无障碍访问

We target the Web Content Accessibility Guidelines (WCAG) 2.2 at Level AA across the marketing site, the portal and the simulation runner. WCAG 2.2 is a superset of 2.1, so a customer whose own obligation is stated against 2.1 AA — as it is under the Australian Government's Digital Service Standard — is covered by the same work.

This is a self-assessment. We have not commissioned an independent accessibility audit and do not claim certification. We would rather say that plainly than imply an assurance nobody has given. We are glad to complete an accessibility questionnaire, provide detail against individual success criteria, or walk your team through how a particular surface behaves.

We also treat accessibility as an obligation in its own right under the Disability Discrimination Act 1992 (Cth), not only as a procurement requirement.

Marketing site and legal documents

These are ordinary content pages — text, headings, links and forms. We assess them as meeting WCAG 2.2 AA.

The portal

Dashboards, authoring, reporting and administration. We assess these as substantially meeting WCAG 2.2 AA. The data-dense reporting views are the part we continue to work on, and they are the area where we would most welcome being told about a problem.

The simulation runner

The practice conversation itself. The conversation, the transcript and every control around them meet the standard. The animated character that speaks the other side of the conversation does not, and is described below.

• Keyboard access throughout, including a skip link to the main content on every surface — the marketing site, the portal and the simulation runner — and focus that stays visible rather than disappearing behind a fixed header.
• Screen-reader labels in each of our six languages. Every control speaks the language of the page it is on, rather than reverting to English.
• A live text transcript of the conversation, announced to screen readers as it arrives, so the spoken side of a simulation is always available in writing.
• Reduced motion honoured. If your device asks for less animation, the interface stops animating.
• Password managers work everywhere. We never block paste in a password field, and every sign-in, sign-up and password-change form is filled and saved correctly — so logging in never depends on remembering or transcribing a password.
• Controls sized to be operable without fine pointer control, and no task that requires dragging.
• Automated accessibility checks run against every change before it can be released, alongside our other tests.

A simulation can be completed entirely in text. The typed-response box sits alongside the microphone at every turn of the conversation and needs no setup, no permission prompt and no account — a workshop participant who joins by link has it too. Nothing about the assessment changes: the same character responds, the same scoring applies, and the same feedback comes back.

It can also be the way you always practise. Turning on text-only simulations stops the animated character loading at all: no motion, no synthesised speech, and none of the download a three-dimensional character needs — which matters on an older machine or a metered connection as much as it does for motion sensitivity. The written conversation stays open for the whole run.

The choice is remembered, and it does not require an account. A workshop participant sets it on the screen before the run starts, which is the point at which someone on a borrowed or locked-down machine can still do something about it. Signed-in users can set it once in their profile and have it follow them to another device. Two other settings sit alongside it: force reduced motion even when the device does not ask for it, and keep the transcript visible throughout.

The animated character. The three-dimensional character that speaks during a simulation is rendered in real time with synthesised speech. Its performance — expression, gesture, lip movement — is not conveyed to assistive technology, and it cannot be operated by keyboard, because there is nothing to operate: it is a depiction of the person you are practising with, not a control.

We name this rather than claim around it. What matters is that nothing depends on it: everything the character says appears in the transcript as it is spoken, and the conversation can be completed in text without it. We are not able to make a real-time rendered character fully conformant, so the text path is the accessible route — and since it is a setting you can turn on once and keep, it is a first-class way to use the product rather than a fallback you drop into.

If something on the platform gets in your way, please tell us at soren@evolvesimulations.com. It helps if you can say which page you were on and what you were trying to do, but tell us in whatever form suits you — we will not turn a report away for missing detail.

We aim to acknowledge every report within five business days, and to tell you what we intend to do about it and roughly when. If you are a learner at an organisation that uses Evolve, you can raise it with your administrator instead and they can reach us on your behalf — but you are equally welcome to contact us directly.

If a barrier is stopping you from completing an assessed simulation, say so and we will treat it as urgent. An administrator at your organisation can also override or void a result — see AI transparency for how that works.

This statement is reviewed when the platform changes in a way that affects it, and the date above is the last substantive revision. Our accessibility work is tracked alongside our other controls; a summary is available to customers and prospective customers on request.