Better website enquiries
Can customers use your enquiry form without a mouse?
Try reaching and completing your enquiry form using only a keyboard. Can you see where you are, understand each field and correct a mistake without starting again? These simple observations can reveal barriers that a good-looking page hides. For service businesses in Stoke-on-Trent and beyond, they provide a useful starting point for a focused website repair. Passing them does not establish that a website is fully accessible.
Choose one enquiry journey and a safe place to test
Start on a service page and follow the route to requesting a quote or consultation. Include the menu, cookie panel and any embedded booking tool that a customer encounters. A usable form is little help if someone cannot reach it.
Write down the page address, browser, device, cookie choice and date. You can inspect navigation, labels and zoom on the public site without sending a message. Use a preview with test delivery disabled for error, failure and success checks. If testing on a live website, agree the submission and receiving inbox with the owner first; do not enter real customer details or repeatedly send enquiries to find out what happens.
Keep a short record for each observation: the step attempted, what happened, what should happen, a screenshot or exact message, and whether you could continue. Mark untested states as unknown. A repeatable record gives your developer something more useful than “the contact page is awkward”.
Put the mouse aside and follow the visible focus
Use Tab to move forwards through links and controls, and Shift with Tab to move backwards. Look for the highlight, outline or other visible indicator showing the active control. Follow it from the service page through to each form field. You should be able to tell where you are without guessing.
Use Enter to activate links; ordinary buttons generally respond to Enter or Space. A standard dropdown usually uses arrow keys to change options. Check that the actual control works: merely reaching it with Tab is not enough. Stop before activating the final send button outside the agreed test environment.
Record skipped fields, invisible focus, a date picker that needs a mouse, or a panel you can enter but cannot leave. A fixed header or cookie panel can also hide the active control. W3C guidance explains both keyboard operation and visible focus; a developer should investigate the underlying component rather than simply adding more tab stops.
Check labels before and after typing
Can you still see what a field asks for after entering text? Faint example text inside a box can disappear as you type. An enquiry form should make the purpose of ordinary fields clear with persistent labels, and explain which answers are required before the customer attempts to send.
Ask whether each label is connected to its field in the page code. A screen reader, which presents content through speech or braille, needs that relationship too. A visible word above a box does not prove it is connected. W3C shows how labels and controls can be associated; ask your developer to verify the relationship, not just its appearance.
Use instructions that match the actual business need. If a postcode is needed to check coverage, say so. If a phone number is optional, make that clear. Avoid asking for an exact project budget or appointment time unless the customer can reasonably answer and the business needs it at this stage.
Enlarge the page and check what disappears
On desktop, use the browser zoom control to enlarge the page to 200%. Can you still read the instructions, find every field and reach the buttons? Look for text cropped inside boxes, overlapping labels, and fixed panels covering the form. Then check a narrow phone screen, including the area above the on-screen keyboard.
A 200% check is an early observation, not the complete accessibility requirement. W3C’s reflow guidance describes keeping ordinary vertically scrolling content usable at a width equivalent to 320 CSS pixels, such as a 1280-pixel-wide viewport at 400% zoom. Ask a developer to check that layout as well. A form should not require repeated sideways scrolling to read each instruction.
Record the browser and zoom setting with the screenshot. “The button disappears at 200% in this browser” is a reproducible repair request. Restore normal zoom afterwards so the next comparison starts from the same conditions.
Make a mistake and check that you can recover
In the agreed test environment, leave a required field blank or enter an incomplete email address. Try to send once. Does the response identify the field and explain how to fix it? A red border alone cannot tell everyone what went wrong. Check with the keyboard that you can reach the problem, correct it and continue without retyping unrelated answers.
Browser-built validation messages and custom website errors behave differently. Ask your developer to check both field-level feedback and the response to a server failure. Include screen-reader checks of labels, required information, errors and confirmation: seeing text appear does not establish that an assistive tool announces it.
Check a successful test separately. A clear confirmation should explain what was received and what happens next; it should not promise a confirmed booking if the business has only received a request. Keep the on-screen confirmation, server acceptance and actual inbox receipt as separate findings. Do not press Send repeatedly when the result is uncertain.
Worked example: a postcode error that blocks a quote
Hypothetical example: a landscaping business asks for a postcode to check its service area. In a preview test, a customer enters a postcode with a space. The form rejects it, shows only a red outline and clears the job description. The customer now has three problems: an unexplained rule, an error that relies on colour and lost work. This is an illustrative scenario, not a Zero Labs client result.
The repair request is specific: accept the postcode formats the business intends to support, explain any genuine restriction in text, connect the error to the postcode field, and preserve unrelated answers. Retest the same entry with keyboard navigation and the agreed assistive tools. Also try an empty postcode and an obviously incomplete one. Accepting the original example alone does not prove that validation is correct.
If the actual issue is that the form stops working after cookies are rejected, investigate that dependency separately using our consent-state guide below. A keyboard repair and a delivery repair may involve different parts of the website.
Copy this brief when asking for a repair
Page and journey: [exact URL and the service-page link used to reach it]. Test environment: [preview or authorised live test]. Browser, device, zoom and cookie choice: [details]. Steps: [keys pressed or answers entered]. Observed result: [exact error or obstruction]. Expected result: [what the customer needs to complete]. Evidence: [screenshot with private details removed].
Requested scope: reproduce the fault, explain its cause and quote for the affected component. Acceptance: repeat the same steps successfully, keep unrelated answers after correction, and check keyboard operation, labels, zoom, error recovery and the agreed test confirmation. Include other forms using the same component. Record screen-reader and delivery checks separately, with the tools used and any states still untested.
Start with the part preventing a customer from progressing. A form label, focus style or validation rule can sometimes be repaired without rebuilding the site. An embedded third-party tool may require a vendor change or a different integration; ask who controls the component before agreeing the work.
Know when the first check needs a deeper review
These checks help you find and describe obvious barriers. They do not cover every disability, browser, assistive technology or accessibility criterion. W3C’s introductory checks make the same distinction: a page can pass a first review and still contain significant barriers. Arrange a broader assessment, including relevant assistive technology and disabled-user input, when you need dependable assurance.
Zero Labs is based in Stoke-on-Trent and works with businesses across the UK. If an enquiry journey is difficult to use, share the public page and the step that fails. We can discuss a focused website review, establish which parts need deeper testing and agree the repair scope before considering a larger redesign.
Take the next step.
Explore the related services and guides to plan your next step.