Data Policy

1. What “Decision Data” Means

Decision data refers to the structured content of a submission: the decision made, the category, the year, the age of the submitter at time of decision, the outcome, the regret score, and any optional fields (wishes known, advice given). This data is the core product of the Platform.

2. Collection

Decision data is collected exclusively via the submission form at /submit. We do not infer, scrape, or purchase decision data from third parties. All data in the archive is voluntarily submitted by real users who have personally experienced the decision.

3. Anonymisation Process

Before any submission enters the public archive, the following anonymisation steps are applied:

• The submitter's user ID is decoupled from the record.
• Age is rounded to the nearest 5-year bracket for public display.
• Free-text fields are scanned for named individuals, locations, and company names, which are replaced with category labels (e.g., “[Company]”, “[City]”).
• The year of decision may be generalised (e.g., “2019–2021”) if specificity would re-identify the submitter.

A privacy-preserving link back to the submitter's account is retained only for follow-up purposes (if opted in) and is deleted upon account deletion.

4. Storage & Security

Data is stored in MongoDB Atlas clusters in EU-West regions (Ireland). Data is encrypted at rest (AES-256) and in transit (TLS 1.3). Access to the production database is restricted to application service accounts and principal engineers; access logs are retained for 90 days.

5. Public Archive Access

Anonymised decision data is publicly readable. It is accessible via the search interface at /search, via individual decision pages, and via the API for authorised users. The archive is indexed by search engines.

6. Research & Bulk Access

Academic and institutional researchers may apply for bulk dataset access. All bulk exports use the same anonymised schema as the public archive, with no additional de-anonymisation. See our Researchers page for access tiers and licensing terms.

7. Data Deletion

You may request deletion of all submissions linked to your account. This removes the personal linkage; the anonymised public record may persist in the archive if it has already entered the public dataset, as it no longer constitutes personal data under applicable privacy law (GDPR Recital 26).

8. Longitudinal Follow-up Data

If you opt in to follow-up, we will contact you at 1, 3, and 5-year intervals to update your regret score and outcome. Follow-up responses are attached to your original anonymised record and update the public archive entry. You may withdraw from follow-ups at any time via account settings.

9. AI & Machine Learning Use

We use aggregated, anonymised decision data to train and refine the AI-assisted pattern analysis features (regret curve, decision similarity). We do not train on personally identifiable information. We do not sell data to external AI training providers.

10. Contact

For data-specific enquiries, contact our data team at data@regretindex.com. For privacy and personal data requests, use privacy@regretindex.com.