Plan a Keyboard-Only Review of a Practice Web Form

Nov 7, 2026

A keyboard-only review of a fictional practice form can show a portfolio reader how you investigate a user journey, describe a problem, and check a proposed improvement. Label the project as fictional, record exactly what you tested, and present your findings as observations rather than a claim that the form conforms to an accessibility standard.

Choose one complete task: navigate to a form, fill it in, submit it with an error, correct the error, and submit again. This narrow scope gives you room to document the sequence and its limitations clearly. It also makes the work sample useful to someone assessing your communication as well as your technical judgment.

What should the practice project demonstrate?

Use a small fictional form with a recognizable purpose, such as signing up for a community workshop. Give it a name field, an email field, a session selector, a consent checkbox, and a submit button. Make the email field required and provide a local success message so the journey can be tested without collecting real personal information.

A strong portfolio entry explains the task, environment, steps, observed behavior, and changes. Include a short screen recording or annotated screenshots if you can show the focus indicator and the result of submission clearly. Describe what the evidence shows: a screenshot of a ring around a button can support an observation about visible focus at that moment, but it cannot establish that every control works.

State whether the form and its issues are hypothetical or whether you built and tested a practice version. Do not describe a fictional organization as a client or imply that someone commissioned the work. When adapting the entry for an application, connect the evidence to the role’s stated tasks; Talyca application resources can provide context for presenting work in a remote-job application.

How do you make the review reproducible?

Record the form version or revision, operating system, browser and version, date, and any relevant browser settings. Begin from a known starting point, such as a freshly loaded page with no form data. Note whether browser validation or custom validation supplies the error, because the behavior you observe may differ between environments.

Put the pointer aside for the journey. Use Tab to move forward and Shift+Tab to move backward. Enter text in fields, use the keyboard controls appropriate to the selector and checkbox, and activate the submit button with the keyboard. Record the actual keys and results rather than assuming every control behaves identically.

Before testing, write down the intended sequence in plain language. Then compare it with what happens. W3C’s WCAG 2.2 guidance on focus order says that, when sequential navigation affects meaning or operation, focus must follow an order that preserves both. That does not require a single universal order: the question is whether a person can understand and operate this form as focus moves through it.

Check whether focus can leave each component as well as enter it. If you appear stuck, record the last control, the keys you tried, and whether you could continue or return. Avoid repeatedly pressing keys until an unexplained escape occurs; a reader needs the sequence that exposed the problem.

Review moment | Question to answer | Evidence to capture

Forward and reverse navigation | Can you reach the controls in an order that makes sense? | Actual focus sequence and any skipped or unexpected stop

Visible focus | Can you tell which control will receive the next action? | Control, visual indicator, and surrounding background

Invalid submission | Can you tell which input needs attention? | Submission key, error text, and where focus remains or moves

Correction and resubmission | Can you return to the field, correct it, and finish? | Return route, retained values, and resulting message

What should you look for as focus moves?

Follow the visible indicator at every stop, including links before the form and controls after it. Note whether it is easy to distinguish from the control’s ordinary border and whether it remains visible while that control has focus. Check against the actual background, including any selected or error state. If focus seems to disappear, continue carefully and record the point at which you can identify it again.

W3C’s Focus Visible guidance explains that a keyboard-operable interface needs a mode in which the focus indicator is visible. Its Keyboard guidance also says that functionality must generally be operable through a keyboard interface without requiring specific timing for individual keystrokes. These are useful references for framing observations, but a short walkthrough alone does not establish conformance.

Pay particular attention to a visual order that differs from the keyboard sequence. For example, if the session selector appears before consent but focus jumps to consent first, describe the mismatch and its effect on the task. A different sequence is not automatically wrong; assess whether it still preserves meaning and lets the person operate the form.

Use descriptive language in the log: “The submit button received focus, but I could not distinguish its focused state from its unfocused state” is more useful than “focus is broken.” If the indicator is present but difficult to see, say that instead. A precise observation gives the next reviewer something to reproduce.

How do you test submission and recovery?

Enter a fictional name, leave the required email blank, select a session, and activate Submit. Record whether submission is blocked, what message appears, and where focus is after the attempt. W3C’s Error Identification guidance says that when an input error is automatically detected, the item in error must be identified and the error described in text. A color change alone is not an adequate description.

From the resulting state, use the keyboard to find the email field. Note whether the error message is discoverable visually, whether it identifies the email field, and whether previously entered values remain. Enter a fictional address, return to Submit, and try again. Record the success message or any remaining obstacle.

A form can display a useful error without moving focus in the particular way you expected. Report the actual route and its practical effect rather than treating one focus movement as the only possible solution. If you cannot determine whether an error is announced to a screen reader, mark that question as untested; a sighted keyboard-only review does not answer it.

What might an original test log look like?

The following is an illustrative log for a fictional workshop form, not a report of a real test. Replace its environment and observations with your own if you carry out the exercise. Keep individual entries small enough that another person can repeat them without guessing what happened.

Step | Action | Illustrative observation | Follow-up

1 | Tab from the start of the form | Focus reaches Name, Email, Session, Consent, then Submit. | Repeat with Shift+Tab to check the reverse route.

2 | Tab to Session | The selector receives focus, but its border looks the same as when unfocused. | Add a distinct focus treatment and retest against the surrounding background.

3 | Leave Email blank and activate Submit | A red border appears around Email; no error text is visible. | Add text identifying the missing email and repeat the submission.

4 | Return to Email, enter a fictional address, and submit | The name and session remain populated; a success message appears. | Record whether the same route remains possible after the change.

How can you explain a before-and-after change?

Before, imagine that the fictional selector has no distinguishable focus treatment and the empty email field receives only a red border after submission. The keyboard user can reach the selector but may not know it is active; after an unsuccessful submission, the form does not explain in text what needs correction.

After, imagine adding a visible outline to the selector and an error message beside Email that says, “Enter an email address.” Repeat the original journey from a fresh page. Record whether the outline is visible when the selector receives focus, whether the text appears after the invalid submission, and whether the user can return to Email and submit successfully. If you have not performed that retest, present these as proposed changes and expected outcomes, not completed results.

Close the portfolio entry with the boundaries of the review: one form, the recorded environment, and the keyboard paths exercised. A fuller accessibility evaluation would need additional checks, including assistive technology and other criteria. That limitation makes the work easier to assess because readers can distinguish what your evidence supports from what remains open.

Sources