Compliance & Lawsuits

What to Do When You Receive an ADA Website Demand Letter

(and why adding an accessibility widget will make it worse)

Last updated May 24, 2026

An ADA website demand letter is a pre-litigation notice from a plaintiff or their counsel alleging that your website fails to comply with Title III of the Americans with Disabilities Act and demanding remediation, a settlement payment, or both. It is almost always a precursor to a federal lawsuit if ignored. The window to act is usually counted in days, not weeks, and the choices you make in the first forty-eight hours shape the entire legal exposure that follows.

The short, honest answer: forward the letter to qualified counsel within twenty-four to forty-eight hours, do not install an accessibility widget or overlay in an attempt to settle, begin a documented automated and manual audit of the site, and prepare to communicate a good-faith remediation timeline through counsel. Every one of those steps creates the kind of contemporaneous record that plaintiffs' firms and the courts actually credit. None of them requires you to admit liability, and none of them requires you to panic.

The single most common mistake is installing an accessibility widget after receiving a letter. The Federal Trade Commission has now formally rejected the marketing claims those products are built on. On April 22, 2025, the FTC approved a final consent order requiring accessiBe to pay $1 million, barring accessiBe from representing that its automated products can make any website WCAG-compliant or can ensure continued compliance with WCAG over time, unless it has the evidence to support such claims (FTC press release). Per UsableNet's 2024 Year-End Report, twenty-five percent of 2024 ADA web-accessibility lawsuits cited an accessibility widget or overlay as a barrier rather than a solution. Installing one mid-case is, in practice, evidence against you.

The volume is real. Per Seyfarth Shaw's 2025 lawsuit tracker, there were 3,117 federal ADA Title III website-accessibility cases filed in 2025, up 27% YoY (Seyfarth ADA Title III blog). Per UsableNet's 2024 Year-End Report, 67% of ADA website lawsuits targeted companies with less than $25 million in annual revenue. If you are a small or mid-market ecommerce operator, the demand letter in your inbox is part of a much larger pattern, and there is a well-understood playbook for responding to it.

The 48-hour playbook

The objective of the first two days is not to fix accessibility. The objective is to put yourself in a defensible position before counsel opens a dialog with the plaintiff. That means routing the matter to someone who can speak for you legally, freezing the evidentiary state of the site, and beginning a documented audit whose timestamps will later show good-faith engagement. Each of the steps below should be calendared and assigned to a named owner.

Hour 0–4: Forward to counsel and freeze undocumented site changes

Send the original demand letter to your general counsel, outside counsel, or, if you have neither, to an attorney who handles ADA Title III website matters. Many small businesses do not have a standing relationship with such counsel; a referral from a state bar ADA practice group or from your insurance carrier is appropriate here. Most demand letters allow ten to thirty days before litigation, but some are shorter, and the response deadline is generally calculated from the date of receipt. Note that date in writing.

At the same hour, pause all undocumented production changes to the website. That does not mean freezing the business — planned deploys, security patches, and the like continue. It means that any unscripted attempt by a well-meaning team member to remove an image here or tweak a button color there must stop until a documented remediation track is open. If a fix later goes in, you want it attached to a ticket, a commit hash, and a deploy log, not lost in the shuffle. This is also the moment to preserve a full crawl of the site as it existed when the letter arrived: rendered HTML, a sitemap, and a screenshot set are sufficient.

Hour 4–24: Begin a documented audit (combined automated + manual)

Run a credible automated WCAG scan against your highest-traffic pages and the user flows that matter to your business: home, search and category pages, product detail, cart, checkout, account, and any page named or screenshotted in the demand letter. Save the raw output with timestamps. Per Deque's Automated Accessibility Coverage Report, Deque reported automated tests identified 57.38% of issue instances by volume in its dataset, based on 2,000+ audits across 13,000+ pages and ~300K issues (Deque coverage report). That is a strong floor, not a ceiling. The other half — focus order, keyboard traps, ARIA misuse, screen-reader intelligibility, error recovery in forms — requires human-driven testing.

Pair the automated pass with manual review using a real keyboard (no mouse), a screen reader (VoiceOver on macOS, NVDA on Windows), and high-zoom and high-contrast settings. The point of the first-day audit is not to finish remediation. It is to produce a dated artifact that says: here is what we found, here is who looked at it, here is the date. See what automated scanners miss for the categories of issues your audit must cover by hand.

Day 1–2: Inventory and triage findings

By the end of day two you should have a single spreadsheet, ticket board, or audit document that lists every confirmed finding, the URL where it appears, the WCAG success criterion it implicates, and a severity classification. The triage order is reliable: anything that blocks a screen-reader user from completing a purchase or a government services interaction is at the top; keyboard traps, unlabeled form fields, and missing or wrong alt text for functional images come next; everything else cascades from there.

