EasyTaxer is Making Tax Digital software provided by Agile Release Limited (of which EasyTaxer is a trading name), a company registered in England and Wales under company number 13237670, registered office 7 Larkshall Road, London, England, E4 7HS. We want as many people as possible to be able to use it, including disabled people.
This statement is deliberately specific about what we have checked and what we have not. Filing a tax return is not something you can take your business elsewhere for on the day of the deadline, so we would rather you knew where the gaps are before you rely on us than discover them yourself.
1. What this statement covers
This statement applies to www.easytaxer.com — both the public pages (the home page, the guides, and the legal pages such as this one) and the signed-in EasyTaxer application you use to record income and file with HMRC.
It does not cover websites we link to, including GOV.UK and HMRC's own sign-in pages. Part of using EasyTaxer involves authorising us on HMRC's website, and HMRC publish their own accessibility statements for those pages.
2. Our current position — we do not claim conformance
We do not claim that EasyTaxer conforms to the Web Content Accessibility Guidelines (WCAG) 2.1 or 2.2 at level AA.
We have not carried out a full assessment against those guidelines, so we cannot honestly state a conformance level — and that includes the partial ones. Every such level is a claim about a completed evaluation. Ours is not complete.
What we have done is a partial automated self-assessment. Sections 3 and 4 set out exactly what it covers and what it leaves out. We would rather publish a narrow true statement than a broad one we cannot stand behind.
3. What we have tested, and how
We run an automated accessibility scan as part of our end-to-end test suite. The details are:
- Tool: axe-core 4.12.0, driven through
@axe-core/playwright4.11.3 in a real browser. It is the only accessibility tool we use. - Rules applied: the WCAG A and AA rule sets —
wcag2a,wcag2aa,wcag21a,wcag21aaandwcag22aa. No rules are switched off and no findings are suppressed. - How often: on every full run of our end-to-end test suite, which in practice is daily.
- Browsers: one browser engine per run — Chromium or Firefox. Safari and WebKit are not tested at all.
The pages it covers:
- the home page (
/) - the privacy policy
- the terms and conditions
- the cookie policy
- the about us page
- this accessibility statement
- the page for accountants
- the sign-in page
- the forgotten-password page
- the registration page for individuals
- the registration page for accountants and agents
- the account activation page
- the signed-in dashboard
- the income sources page
- the Construction Industry Scheme income page
- the profile and residency page
- the guides index (
/guides/) - the getting-started guide
- the guide to what Making Tax Digital for Income Tax is
- the guide to connecting EasyTaxer to HMRC
- the guide to quarterly updates
- the guide to the 2026-27 deadlines
- the guide for landlords
- the guide for sole traders
- the guide to filing without an accountant
- the guide to what to do if HMRC has signed you up
- the guide to a missed quarterly update in 2026-27
- the guide to the records you need to keep
- the guide for people who are a sole trader and a landlord
- the guide to exemptions
- the guide to choosing software
- the guide for accountants
- the glossary
- the help centre (
/help) - the help page on signing up and connecting to HMRC
- the help page on a quarterly update for self-employment
- the help page on a quarterly update for UK property
- the help page on checking your deadlines
- the help page on choosing your income sources
- the help page on choosing standard or calendar quarters
- the help page on signing in and resetting your password
- the help page on adding and editing transactions
- the help page on attaching a receipt
- the help page on importing transactions from a spreadsheet
- the help page on adding a bank account
- the help page on claiming business mileage
- the help page on a quarterly update for foreign property
- the help page on the final declaration
- the help page on your tax calculation
- the help page on accounting adjustments
- the help page on losses and claims
- the help page on your HMRC account
- the help page on tasks you do with HMRC
- the help page on tax liability adjustments
- the help page on employment income
- the help page on CIS deductions
- the help page on taxable state benefits
- the help page on updating your details and password
- the help page on downloading a copy of your data
- the help page on turning on an authenticator app
That list includes the whole guides section — all 17 guide pages. They were previously checked for two specific problems only (colour contrast, and links told apart from the text around them by colour alone); they are now checked against the same full WCAG A and AA rule set as every other page above.
