Accessibility statement

This says what should work today, what we know is wrong, and what has never been tested. It is written plainly because the people who need it are deciding whether to bother with us at all.

The honest headline

No one has audited BrightWren against WCAG or any other standard, so this page claims no conformance level. We are a private company, not a public body, so no law requires this statement. We publish it because a parent, a child or a council officer deserves to know what they are getting before they start, rather than finding out halfway through a maths question.

Two different things, with different answers

The free website at brightwren.co.uk needs no account and runs in a browser. The BrightWren Plus app at app.brightwren.co.uk is signed into by a grown-up. They were built at different times and they are not equally good, so every answer below says which one it is about.

What should work now

• A keyboard, with no mouse. Every control is a real button or a real link, so it can be reached with Tab and used with Enter or Space. Nothing is a picture pretending to be a button. Answers can be typed on a physical keyboard everywhere a number is wanted, including the tables check, the SATs paper and the app's questions, so the on-screen number pad is an option rather than the only way in.
• A visible focus ring. Both surfaces draw the same 3px ring, and it appears for keyboard users without flashing up for mouse users. In the space game it is redrawn to follow the shape of each planet, in a pale blue, because the usual dark ring disappears against a night sky.
• Dialogs behave. When the question player or a pop-up is open, everything behind it is switched off for both the keyboard and assistive technology, Escape closes it, and focus goes back to whatever you were on.
• The skip link on the free website jumps past the navigation to the content.
• Nothing moves that you cannot stop. Both surfaces honour the operating system's "reduce motion" setting, in about eighty places, so animations are cut rather than softened.
• Nothing needs sound. There is no audio, no video and no speech anywhere on either surface, so nothing needs captions, a transcript or a description, and nothing is lost with the volume off.
• Zoom works. Nothing blocks pinch zoom or browser zoom, and the maths drawings are drawn to grow with it rather than turning fuzzy.
• Colour is never the only signal. Where a topic is marked on track, getting there or needs a hand, the words are there beside the colour, and so is a shape.
• Old devices. The build refuses to ship stylesheets an iPad stuck on iOS 15 cannot read, so a child on a hand-me-down tablet gets the real layout rather than a broken one.
• In the tables check only: four alternative colour schemes (yellow on black, black on blue, black on peach, blue on cream), bigger text, and on a computer the option to turn the number pad off. These are the access arrangements the real Year 4 check offers.

What we know is not right yet

These are real, we found them ourselves, and they are listed here rather than left for you to discover.

1. Most maths drawings do not say what they show. A screen reader is told the kind of picture and not the value: "an analogue clock", not the time it shows; "a bar divided into eight equal parts, with some shaded", not how many. Four of our nineteen drawings do read out their values. The rest do not, so a question that depends on reading the picture cannot be answered by listening today. This is the one on this page that most affects whether a child can use us at all, and it is the one we most want to fix.
2. The app has no skip link. The free website does. In the app you tab through the navigation on every page.
3. Two of the three answer boxes are not real form fields. The SATs practice box and the tables check box are built out of other elements and fed by a keyboard handler. They take typing, but they may not announce themselves to a screen reader the way a normal box would.
4. About two dozen small pictures have no label and are not hidden, so assistive technology may read them out as unnamed images. Most of our pictures are handled properly; these were missed, and nothing in the build catches them yet.
5. One page in the app, Daily 5, has no main heading, which makes it harder to find your place by headings.
6. Four pages in the space game say they are in English rather than British English, which can change how a screen reader pronounces things.
7. One pop-up in the space game does not switch off the page behind it, so the keyboard can wander behind it. Every other pop-up does.
8. Text size is set in pixels, so raising your browser's own minimum font size does not change anything here. Zoom does work, and that is the way to make it bigger today.

What nobody has tested

We would rather say this than let you assume it.

• No screen reader has ever been used on either surface. The code was written with screen readers in mind, carefully, and that is not the same as trying it. We cannot tell you how it sounds in VoiceOver, NVDA or JAWS, because nobody has listened.
• Nobody has completed a whole journey on a keyboard alone. Individual pieces were built for it and are checked by eye; a full pass has never been done.
• Nothing has been tested at 200% text, or with a reader's own font settings.
• No disabled person has used or reviewed either surface, as far as we know. The list of assistive technologies we have tried is empty, and we are not going to dress that up.
• Nothing has been assessed for dyslexia, attention or reading age. Questions are written in short sentences for the age group, and that is a choice about teaching, not a tested accessibility claim.

How we check what we do check

The app is walked by a machine through 38 journeys at four screen sizes, a phone, an iPad held both ways and a laptop, in two different browser engines, and it is checked for text that is covered, clipped or scrolls sideways, for colour contrast and for tap targets smaller than 44 by 44 pixels. Contrast and tap size are reported as warnings rather than failures, which means they are seen and not enforced, so we will not quote you a contrast figure we cannot stand behind. The free website has no automated checking of this kind at all, which is worth knowing, because it is the surface most people meet first.

If something stops you

Email hello@brightwren.co.uk and say what you were trying to do and what happened. BrightWren is made by one person, so you will get a real reply rather than a ticket number, usually within a few working days. If something on this list is blocking a child right now, say so in the subject line and it goes to the front.

There is no formal complaints route to escalate to, because the regulations that create one apply to public bodies and not to us. If we cannot fix something, we will tell you that plainly rather than leave you waiting.

When this was written

Checked against the code on 20 October 2026, and this is the first version. We will revisit it whenever one of the items above is fixed, and otherwise at least once a school year. Anything fixed is struck from the list rather than quietly removed, so you can see what moved.