A practical heuristic: if a finding sits on a page that is in the critical path from landing to conversion, treat it as urgent. If a finding sits on a deep blog post from 2019, it is real, but it is not the page the plaintiff is going to demonstrate to a judge. The inventory document is also the foundation for the timeline you will communicate next; if it is honest and complete, the timeline writes itself.

Day 2–7: Communicate a good-faith remediation timeline to opposing counsel

Through your attorney, respond in writing. A typical good-faith response acknowledges receipt without admitting liability, confirms that an audit is underway, identifies the standard you are remediating toward (WCAG 2.1 AA is the conventional reference; many businesses are moving to WCAG 2.2 AA), commits to a remediation schedule with intermediate milestones, and offers a path for the plaintiff's counsel to verify progress. The exact language is counsel's job. The substance — that you have an audit, a plan, and a schedule — is yours.

Why widgets / overlays fail (and worsen outcomes)

The temptation, when a demand letter arrives, is to reach for the fastest visible fix. Overlay vendors have spent years marketing themselves as exactly that fix: paste a one-line script, AI-powered remediation, settle the case. The problem is that the regulator who polices that marketing has now publicly rejected it, and the plaintiffs' bar treats the presence of an overlay as a live target, not a defense.

The mechanical reason is straightforward. Overlays run in the browser and attempt to patch accessibility metadata after the page has rendered. They cannot reliably fix structural issues like incorrect heading hierarchy, missing labels in dynamic forms, keyboard traps in third-party widgets, or focus management in single-page applications. They frequently break assistive technology that users have already configured to their own preferences — a NVDA or JAWS user with a tuned configuration does not want a third-party script renaming controls underneath them.

The regulatory reason is now on the record. On April 22, 2025, the FTC approved a final consent order requiring accessiBe to pay $1 million, barring accessiBe from representing that its automated products can make any website WCAG-compliant or can ensure continued compliance with WCAG over time, unless it has the evidence to support such claims. The order specifically targets the marketing claims overlay vendors have relied on; the underlying mechanics that produced those claims have not changed.

The litigation reason is the most direct. Per UsableNet's 2024 Year-End Report (as cited by TestParty.ai), 67% of ADA website lawsuits targeted companies with less than $25 million in annual revenue, and 25% of 2024 ADA web-accessibility lawsuits cited an accessibility widget or overlay as a barrier rather than a solution. In practice, that means one in four cases now treat the overlay itself as evidence. If you install one after a demand letter, you have done two things: paid for a product that does not actually fix the underlying barriers, and created a fact pattern that the plaintiff's counsel will read as a pretextual quick fix.

The honest alternative is documented, source-code remediation. For the full mechanical breakdown of why overlays underperform on each WCAG category, see why accessibility overlays fail.

Where lawsuits are filed and who files them

Federal ADA Title III website-accessibility filings are concentrated in a small number of jurisdictions and a small number of industries. That concentration matters when you are evaluating how at-risk you are and what venue you may end up litigating in.

Per Seyfarth Shaw's 2025 lawsuit tracker, there were 3,117 federal ADA Title III website-accessibility cases filed in 2025, up 27% YoY. New York's federal courts have historically led the country in this category; Florida has grown sharply, with federal website-accessibility filings rising from 470 in 2024 to 961 in 2025 (Seyfarth: 2025 federal filings bounce back). Concurrent state-court filings, particularly under New York's state human rights statutes, are not counted in those federal numbers and represent additional exposure.

The industry concentration is even sharper. Per TestParty's analysis of H1 2025 lawsuit data, ecommerce sites accounted for 69–77% of all H1 2025 ADA web-accessibility filings (TestParty H1 2025 analysis). If you run a Shopify, WooCommerce, BigCommerce, or custom-storefront ecommerce business serving the United States public, you are in the most heavily targeted segment of the most heavily targeted statute.

The plaintiff pool itself is shifting. Per a Seyfarth Shaw analysis published October 2025, pro se ADA Title III filings grew roughly 40% YoY in 2025 — Seyfarth attributes the surge to plaintiffs using AI tools like ChatGPT, Gemini, and Copilot to draft complaints (Seyfarth: pro se surge powered by AI). A pro se complaint, drafted with the help of a chatbot, is still a real lawsuit; the docket sheet does not distinguish. What changes is that opposing counsel may not have the procedural sophistication of a repeat plaintiff firm, the boilerplate may be weaker, and the willingness to settle quickly is sometimes higher — but the cost of ignoring the complaint is the same.

Repeat-defendant risk is the other pattern to keep in mind. Per UsableNet's 2026 annual report: in federal court alone, 46 percent of 2025 cases involved repeat defendants — 1,427 cases targeted companies that had already faced an ADA web accessibility claim (UsableNet 2026 annual report). Settling a single case without fixing the underlying barriers materially raises the probability of a second claim. The work you do after the letter has to actually address the violations, not just the lawsuit.