It also includes the whole help centre — all 27 help pages — which was checked against the full rule set from the day it was published.
The threshold it enforces, stated plainly: the scan fails our test suite on violations that axe rates critical or serious. Violations rated moderate or minor are recorded in the test log but do not fail anything. So a page passing our scan is not a page with no automated findings — it is a page with no critical or serious ones.
We do not publish the result of any single run here, and that is deliberate. A result is true of one run on one day; quoting one would tell you about a version of the software that no longer exists, and it would be out of date again by the following morning. What we can tell you is the rule that is always in force: a run that finds a critical or serious violation on any of the pages listed above fails. The public pages in that list are checked on every run; the four signed-in pages need a logged-in session and are checked on our full authenticated run. To be exact about what that is and is not: it is a standing check that puts a failure in front of us, not a barrier that stops a release — so it tells you such a problem would not go unnoticed for long, and it does not promise that no page is carrying one today.
4. What we have not tested
This is the part of an accessibility statement that is usually vague. We have tried to make ours countable.
4.1 Most pages are not scanned
EasyTaxer has roughly 151 pages. The scan covers 60 of them — about one in three. Not scanned includes:
- Every page you actually file tax through — calculations, submissions, transactions, assets, all settings pages, the quarterly and annual submission pages for every income type, losses and claims, and the remaining income landing pages.
- All administrative pages (these are internal, not customer-facing).
4.2 Only one state of each page is scanned
The scan loads each page once, in one window size, in light mode, and looks at it as it first settles. It has never looked at:
- any dialog or pop-up window, including every confirmation step that stands in front of sending data to HMRC
- the keyboard command palette
- any menu or drop-down while it is open
- any form showing validation errors
- hover and keyboard-focus states
- dark mode
- narrow screens, including phone-sized windows
4.3 Automated tools only find some problems
Automated checking can detect roughly 30–40% of WCAG success criteria at all. Of the 50 A and AA criteria in WCAG 2.1, about 12 are meaningfully exercised by our setup. The AA criteria with no automated coverage here include: 1.2.4 and 1.2.5 (captions and audio description for live and prerecorded media), 1.3.4 (orientation), 1.3.5 (identify input purpose), 1.4.4 (resize text), 1.4.5 (images of text), 1.4.10 (reflow at 320px), 1.4.11 (non-text contrast), 1.4.13 (content on hover or focus), 2.4.5 (multiple ways), 2.4.6 (headings and labels), 2.4.7 (focus visible), 3.1.2 (language of parts), 3.2.3 and 3.2.4 (consistent navigation and identification), 3.3.3 (error suggestion), 3.3.4 (error prevention for legal, financial and data submissions) and 4.1.3 (status messages). Level A criteria with no or negligible coverage include 2.1.1 and 2.1.2 (keyboard operation and keyboard traps), 2.4.3 (focus order), 1.3.2 (meaningful sequence) and 1.4.1 (use of colour).
We want to be direct about one of those. WCAG 3.3.4 — error prevention for submissions that carry a legal or financial consequence — is the single most important AA criterion for tax software, and it has no automated coverage in our setup. EasyTaxer does have relevant safeguards built in: figures are shown for review before anything is sent to HMRC; actions that change your HMRC record sit behind a confirmation step; the most serious actions (final declaration, and deleting a submission at HMRC) require you to type a specific word before the button becomes available; and where an action cannot be undone, the confirmation says so in those words. We believe that is a reasonable basis, and we are not going to describe it as conformance, because no one has assessed it against the criterion.
