Risk Graphprototype

Three audiences, one graph. Each pack is a view over the same objects, so the board, a regulator, and a customer can never be told different facts, only different levels of detail. The customer pack is checked on every build for dollar figures, internal codes, and staff names.

Issuance warning: 8 metric(s) in this pack have broken lineage (“ATO attempts / 10k logins” on RSK-014; “Successful ATO rate” on RSK-014; “MFA adoption” on RSK-014; “90-day churn of ATO'd accounts” on RSK-014; “Mean time to lockout” on RSK-014; “DSAR responses on time” on RSK-041; “Privileged exports / week” on RSK-033; “Chargeback rate” on RSK-052). It can only be issued with a written annotation.

Board pack · quarterly risk report

Risk committee briefing

For: Board risk committee

Live draft · 2026-11-21

1. Executive summary

Quantified residual exposure across the register stands at $12.67M annualized (inherent $18.37M, reduced 31% by the control environment). 2 risks exceed board-approved appetite (RSK-014, RSK-021) and require treatment or formal acceptance. Customer churn is the largest lens at $8.72M inherent.

2. Exposure by lens

LensInherentShare
Direct monetary$6.55M36%
Regulatory$3.10M17%
Customer churn$8.72M47%
Reputationalscored 1–5 per risk—

Churn is priced as exposed customers × churn probability × lifetime value. The probability starts as an assumption and is recalibrated by every incident that records how many affected customers actually left.

3. Top risks vs. appetite

RiskResidualAppetiteStatus
RSK-014 · Account takeover — customer auth$5.10M$5.00Mabove
RSK-008 · Vendor concentration — cloud region outage$2.48M$3.00Mwithin
RSK-021 · SMS pumping / artificially inflated traffic$2.10M$1.00Mabove
RSK-041 · Customer PII handling (GDPR / CCPA)$1.56M$2.50Mwithin
RSK-033 · Insider data exfiltration$980k$2.00Mwithin

4. Indicator breaches

  • ATO attempts / 10k logins on RSK-014 is at 4.1 per 10k against a red threshold of 4 per 10k, and worsening.
  • 90-day churn of ATO'd accounts on RSK-014 is at 67% against a red threshold of 60%.

5. Loss events

  • Contractor bulk export (2026-10-14, RSK-033): loss $0. Status: investigating.
  • AIT spike via trial accounts (2026-09-05, RSK-021): loss $96k. Status: remediating.
  • ATO cluster via legacy login (2026-07-18, RSK-014): loss $412k; 79 of 118 affected customers churned (67%), which recalibrates that risk's churn estimate. Status: closed.
  • Deletion requests missed in backup tier (2026-06-03, RSK-041): loss $35k; 3 of 40 affected customers churned (8%), which recalibrates that risk's churn estimate. Status: closed.

6. Where we are digging next, and why

  • 1. RSK-014 · Account takeover — customer auth: priority 2.6 ($5.10M residual, 30% blind: 100% of KRIs hand-fed or stale; 1 control(s) untested >180d).
  • 2. RSK-008 · Vendor concentration — cloud region outage: priority 2.1 ($2.48M residual, 50% blind: 100% of KRIs hand-fed or stale; assessment >12 months old; 1 vendor(s) past due-diligence review).
  • 3. RSK-021 · SMS pumping / artificially inflated traffic: priority 1.5 ($2.10M residual, 60% blind: no KRIs defined; 1 linked attack vector(s) with no detections).

Investigation priority = exposure × blindness × velocity. You dig where you are most exposed and least able to see, not where the biggest number sits.

Issue this pack

Issuing freezes the pack exactly as rendered above. The snapshot is hash-chained to the one before it, and the database refuses any edit or delete, so what was sent stays provable after the underlying data moves on.