Privacy & Accessibility

What is stored locally

Prior Work uses browser storage to keep working material between visits. Refreshing the page or closing the browser does not delete this material. It can include:

• •The active model
• •Report context fields you have entered (organisation name, scope, preparer)
• •Report section content generated during the session
• •Baseline survey rows and demographic information, saved in this browser for restoring group analysis
• •Pulse analysis and report results. Uploaded pulse rows are held in memory and need re-uploading after a full reload.

AI report writing and your API keys

Report writing can use Anthropic or OpenAI, selected by you for all stages of a run. If you choose to use these features:

• •You supply your own provider key. It stays in memory until reload or closing unless you explicitly choose to remember it. Remembered keys are stored separately for each provider, never in a project or report export. The browser contacts that provider directly; your key does not pass through Prior Work's calculation server.
• •Report writing sends analysis results, factor descriptions, organisational context and relevant report text to the selected provider. Survey response rows are not automatically included in the report prompt. Avoid putting identifying or confidential individual details in the context or report text you provide.
• •Review Anthropic's usage policies or OpenAI's service agreement. As the API key holder, you contract with the selected provider.
• •Anthropic retention. Retention depends on the model, API feature and your organisation's agreement. Check Anthropic's current API retention terms before generating a report. Removing data from this browser does not delete copies held by Anthropic.
• •OpenAI retention. Requests disable saved response state and run in the foreground without conversation chaining. This does not mean zero retention: abuse monitoring and prompt caching have separate rules. Review OpenAI's data controls for your account and model.
• •Page scripts can access browser-held keys, including memory-only keys. This direct-browser design is a Prior Work product choice. OpenAI recommends keeping keys on a server. Removing a local key stops requests using it here; revoke it in the provider account to prevent its use elsewhere.
• •Check your organisation's policies and the selected provider's data-processing terms before using report writing. Removing browser data does not delete copies held by a provider.

Backend processing

The application processes uploaded models, survey rows and analysis requests on a calculation server. Report document downloads send report text and supporting analysis to that server for formatting. The application has no database for saving your work; it returns results to your browser. Recently used models can remain in a small server-memory cache until they are replaced or the server process stops. This describes application behaviour, not a verified retention period for hosting-provider logs. Contact us if your organisation needs those details confirmed before uploading.

Analytics and tracking

Prior Work does not use third-party analytics, advertising trackers, or session recording tools. No cookies are set beyond those required for application function.

Workforce data and privacy obligations

Worker survey data can include sensitive information. The organisation conducting the survey remains responsible for ensuring that collection, use, and disclosure of worker survey data complies with applicable privacy law, enterprise agreements, and any relevant workplace policies.

Group-level by design, never individual

Prior Work's analysis operates on the workforce as a whole, or on groupings within it (a team, a site, a department), but never on named individuals. There is no individual scoring, no per-worker risk profile, and no mechanism for the tool to report on a specific person. This is a deliberate design choice, not a limitation we plan to remove.

The reason is straightforward. Psychosocial conditions are produced by how work is organised, supported, and led. The levers for change sit at that organisational level too. A tool that names individuals invites the wrong response, which is to treat the worker as the problem, and crowds out the response that actually shifts outcomes, which is to redesign the conditions. Designing the tool so it can only see groups keeps the conversation where the leverage is.

Practical consequences:

• •Reports describe the working environment, not specific workers.
• •Recommendations target work design, leadership, resourcing, and policy. They do not target individuals for assessment, screening, or support pathways.
• •Combining survey data with a starting model does not guarantee anonymity. Avoid names and identifying detail, combine small groups where appropriate, and check whether combinations of demographic categories could identify someone before sharing results.
• •Decisions about individual workers, including support, performance, or wellbeing conversations, remain a matter for direct conversation, supervisor judgement, and where appropriate clinical or HR support. They are never derived from this analysis.

This stance aligns with ISO 45003's transparency requirements for psychosocial risk management and reflects the broader principle that the leverage to improve psychological safety at work sits in how the work is structured, not in profiling the people doing it.

Target standard

We aim to meet WCAG 2.1 Level AA across all process pages and content pages. This includes:

• •Sufficient colour contrast for all text (minimum 4.5:1 for normal text, 3:1 for large text)
• •Keyboard navigability for all interactive elements
• •Descriptive labels and ARIA attributes on form controls
• •Logical heading hierarchy and landmark regions
• •Meaningful text alternatives for non-decorative images

Known limitations

The following areas have not yet been fully assessed against WCAG 2.1 AA:

• •Model diagram (Data Analytics): The interactive model diagram is navigable by mouse but not yet by keyboard or screen reader. A tabular fallback view is on the roadmap.
• •Report generation output: AI-generated text is presented as prose HTML. The complete reading and generation workflow has not been formally tested with assistive technology.
• •Downloadable templates: CSV and Word document templates have not been assessed for document-level accessibility (e.g. tagged PDF, accessible table structure).

Feedback

If you encounter an accessibility barrier or have a specific access need, please contact patrick@priorwork.au. We aim to respond within five business days.

Assessment approach

Selected layouts and controls have had local browser and keyboard checks. These checks do not establish WCAG conformance. A complete assistive-technology assessment and independent accessibility audit have not been completed.