4.4 No manual or assistive-technology testing
We have not carried out:
- any manual accessibility review, by us or by anyone else
- any testing with a screen reader (such as NVDA, JAWS or VoiceOver), screen magnification, speech input or switch access
- any keyboard-only walkthrough of the filing journey
- any testing with disabled people
- any external audit, and we hold no VPAT or EN 301 549 evaluation
5. Known accessibility problems
5.1 The two walkthrough videos: a text alternative is published, and nobody outside the team has checked it
Our getting started and quarterly updates guides each embed a short screen recording (about 13 and 10 seconds). Both are silent — they carry no audio track at all. That means the WCAG criteria about captions (1.2.2) and audio description (1.2.3 and 1.2.5) do not apply to them, because there is no spoken content to caption or describe.
What does apply is WCAG 1.2.1 (prerecorded video-only content), which requires an alternative that presents equivalent information. An earlier version of this statement recorded this as an open gap: written step-by-step instructions covering the same ground appeared below each video, but nothing identified them as an alternative to it and nobody had checked that they said the same thing. Both videos now carry a written alternative of their own. Each one sits under a visible heading, "Text alternative for this video", so that it can be found by a reader who is not using a screen reader, and it is also joined to the video in the markup so that a screen reader announces the two as belonging together. It was written screen by screen from the recording itself rather than adapted from the steps that were already on the page — those steps were compared against the recordings and did not match them: one recording ends on a screen the written steps never mention, and the other stops before the point the written steps describe. Neither video is needed in order to follow the guide.
What we are still not claiming. No one has evaluated either alternative against the criterion, and — as section 4.4 says of the whole site — we have not tested these pages with a screen reader or any other assistive technology. We believe each alternative carries the same information as the video it describes; that is our own judgement and it has not been checked by anyone outside the team. If either one leaves something out, please tell us — section 7 says how.
5.2 Anything outside section 3 is simply unknown
The pages, page states and criteria listed in section 4 have not been assessed. We are not claiming they are inaccessible; we are saying we do not know, and it would be wrong to let a short "known issues" list imply otherwise.
5.3 Moderate and minor findings are logged, not fixed
Our scan records moderate and minor violations without failing on them, and we do not currently publish that log. Some of those findings relate to real WCAG criteria — page structure and landmarks, heading order — on pages that otherwise pass.
This is not a complete list of the accessibility problems on this website, and we ask that it is not read as one. It is a list of the problems we know about, which is a smaller thing.
6. What we are doing to improve accessibility
We are committing to the following, and we will update this statement as each is done:
- Provide a clearly identified text alternative for both walkthrough videos, or remove the videos if we cannot. Done — both are published, and section 5.1 says what has and has not been checked about them.
- Extend the automated scan to every public guide page and to the pages used to record income and file with HMRC. The guide pages are done — all of them are now scanned against the full rule set and are listed in section 3. The pages used to record income and file with HMRC are not yet covered. Target: 5 April 2027.
- Carry out a manual review of the core filing journey using only a keyboard, and with a screen reader, and publish what it finds here. Target: 5 April 2027.
- Extend the scan to dialogs, validation-error states, dark mode and narrow screens, which are the largest untested areas.
- Raise the scan threshold so that moderate and minor violations also have to be fixed, once the current findings are cleared.
- Review this statement at least every 12 months, and whenever the scope of our testing changes.
7. Reporting an accessibility problem
If you find a problem, or you need something from EasyTaxer in a way you cannot currently get it, please tell us — email info@easytaxer.com.
It helps if you can include:
- the page address, or a description of where you were
- what you were trying to do, and what happened instead
- any assistive technology, browser or operating system you were using
We aim to reply within 5 working days. If you are up against an HMRC deadline, say so in the subject line and we will treat it as urgent.
If you need information from this website in a different format — for example a plain-text version of a guide, or larger print — email the same address and tell us what would work for you. We will discuss what we can reasonably provide and how long it will take.
We do not currently operate a telephone helpline. Email and the support feature inside the application are the ways to reach us.
