Overview
Repeat Findings Detection means Centrik recognises when a new audit finding resembles a past one - a safety net for catching systematic issues early. It flags likely repeat findings automatically, using AI to compare findings by meaning rather than relying on exact wording - the reviewer then confirms or rejects the suggested link.
Before you start:
This feature is optional and disabled by default. To use it, an organisation admin must:
- Opt in to the feature for your organisation
- Agree to the AI data processing terms, as part of enabling AI-powered functionality
Once enabled, the system begins analysing your findings to detect possible repeat findings. There's no per-user setup - it applies organisation-wide once switched on.
If you're interested in activating this functionality, please contact TrustFlight Support.
Access Rights
To view and action suggestions the Quality Manager access right for the appropriate subsystem, and/or anyone with permission to perform the audit that the finding belongs to.
How It Works
Step 1: Finding Processing & Analysis (background)
Once enabled, each finding is processed into an AI-readable representation that captures its meaning - the specific finding text, the regulation or requirement it was raised against, and its context - not just the literal wording. This runs automatically in the background, for existing and new findings alike.
Step 2: AI Comparison (background)
When a new finding is raised, it's compared against previously processed findings to identify ones that describe the same or a very similar underlying non-conformity, even where the wording, audit, or department differs.
Step 3: Review & Confirm (reviewer-facing)
Within the finding, a Smart Suggestions banner appears, showing how many suggestions are ready - e.g. "You have 1 suggestion ready to review and apply" - with a Review suggestions button.
- Review suggestions opens a panel from the right, titled SMART SUGGESTIONS – Possible repeat findings. This shows the candidate findings, along with a short excerpt of its content for context. An About AI-powered suggestions note sets expectations:
- These are suggestions only - nothing is linked until it's reviewed
- Only clear matches appear - low-confidence matches aren't surfaced
-
Confirming or dismissing a suggestion helps improve future runs
2. Viewing the suggestion takes the reviewer to a side by side view, showing the proposed relationship - with Approve and Reject buttons within a Confirm repeat finding.
3. If approved, both findings keep their own separate records, but a link is created between them. Each finding's Relationships section then shows "This Finding relates to" the other finding, with the relationship Type: Repeat.
Unlike Same Event/Hazard Detection in the Safety module, this feature links rather than merges - both findings remain distinct records, each retaining its own audit trail, categorisation, and status, with a visible cross-reference between them.
Things to Know
- Suggestions, not automatic links. A reviewer must approve the suggestion before any relationship is created; rejecting it leaves both findings unlinked.
- This links records, it doesn't merge them. Both findings keep their own identity, status, and history - the link is purely a cross-reference.
- All matches are listed, each with a confidence score. If more than one existing finding plausibly matches, the system lists every match rather than just the single closest one, with a confidence score against each so the reviewer can judge how strong each potential link is.
- To undo an approved Repeat Finding link, contact TrustFlight Support.
Key Benefits
- Catches systematic or recurring issues early, before they're treated as isolated one-offs
- Surfaces likely repeat non-conformities that a manual search across audits might miss
- Uses the meaning and context of a finding, not just keyword overlap or shared regulation reference
- Keeps both findings intact as separate, fully auditable records
- Reviewer stays in control - every suggestion requires an explicit Approve or Reject
Summary
Once enabled, AI Smart Suggestions – Repeat Findings Detection:
- Recognises when a new audit finding resembles a past one, acting as a safety net for catching systematic issues early
- Continuously processes findings in the background
- Compares new findings against existing ones using semantic matching
- Flags likely repeat findings for review, showing only clear matches
- Lets the reviewer approve or reject each suggested link
- Creates a "Repeat" relationship between findings on approval, without merging or altering either record
Helping compliance teams:
- Spot recurring non-conformities across audits early
- Maintain a clear, linked audit trail between related findings
- Keep the decision to link findings firmly with the reviewer