, a responsive menu can look correct in a narrow mobile viewport while keyboard focus still lands on links that are hidden. The article suggests a short manual test before release, using a narrow viewport and a physical keyboard.
For the closed menu, start at the top of the page and press Tab until the menu control gets focus. Its focus indicator should be visible. Links inside the closed menu should not be reachable by Tab. Pressing Enter on the control should change its expanded state.
For the open menu, tab to the first visible link, then use Shift+Tab to go backwards. If the menu is a modal drawer (one that sits over the page and takes over interaction), Escape should close it and send focus back to the control that opened it. If it is a non-modal disclosure, people should be able to move on to the page content in a logical order.
The article also recommends testing a real task. Pick one destination, such as a sizing guide, open the menu, reach the link, activate it and confirm the page loads. If testers keep opening the wrong menu section, the labels and the site structure behind the navigation may need review.
For bug reports, the article says to record the browser, viewport, starting state, expected focus sequence and the first point where it breaks. After a fix, repeat the same steps with the menu closed and open, and judge modal and non-modal menus by their own rules.
Why it matters
Many menu checks happen only with a mouse or touch in a resized browser window. That can miss problems that only show up for people who navigate with a keyboard, such as focus moving into links they cannot see.
The method in the article is simple and needs no special tools. It also gives teams a repeatable way to describe a failure, which makes fixes easier to verify and stops the same bug returning.
Source: DEV Community. This brief summarises their reporting; read the original for the full detail.