8. Your rights, and the law that applies to us
EasyTaxer is provided by a private company. The Public Sector Bodies (Websites and Mobile Applications) (No. 2) Accessibility Regulations 2018 apply to public sector bodies, and they do not apply to us. We publish this statement voluntarily, not because we are required to, and we are not claiming compliance with those regulations.
The duty that does apply to us is under the Equality Act 2010: as a provider of services to the public we must make reasonable adjustments so that disabled people are not put at a substantial disadvantage compared with people who are not disabled. If you need an adjustment in order to use EasyTaxer, please tell us at info@easytaxer.com and we will do what we reasonably can.
If you are not satisfied with how we respond, please raise it with us again first — we would rather fix it. You can also get free, independent advice about your rights under the Equality Act 2010 from the Equality Advisory and Support Service (EASS) at equalityadvisoryservice.com.
9. Technical information about this website's accessibility
The public pages are standard HTML and CSS and remain readable without JavaScript. The signed-in application is a JavaScript application and requires JavaScript to be enabled in your browser.
Our automated testing runs on the Chromium and Firefox browser engines. We do not test on Safari or any other WebKit browser, which includes every browser on iPhone and iPad.
EasyTaxer is not available in Welsh or any language other than English.
10. How we prepared this statement
This statement was prepared on 11 August 2026. It was last reviewed on 22 August 2026.
Its text was last changed on 28 September 2026, when four more guide pages were published — written versions of four of our short explainer videos — and added to the automated scan in the same change: they are listed in section 3, the guide-page count in section 3 went from 13 to 17 (section 6, which had repeated the old count, now says "all of them"), and in section 4.1 the scanned-page count moved from 56 to 60 and the total-page count from roughly 145 to roughly 151 (the four new guide pages, and two signed-in pages added since the previous count), while the share of pages scanned stayed at about one in three, all re-derived from our source code rather than edited by hand. The day before, 22 more help pages were published and added to the automated scan in the same way: they are listed in section 3, the help-page count in section 3 went from 5 to 27, and in section 4.1 the scanned-page count moved from 34 to 56, the total-page count from roughly 123 to roughly 145, and the share of pages scanned from about one in four to about one in three. The day before that, the new help centre was published and its five pages — the help centre's front page and four help pages — were added to the automated scan in the same change: they are listed in section 3, section 3 now states the help-page count, and the scanned-page count in section 4.1 moved from 29 to 34 and the total-page count from roughly 112 to roughly 123, re-derived from our source code rather than edited by hand. Earlier that day two new guide pages were published and added to the scan in the same way, taking the guide-page count in sections 3 and 6 from 11 to 13. The change before that recorded a widening of the automated scan: the 11 guide pages, which were previously checked for two specific problems only, are now checked against the same full WCAG A and AA rule set as every other page in section 3. Sections 3, 4.1 and 6 were updated to say so, and the counts in them were re-derived from our source code rather than edited by hand. A statement that keeps reporting a gap we have closed is as inaccurate as one that hides a gap we have not. The review date above is a separate fact, deliberately kept apart from this one: it records the last time a person read the whole statement through — 22 August 2026 — and that is the date section 6's twelve-month promise is judged against.
The change before those corrected the figures in sections 3 and 4 — the tool version, the list of pages the scan covers, and the counts derived from it — by re-deriving them from our own source code, and removed a sentence quoting the result of one particular scan run, because a result like that is true of a single day and nothing can keep it true afterwards. That wording was read through and approved by a person on the same day.
It is based on a partial automated self-assessment carried out by us, using axe-core as described in section 3. The scope figures in sections 3 and 4 are not typed in by hand: an automated check re-derives them from our own source code and test configuration, and fails our test suite whenever this page and that code disagree. The tool version, the list of covered pages and the number of them are checked exactly; our total page count is a round figure, so it is checked as an approximation and the word "roughly" is meant literally. No external audit was commissioned and no manual evaluation was performed.