Level Access sells a programme. We sell the automated half of it.
Expert audits, training, conformance documentation and coverage across software and documents are theirs, and they matter. Ours is the layer that runs on every change: the pages you name, fixes ranked by reach, and an honest account of what was never checked.
Scan a public URL free · no credit card required
Choose Level Access if
- You need a person to examine your product and write a defensible finding. They employ trained evaluators, including people who use assistive technology daily.
- Procurement is going to ask for a VPAT — an Accessibility Conformance Report — and no automated tool can produce one honestly.
- The surface is wider than web pages: mobile, software and documents.
- The highest-leverage part of the contract is raising a standard across many teams, through courses and certification.
Choose TestKase if
- You want a specific change in front of a specific developer quickly, ranked by how many issues it clears.
- You want to know whether the change worked: fixed, introduced and unchanged against the previous run.
- You want a report that marks criteria nothing checked as not checked, so a clean scan is never mistaken for conformance.
- You want it without a scoping call, without a per-person charge, and in the same workspace as the rest of the testing.
The head-to-head
A tool and a programme are not competing products. Most of these rows go their way, and the ones that come to us are about speed and specificity rather than depth.
| Item | How the two compare | For us |
|---|---|---|
| Expert manual evaluation | Level Access pairs automated testing with expert manual evaluation, which is the only route to a finding about the criteria a machine cannot reach. Nobody here examines your product and writes a finding; we flag where a human decision is needed and stop. | Theirs |
| Conformance documentation | Accessibility Conformance Report work — the VPAT a procurement team will ask for. We produce scan reports, and a scan report is not a conformance claim. | Theirs |
| Surface covered | Web, mobile, software and documents. Our scanner works on web pages in a real desktop browser. Manual session testing on real mobile devices sits alongside it, which is a different activity and not an accessibility audit of a native app. | Theirs |
| Training and certification | Level Access Academy runs courses and certification. We publish documentation and an honest account of what automation can and cannot check. That is not training. | Theirs |
| Remediation and programme governance | They sell remediation help rather than only reporting a problem, with monitoring and documentation aimed at maturing a programme over years. We report and stop. | Theirs |
| Frequency | An audit is a photograph of one moment and its value decays from the day it is delivered. A scan here can run daily, weekly or monthly, which is what makes the run-to-run comparison useful: eleven issues appeared between Tuesday and Wednesday is a lead, eleven since last quarter is an investigation. | Ours |
| What the report opens with | A ranked list of code changes ordered by how many issues each one clears, with severity as the tie-break, a running total per row, and the selector the fault concentrates on. | Ours |
| Buying it | Level Access engagements are scoped and quoted, which is the correct model when a large part of what you are buying is expert time. Ours is published, has a free tier that scans real addresses, and has no per-person component on the device side. | Ours |
| Overlays | Level Access acquired UserWay, an overlay vendor, and has published its own position on the category — set out with their words below. We do not sell an overlay of any kind, and nothing we sell modifies your site at runtime. | In their words |
- WCAG 2.2 A and AA criteria with an automated rule behind them
- 23 of 55
- Addresses per scan, on every plan including the free one
- 10
- Scan history and schedules kept
- 90 days
- Auditors, VPATs, training courses and overlays sold here
- 0
Our figures, from the Accessibility product pages in this site. There is no comparable Level Access figure to put beside them: engagements are scoped and quoted, and third-party estimates are not the vendor's number, so none is repeated here.
A tool and a programme are not competing products
Accessibility work splits into two halves that are usually bought separately and confused constantly. One half is mechanical: contrast ratios, missing labels, controls without an accessible name, faults that repeat wherever a component repeats. A machine finds those in seconds and a developer clears them in an afternoon.
The other half is judgement. Whether the alternative text describes the image or merely exists. Whether the focus order matches the reading order. Whether the error message tells somebody how to fix the error. Whether a person using a screen reader can finish the task at all. No program decides any of that, and fewer than half of the WCAG 2.2 AA success criteria have an automated rule behind them.
Level Access sells the second half, with the first half attached. We sell the first half properly and are explicit about not selling the second. That is the entire comparison, and which one you need depends on whether your problem is a backlog or an obligation.
The overlay question, stated carefully
One fact is worth knowing when comparing vendors in this field, and it is worth stating without editorialising.
Level Access acquired UserWay, which sells an accessibility overlay, in a transaction agreed at the end of 2023 and completed in 2024; UserWay has continued to operate under its own brand.
Level Access has published its own position on the category. It describes overlay technology as something that “can be an amazing part of an overall solution for accessibility but, on a standalone basis, isn't sufficient to substantiate compliance with most accessibility requirements”, and frames the acquisition as making a low-cost starting point available to organisations that cannot yet fund a full programme. Those are their words and their reasoning, and readers should weigh them directly rather than through us.
Our own position is simpler because our product range is narrower: we do not sell an overlay, we do not offer one as an option, and nothing we sell modifies your site at runtime. The category page on overlays sets out the documented criticism, including the position taken by disability advocacy organisations, with sources rather than adjectives.
