How-to · Insurance claims · FRISS · Oct 20, 2026
Set FRISS referral thresholds, and have the SIU lead decide every referral before anyone is investigated
FRISS Claims Analytics returns a risk signal with reasons on each claim, flags those above the insurer’s threshold, and lets low-risk claims fast-track. The threshold is the real decision. This page shows how to set it with evidence and keep every referral with the special investigations unit (SIU) lead, and every payment with the adjuster.
Who this is for
Claims leaders, SIU leads, investigators, and adjusters at P&C insurers using FRISS. The job here is the referral threshold and the SIU lead’s decision on each referral. For having adjusters approve coverage, payments, and messages on claims Clive agents prepared — a different job — see Five Sigma Clive coverage position approval.
How to do it
This is a how-to, not a ranking of fraud tools. Per the FRISS AI review, FRISS calls its platform Trust Automation; a reviewer can always see why a file was flagged and overrule it; the same threshold decides which low-risk claims fast-track; and Media Check looks for reused, edited, and AI-made images and altered PDFs. The ground rule: a FRISS flag is a risk signal, not a finding of fraud.
1. Pick a starting threshold and name its owner
What it is: Agree a starting threshold with the SIU lead and claims supervisor, and write who may change it and what evidence a change needs.
- Start from the threshold FRISS and your team set at implementation; do not lower it in week one.
- Name the SIU lead as owner of the referral threshold.
- Name the claims supervisor as owner of the fast-track rules.
- List the claim types that may fast-track with no concern shown.
- Write that any threshold change needs a month of tallies first.
An SIU lead at a mid-sized auto and home insurer keeps the current threshold and writes the rule: no change without a month of referral tallies. The claims supervisor lists three fast-track claim types: small water damage, glass, and small theft with a police report.
Write one line each: who owns the referral threshold, who owns the fast track, and what evidence a change needs.
2. Read the signals on twenty claims
What it is: Take the last twenty flagged claims and read the reasons FRISS gives, then record whether each was really worth investigating.
- Open the reasons behind each signal.
- Mark worth investigating, quickly cleared, or flagged on a bad data field.
- Look for one field causing many flags, such as a stale address.
- Sample fast-tracked claims too, for gaps like duplicate invoices.
- Keep the tally going monthly.
Twelve were worth investigating, five were quickly cleared, and three were flagged on one stale address field. A sample of fast-tracked claims finds one with a duplicate invoice, so the supervisor adds an invoice check for that claim type.
List the last twenty FRISS flags with three columns: reason given, worth investigating (yes or no), data problem (yes or no).
3. SIU lead decides each referral; run Media Check on photo and PDF evidence
What it is: The SIU lead decides whether each flagged claim is referred. Adjusters treat Media Check flags as a reason to ask for originals, not a reason to deny.
- SIU lead decides refer or clear on every flagged claim, with a one-line reason.
- The adjuster still decides every payment.
- On a Media Check flag, ask the policyholder for original files or schedule an inspection.
- Record how many Media Check flags cleared and share the tally monthly.
- Every closed case records the named investigator who decided.
An adjuster gets a Media Check flag on a roof photo whose metadata does not match the loss date. She asks for the original files and schedules an inspection. The originals explain the mismatch; she notes the outcome so the SIU lead can see how often flags clear.
On the next ten Media Check flags, ask for original files before any referral, and record how many cleared.
Pricing context (October 2026)
This how-to isn’t a price sheet. Per the FRISS AI review, FRISS does not publish prices. When we checked on Oct 20, 2026, friss.com/pricing returned not found and the site offers demos. Treat it as quote-only and ask whether claims, underwriting, investigations, compliance screening, and Media Check are priced separately. Dated October 2026.
Checked Oct 20, 2026 on the vendor’s own site: FRISS publishes no prices; friss.com/pricing returned not found and FRISS’s site lists no prices. This page quotes no FRISS dollar figure; ask FRISS for a written quote and see our review for what is known.
Uncertainty: The only prices this site sells here are the optional $9 AI toolkit spreadsheet and $19 AI toolkit audit.
When this helps (honest take)
Useful when you want consistent fraud and risk signals on claims and a named SIU lead will own the referral threshold.
Not useful when your main gap is casualty severity, you want AI to handle whole claims, or you want flags to trigger denials automatically.
Related options (educational)
| Need | Where to look | Why |
|---|---|---|
| Referral thresholds and Media Check | FRISS AI | Primary path; quote only |
| Adjuster approval on Clive-prepared claims | Five Sigma Clive coverage position approval | A different job |
| Casualty severity and litigation signals | CLARA Analytics AI | Different claim type |
| Where claims AI fits | Stack map: insurance ops | Broader map |
FAQ
Does a FRISS flag mean a claim is fraudulent?
No. Per our review, a flag is a risk signal that a person should look. The SIU lead decides whether to refer, and the adjuster decides whether to pay.
What does Media Check look for?
Reused images, metadata that doesn’t fit, edited images, altered PDFs such as invoices, and AI-generated images, inside the claims workflow.
How much does FRISS cost?
FRISS does not publish prices. Treat it as quote-only (checked Oct 20, 2026).
Optional next step
If you want a one-page toolkit map for the tools you already pay for, use the optional $9 AI toolkit spreadsheet. If you want a short review of seat waste and unused tools, use the optional $19 AI toolkit audit. Neither is required to follow this how-to.