Accessibility gets pitched two ways, as charity or as lawsuit insurance, and both pitches undersell it. Accessible builds are simply better engineering, and they sell to more people. That is the whole case, and it is sufficient. Everything below is just the receipts.
The curb-cut effect
Sidewalk ramps were built for wheelchairs and promptly used by everyone with a stroller, a suitcase, or a delivery cart. Software works identically. Real contrast ratios help the customer reading a cracked phone in direct sunlight. Keyboard navigation helps the power user who never touches a mouse. Captions help the parent watching your demo video with a sleeping kid on their chest. Labeled forms help every autofill on earth.
Build for the edge and the middle gets better; it never works the other way around. This is why accessibility improvements keep showing up in conversion data: faster forms, clearer flows, fewer abandoned carts. NIST’s usability research has been making the same point for years: interfaces that work under stress work better for everyone.
Accessibility is also just more market
Roughly one in four American adults lives with a disability. They run companies, sign purchase orders, and buy things at the same rate as everyone else, right up until a checkout flow locks them out. An inaccessible site is a store with the register closed for a quarter of the town.
And unlike most market-expansion projects, opening that register costs almost nothing when done early. Add search engines to the list too: the same semantic structure that helps a screen reader parse your page helps a crawler rank it. Accessibility and SEO are the same homework, graded twice.
There is a compliance floor as well, and it keeps rising. ADA web lawsuits climb every year, and settlements cost more than doing the work would have. But treat legal as the floor, not the ceiling: a site that merely dodges lawsuits can still lose the customer, and the customer was the point.
What building accessibility in looks like
Semantic HTML before div soup. Focus states that survive the design pass. Contrast checked at the palette stage, when fixing it is free. Forms with real labels, images with real alt text, and a keyboard walk-through as a standard QA step alongside the browser matrix.
On our builds WCAG is the default, not a line item, because retrofit accessibility is expensive archaeology while built-in accessibility is nearly free. It also fits how we work: we ship small and often, so an accessibility miss gets caught in a two-day-old change instead of excavated from a two-year-old one.
Tooling helps but does not absolve. Automated scanners catch maybe a third of real issues; the rest need a human tabbing through the flow and listening to a screen reader hit the checkout. Put that hour in every sprint. It is the cheapest user research you will ever run.
Budget honestly and it stays cheap. On a new build, doing this from day one adds low single-digit percent to the timeline. Retrofitting the same site two years later can eat a whole quarter, because now every fix fights the layout, the plugins, and everyone’s habits. The best time to care was the kickoff; the second-best time is the next sprint.
The tell
Ask any web vendor how they handle accessibility. If the answer involves an overlay widget bolted on after launch, keep shopping. Overlays paper over symptoms, break about as often as they help, and have become a reliable signal that nobody checked the actual work.
The good shops cannot answer the question briefly, because for them it is not a feature they add at the end. It is how the work is done: in the markup, in the design review, in QA, every sprint, on every project, without anyone filing a ticket for it.
Accessibility is a feature, not a favor. Features get specced, built, tested, and maintained. Favors get skipped when the deadline tightens. Decide which one it is at your shop, and understand that your customers, all of them, can already tell the difference.