What we do instead, and do well
An audit is a photograph of one moment, and the value of it decays from the day it is delivered. The automated layer exists to stop the estate drifting back between photographs, which means its job is not to be thorough — it is to be fast, frequent and specific enough that somebody acts on it the same week.
| Item | What it does |
|---|---|
| The next change, not the whole picture | A fix plan ranked by how many issues each change clears, with severity as the tie-break and a running total on every row, plus the selector the fault concentrates on so you can tell a one-component fix from a forty-file one. |
| Proof that the change worked | Every run is compared with the one before it: fixed, introduced and unchanged, matched on the rule, the page and the element. A total that fell is not evidence that anything specific was repaired. |
| A report that admits what it skipped | Criteria with no automated rule are marked not checked rather than passed, and the machine-checkable set is read from the engine at runtime rather than asserted by us. It makes our reports look worse and makes them worth quoting. |
| In the same place as the rest of the testing | Scans run in a real desktop browser on the same fleet that serves manual sessions on real devices, and a finding becomes a defect in the tracker your team already uses rather than in a separate platform. |
Where Level Access is ahead
Six things they do that we do not. If your requirement is a defensible conformance position for an enterprise, this is the wrong page to be persuaded by.
- Expert manual evaluation — the only route to the criteria a machine cannot reachNot ours
- VPAT and conformance reports — what procurement asks forNot ours
- Remediation services — help fixing it, not only reporting itNot ours
- Academy training and certification — raising a standard across many teamsNot ours
- Mobile, software and document testing — we cover web pages onlyNot ours
- Programme governance and maturity — monitoring and documentation over yearsNot ours
How each one is bought
One evaluation begins with a conversation about your organisation; the other begins with a scan of your site.
| Item | How it is bought |
|---|---|
| Level Access | No published pricing. Engagements are scoped and quoted, which is the correct model when a large part of what you are buying is expert time. Estimates circulate on third-party sites; none of them is the vendor's number and we are not going to repeat one. |
| TestKase Accessibility | Published, with no per-person component on the device side. The accessibility scanner has a free tier that scans real addresses, test management is free for a small team, and device testing is billed per parallel thread, so widening access inside a workspace costs nothing. Everything is on the pricing page. |
The rest of the category
Four answers to the same question, and what each of them is actually for.
| Item | What it is for |
|---|---|
| Level Access | A programme purchase: automated testing plus expert manual evaluation, remediation help, training and certification, conformance documentation, and coverage across web, mobile, software and documents. |
| Siteimprove | Siteimprove is the closest neighbour: estate-wide crawling, a score and dashboards, with analytics and content tooling in the same platform. |
| Deque | Deque is the strongest engineering tooling, and its axe-core engine is the one our scanner runs. |
| TestKase | The automated layer and nothing more: the pages you name, fixes ranked by reach, a run-to-run comparison, criteria nothing checked marked not checked, no per-seat charge, and no overlay. |
What this does not do
- No manual audit, no auditors.
- Nobody here examines your product and writes a finding. The criteria that need a human decision are flagged and left to you.
- No VPAT and no certification.
- We do not issue conformance documentation, and nothing on this site is legal advice.
- No training or certification programme.
- Documentation, not courses.
- Web pages only.
- No document, native software or embedded-device accessibility testing.
- No crawling.
- Up to ten addresses per scan, chosen by you, kept for ninety days.
- No overlay, by choice.
- Nothing we sell injects itself into your site at runtime or claims to repair it from the outside.
Frequently asked questions
No. We do not employ auditors, we do not sell expert evaluation, and we issue no audit report. Level Access does all three, with trained evaluators including people who use assistive technology daily. If what you need is a person to examine your product and write a defensible finding, that is a services purchase and ours is not one.
No. A VPAT — an Accessibility Conformance Report — is a statement about criteria that mostly cannot be judged by a machine, and producing one honestly requires manual evaluation. Level Access does this work. We produce scan reports, and a scan report is not a conformance claim.
Level Access runs an academy with courses and certification, and for an organisation trying to raise a standard across many teams that is often the highest-leverage part of the contract. We publish documentation and an honest account of what automation can and cannot check. That is not training.
Level Access acquired UserWay, an overlay vendor, in a transaction agreed at the end of 2023 and completed in 2024, and UserWay has continued under its own brand. Level Access has published its own position on the category, stating that overlay technology can be part of an overall solution but on a standalone basis is not sufficient to substantiate compliance with most accessibility requirements. We do not sell an overlay of any kind.
Getting a specific change in front of a specific developer quickly, and telling them whether it worked. The report opens with a ranked list of code changes ordered by how many issues each one clears, and every run is compared with the previous one to show what was fixed, what was introduced and what did not move. It also costs nothing per person and does not require a scoping call.
No. Our scanner works on web pages, in a real desktop browser on our own machines. Level Access covers considerably more surface than that. What we do have alongside it is manual session testing on real mobile devices, which is a different activity and not an accessibility audit of a native app.
Often enough that a regression is attributed to the change that caused it. A scan can be scheduled daily, weekly or monthly at a time in a named timezone, and because each run is compared with the previous one, a short interval makes the comparison useful: eleven issues appeared between Tuesday and Wednesday is a lead, whereas eleven appeared since last quarter is an investigation. An expert audit cannot run at that frequency and is not meant to.
Reasonably, yes, and some should. An automated layer that runs on every change and a periodic expert audit answer different questions, and the audit is the one that gets you a conformance position. Using a cheap fast tool to keep the machine-detectable findings at zero makes the expensive human hours go further, because the auditor spends them on the criteria only a human can judge.
Related
- Accessibility overlaysWhy we do not sell one, and what the evidence says
- TestKase vs Sauce LabsReal and virtual devices, per-thread pricing, and one workspace
- TestKase vs LambdaTestHow we compare with LambdaTest, now TestMu AI
- All of Why TestKaseHow we compare with the tools you are probably already paying for
- PricingPer-seat test management and per-thread device testing, priced on the page
Clear the machine-detectable half first
It is the cheapest work you will do, and it is what makes expensive expert hours go further. Scan a public address and see the ranked plan.