Recordstake

West Jordan, Utah

Document accessibility inventory · ADA Title II

Most agencies cannot answer the first question: how many documents are we actually on the hook for?

We count your documents, test them, and label every answer as measured, judged, or not assessed — never blurred together.

  • Measured
  • Judged
  • Not assessed
Get the free scan

One real city, measured end to end

The only library we have measured completely: roughly 2,700 documents, more than 60,000 sheets, every one opened. Figures are rounded and the agency is not named until it chooses to be — nothing here is rounded in a direction that flatters us.

  • 8%

    of sheets are pictures of text, not text

  • 1 in 5

    documents holds at least one such sheet

  • ~50%

    of budget documents contain a scanned page

  • 0 of 2,691

    documents passed an automated check outright

Those scanned sheets are almost always the signed ones — resolutions, executed agreements — and a screen reader cannot read them at all. Text recognition helps, but it is not infallible: on this corpus it once misread a city's own motto while reporting full confidence. That is why a person checks recognised text before we call anything done. Read the full findings, with the method and every caveat beside it.

What we actually do

  1. Measure — free. We crawl what your website already publishes and count it: how many documents, how many pages, what share a validator flags, what share are scanned images. No login, no cost, no obligation.

    How the free scan works.

  2. Judge — $2,000. The Document Inventory is the same crawl, but a person reviews every document: which specific files carry which fault, and which side of an exception each one falls on. Delivered in five business days.

    What it costs, and why.

  3. Fix — quoted per engagement. Remediation writes the fix into the document itself: real structure, real headings, tables a screen reader can actually follow. We built this engine ourselves after testing showed an off-the-shelf tool can tag a document correctly and still leave it unusable by ear. It is real and it runs against real documents — and today, on its own, it does not get any document fully clean automatically. A person finishes what the machine cannot, and we are building toward needing less of that, on a roadmap we track in the open rather than promise quietly.

    Why this is hard, and what it actually takes.

Read the specific answer

The deadline, stated plainly

  • 26 Apr 2027

    public entities of 50,000 people or more

  • 26 Apr 2028

    smaller entities and special districts

The technical standard is WCAG 2.1 Level AA. Those are the extended dates — DOJ pushed both back by a year, effective 20 April 2026. If a vendor quotes you the old dates, they have not read the rule since it changed. Which date applies to your agency, and where both come from.

The rule does not cover everything you have ever published — it excepts archived material and preexisting documents no longer used to reach a service. Which of your documents those exceptions remove is the single biggest factor in what this costs you. The two exceptions in full, and why being old does not help you.

An exception is not a permanent pass. An entity still owes effective communication under the ADA — if a person asks for an excepted document in a form they can use, it generally has to be provided. Any vendor who tells you otherwise is selling you a smaller number, not a smaller obligation.

We are not attorneys and this is not legal advice. We produce evidence and a recommendation you can hand to your counsel, who decides.

Get the free scan

We read what your website already publishes, using a labelled crawler at one request per second that honours your robots.txt. You send us nothing, and we ask for no login, no credentials and no access to your systems.

Get the free scan
Three fields — your agency, its website, and where to send the result. A person runs each scan and reviews it before it goes out, so we email you when yours is ready. No obligation, no meeting.

What it does not do is name the documents — which specific files carry which fault is the detailed, paid pass, and the scan is what tells you whether that is worth buying. Everything the scan does and does not tell you.

Every statement we make sits in one of three buckets

Any claim about any document is labelled as exactly one of these, and never blurred between them.

Measured
An instrument produced this — a named tool, at a named version, on a named date. Reproducible by you or by anyone else. It settles what is in the file, and nothing about whether that matters to someone trying to read it.
Judged
A person examined it, against a stated question, on a stated sample. We say how many, not just that we looked. No tool decides whether a failure is a real barrier or only bookkeeping.
Not assessed
Nobody looked, or no method could settle it. Stated as plainly as the other two.

Automated tools decide only part of this standard. Published estimates put automated coverage at roughly 30–40% of accessibility problems. Whether a heading is really a heading, whether alt text describes the picture, whether the reading order makes sense aloud: no validator decides those. Any vendor whose report has no "not assessed" section is not telling you what they did not check. What a checker can and cannot prove, in full.

Four questions that make your answer much better

When we inventory a library from outside, roughly three quarters of it comes back undetermined — not because the analysis is weak, but because whether a document is still used to reach a service is a fact about your organisation, not the file. A crawler cannot see it. You can.

None of these four is required, and each is usually a five-minute conversation with somebody who already knows.

  1. Which parts of your website do you consider archived?

    Where you have labelled a section yourself, that is your evidence rather than our inference — this one usually settles the most. Ask: whoever maintains the website.

  2. Can you share page-view figures for your documents?

    A file nobody has opened in two years is a strong candidate for the exception. Ask: whoever administers your analytics.

  3. Does your records-retention schedule name these materials?

    An adopted document carries more weight with your own counsel than anything we could produce. Ask: the city recorder, clerk or records manager.

  4. Which documents do residents actually ask for?

    A document people request is being used to reach a service, however old it is — the answer no crawler will ever find. Ask: front-counter and permit-desk staff.

You will get the free scan either way, and it will say plainly what it could not determine. We would rather hand you a report with an honest gap in it than one that quietly guessed and looked tidier.

What it costs

Quoted per engagement

for remediation work, after the Inventory tells us what is actually there. We do not publish a per-page rate, because the work is not per-page — rebuilding a document's structure and transcribing a map's legend are different jobs at different costs.

$2,000

for the full Document Inventory: a fresh crawl, tested, with a person's review before release, delivered within five business days. Paid by card, or against your agency's purchase order invoiced net 30 if you write to us first. For libraries up to 5,000 documents; larger agencies do the same.

Buy the Inventory

We publish that number because almost nobody else in this field will. The full picture on cost, in one place.

What we will not do

  • We will not tell you that you are covered by the law. That is a legal conclusion, we are not qualified to reach it, and a vendor in this category was fined $1,000,000 by the FTC in 2025 for making claims of that shape.
  • We will not report a document as fine because a validator found nothing. Absence of a detected fault is not evidence of quality.
  • We will not take on work a machine cannot finish and quietly do it badly. Where a document needs human hands, the report says so and prices it separately — or says we are not the right people for it.
  • We will not pretend our own limits away. On the one library we have measured end to end, our automated pass refused 376 of 2,691 documents and said why: 359 were scanned images with no text to work from, 9 the tagger declined, and 8 were encrypted. We would rather you read that here than discover it after paying us.

Who this is

Recordstake is a small operation, based in West Jordan, Utah — deliberately, which is why our findings are published rather than kept in anyone's head. The corpus research is public, including the two places we were wrong.

Our own website is audited against the same standard we report on, and that audit is published here, including the parts we have not assessed. It would be strange to ask you to accept a report we would not publish about ourselves.

Getting the free scan

Tell us the name of your agency, the address of its website, and where to send the result. We confirm that address from your own inbox first, so nothing is queued by somebody typing it in for you.

Get the free scan