Settlement ranges and cost realism

There is no single public dollar figure for ADA website settlements, and any vendor that quotes you a precise average is guessing. Settlements are confidential by default, ranges vary widely by jurisdiction and plaintiff firm, and the cost depends heavily on how early in the process you engage and how credible your remediation plan is.

What is publicly visible: Seyfarth Shaw and UsableNet both track the volume and trajectory of cases, and both have written that the larger cost categories are typically defense fees, plaintiff's attorney fees, the remediation work itself, and ongoing monitoring requirements baked into consent decrees or settlement agreements. Cases that move to litigation almost always cost more than cases settled pre-suit. Cases brought against repeat defendants — the 1,427 of them in 2025 — tend to escalate faster because the plaintiff bar treats a prior settlement as evidence of notice.

The honest framing for any executive decision is: total cost equals the settlement payment plus the legal fees plus the remediation cost you would have incurred anyway. Skipping the remediation step does not reduce that total; it shifts the cost into the next case. Treat the demand letter as the budget event that funds the remediation work you have been deferring, and treat the budget as if a second plaintiff is already drafting their complaint — because the data says they often are.

What “defensible remediation” looks like

The word courts and plaintiffs' counsel actually credit is documented. The strongest pre-litigation posture and the strongest mid-litigation posture share the same skeleton: a timestamped audit, code-level changes attached to commit history, a published accessibility statement that names the standard, an ongoing monitoring cadence, and recurring scans that produce dated evidence of regression catches. The point is not to produce a paper trail for its own sake; it is to make the good-faith effort legible to a third party who was not in the room.

• Documented audit with timestamps. Automated scan results saved with the date and tool version, manual review notes signed by the person who performed them, and a clear delineation between confirmed findings and items that require further investigation.
• Source-code changes with commit history. Every fix that materially affects accessibility lives in a commit, with a ticket reference and a clear message describing what was changed and why. Browser-injected fixes via an overlay do not satisfy this; they live and die with the script and leave no durable record.
• Published accessibility statement. A page at a stable URL (commonly /accessibility) that names the standard you are conforming to (WCAG 2.1 AA or 2.2 AA), lists known limitations, provides a contact for accessibility reports, and is itself accessible. The statement is not a substitute for fixes; it is a record of intent.
• Ongoing monitoring. Accessibility regressions arrive in pull requests, third-party embeds, and content edits. A monitoring system that scans on a schedule and alerts on regression is the difference between getting better and ending up as a repeat defendant.
• Recurring scan schedule. A cadence (weekly or monthly is typical) of automated scans against the same critical pages, with results archived. The archive itself is the artifact: a year of dated scans is hard for any plaintiff to characterize as bad-faith neglect.

A note on ADA Title II (state/local government)

Most demand letters target private businesses under Title III, but state and local government entities operate under Title II, and the regulatory calendar there has just moved. DOJ's April 2026 Interim Final Rule extended Title II web-accessibility compliance dates to April 26, 2027 (population ≥50,000) and April 26, 2028 (smaller entities and special district governments) (Federal Register: 2026 IFR). The underlying technical standard — WCAG 2.1 AA — did not change; only the deadline did. Public entities that were on a 2026 plan now have a year of additional runway, but the obligation has not been rescinded. See ADA Title II vs Title III for the full framework.

How SweepHound supports demand-letter response

SweepHound is built around the assumption that the audit needs to be defensible, not decorative. We do not promise to end your lawsuit, make your site “fully compliant”, or guarantee any legal outcome — nobody honestly can. What we do is produce the kind of timestamped, source-code-anchored artifacts that a remediation effort needs to be credible.

• Documented audit trail. Every scan is stored with the date, the page set, and the findings; statements you publish are versioned against the scan they correspond to. Counsel can hand the trail to opposing counsel without curation.
• Dual-engine scan. SweepHound runs axe-core alongside a second engine for the higher-conformance plans, so the automated baseline is broader than any single tool produces alone. Engine breakdowns are visible in the scan summary so you can see exactly which rules fired.
• Code-level fixes with commit-ready diffs. For violations where a mechanical fix exists, SweepHound emits an HTML or attribute-level patch you can apply at the source. Where the fix is ambiguous or context-dependent, SweepHound says so rather than producing an unsafe guess.
• Auto-generated accessibility statement. Published at a stable URL, named to the standard you target, and tied to the most recent scan — not a generic template that overstates conformance.
• Ongoing monitoring. Scheduled scans against the same critical pages, with regression alerts when a new deploy introduces a violation. The archive is the artifact.

If a letter is in your inbox right now, start a free scan to put a dated baseline on the record today, and route the letter itself to counsel within the next twenty-four hours. Plans and monitoring cadences are on the pricing page; for an ongoing remediation engagement, the higher tiers include recurring scheduled scans and the second scan engine. When you're ready, create an account and run the first audit in the next hour.