Is a store liable for its chat widget's barriers?

Yes, in practical terms. The widget is part of the store's public facing service. Vendor contracts should require conformance and fixes, but a shopper's complaint and a regulator's attention land on the store first.

What is the quickest chat widget check?

On a phone, go to checkout and see whether the bubble covers any checkout control, then open and close the panel using only a keyboard or switch device. Those two passes catch the failures that block purchases.

Should chat be removed from checkout pages?

Often, yes. Suppressing the widget during payment removes the overlap and focus risks entirely. Provide a clearly named alternative contact method in the checkout help section instead.

Does the chat button need more than an icon?

It needs an accessible name describing the action, such as chat with support. An unlabelled icon button is announced as just button, which tells a screen reader user nothing about what will happen.

Testing Contrast in Sale Countdown Timers and Urgency Badges

November 2, 2026

Countdown timers ship in brand colours at small sizes over busy photos, and fail WCAG contrast quietly until a sale. How to measure them, including the states after the timer ends.

Countdown timers are designed to be glanced at, which makes their contrast failure almost poetic. Light numerals on a brand red bar, thin digits over a campaign photograph, a gold ends soon badge on white at eleven pixels. A sighted tester squints and approves it. Measured against WCAG 1.4.3, normal text needs 4.5 to 1 against its background, and a remarkable share of urgency furniture sits nearer 2.5 to 1, because nobody measured it: the timer came from an app, in brand colours, three sales ago.

The timer also changes. Digits tick, the final hour turns red, the sale ends and the bar becomes a sale extended message in different colours again. Each state is a separate contrast measurement, and the state most likely to fail, the urgent red on red at the end, is the one nobody screenshotted in the design review.

Measure every state, not the screenshot

Capture the timer at four moments: sale running normally, under one hour, sale ended, and if the app has one, the hovered or focused state of any button in the bar. Measure text against its immediate background, and where text sits over photography, measure against the lightest part of the image behind the words, not the average colour a picker suggests. Photography is where timers go to become illegible politely.

Include the non-text check. WCAG 1.4.11 asks for 3 to 1 on the meaningful parts of graphics and controls. A timer icon, a progress bar showing stock or time remaining, and the boundary of a dismiss button all count if a shopper needs them to understand the offer. A progress bar you cannot see is not neutral decoration; for a low vision shopper it is a missing fact about the deal.

The moving text problem nobody logs

Ticking digits raise a second criterion that rarely makes the report: 2.2.2 Pause, Stop, Hide. Content that updates automatically needs a way to pause or hide it unless it is essential, and timers sit in a grey area that teams resolve by ignoring it. A second by second timer is also a distraction and, for some shoppers with attention or cognitive disabilities, a genuine barrier to finishing checkout under manufactured pressure.

You do not have to kill the timer to handle this honestly. Updating once a minute instead of every second, respecting the reduce motion setting where the platform exposes it, and never putting the only copy of the sale end date inside the moving element all reduce the harm. The end date and time should exist as plain static text somewhere on the page regardless. A deadline a shopper cannot re-read is a deadline designed to pressure, not inform, and that distinction is getting harder to defend.

Put urgency furniture on the sale checklist

Sales launch under time pressure, which is exactly why the check has to be a checklist and not a judgement call. Before the campaign goes live: timer text measured in all states, end date present as static text, dismiss control labelled and reachable by keyboard, bar does not cover focused content when it sticks. Four lines, five minutes, and the evidence is a screenshot set with ratios written on it.

Keep the ratios in the campaign file. When a complaint arrives six weeks later about a sale nobody can reconstruct, that file is the difference between showing your working and starting from a blank page. Contrast work is cheap to prove at the time and almost impossible to prove afterwards.

Frequently asked questions

What contrast ratio do countdown digits need?

If the digits are normal sized text, 4.5 to 1 against their background under WCAG 1.4.3. Large text, at least 24 pixels or 19 pixels bold, needs 3 to 1. Most timer digits are small, so plan on 4.5 to 1 and measure in every colour state the timer uses.

How do you measure text over a photo background?

Against the lightest area directly behind the text, because that is the worst case the reader actually faces in that spot. If the ratio only passes against the photo's average or darkest tone, it does not pass where the words sit. A solid backing panel behind the timer text is the simplest fix.

Does a ticking timer violate Pause, Stop, Hide?

Automatically updating content falls under WCAG 2.2.2 unless it is essential. Teams argue timers are essential to the offer, but the safer pattern is calmer updates, respect for reduce motion settings, and the deadline also written as static text. That combination is defensible in a way a flashing second counter is not.

Who should own this check, design or compliance?

Whoever launches the campaign, with the ratios recorded. Design owns the colours, the app supplies the component, and compliance owns the checklist that forces the measurement before launch rather than after a complaint. In practice the checklist matters more than the org chart.