Ask five department heads what “high risk” means and you will get five answers, each defensible on its own terms. The trouble starts when those answers roll up into one report for the board.
This piece looks at where the definitions diverge, why risk appetite is the translation layer, and how a single scale across domains makes the comparison honest.
Image placeholder: Hero image, 1200 × 630. Devensa AI branded graphic for “One Risk Scale for Every Department”.
Executives feel the pressure: 74% reported a significant operational surprise in the past five years, and 69% see rising risk volume and complexity. Far fewer have the shared structure to respond, with 29% holding a formal risk appetite and 11% seeing strategic advantage from risk management.1
Departments define “high risk” differently because most organizations never set a shared risk appetite or a common scoring scale. Finance scores by dollars, security by exploitability, operations by downtime. The 2026 State of Risk Oversight study from NC State and the AICPA found only 29% of organizations have formally articulated risk appetite in the context of strategic planning.1
Ask five department heads to rate the same event and the answers diverge fast.
Finance calls a risk high when the expected loss crosses a budget threshold. Security calls a vulnerability high when the CVSS score passes 7.0, regardless of what the affected system holds. Legal calls a matter high when a regulator might act. Operations measures in hours of downtime. Product worries about a missed launch window.
Take one event: a payment vendor goes offline for six hours. Operations rates the outage high because checkout stopped. Finance rates the event medium because lost revenue stayed under the quarterly threshold. Security rates the event low because no data left the building. Legal has no rating at all until a customer complains. Four teams, one incident, four conclusions about whether leadership needs to hear about the outage before Friday.
Each scale makes sense inside one department. The problem starts when the five lists roll up into one enterprise heat map. A red cell from security and a red cell from finance look identical on the slide, though one represents a $50,000 exposure and the other a regulatory action. The board reads both as equal.
A heat map built from inconsistent scales does not show enterprise risk. The map shows how loudly each department rates its own concerns.
Risk appetite is the amount and type of risk an organization accepts in pursuit of its objectives. Risk tolerance turns appetite into specific thresholds: a maximum loss, a minimum uptime, zero tolerance for a category of regulatory breach.
COSO’s 2017 framework, Enterprise Risk Management: Integrating with Strategy and Performance, organizes ERM into five components and 20 principles. Defining risk appetite sits in the Strategy and Objective-Setting component.2 Appetite comes before scoring, because scores mean nothing without a shared reference point.
The misconception treats risk management as a register: collect risks, color them, file the spreadsheet. A register without an agreed appetite becomes a list of departmental opinions. ERM done properly connects every rating back to what the organization decided to accept.
Image placeholder: Supporting chart or diagram for this section. Devensa AI branded.
In the 2026 study, 74% of executives reported a significant operational surprise within the past five years, and 69% said the volume and complexity of risks increased over the same period. Only 30% described their risk oversight as mature or robust.1 The authors pointed to siloed risk processes as one barrier to seeing how risks connect.1
The 2026 State of Risk Oversight, the 17th annual study from NC State University’s ERM Initiative and the AICPA, surveyed 331 executives and board-level leaders in the second quarter of 2026.1
Credit Suisse lost $5.5 billion when Archegos Capital collapsed in 2021. The independent investigation by Paul, Weiss found numerous warning signals, including large and persistent limit breaches, which risk managers and senior executives failed to act on.3 Archegos also received temporary, bespoke increases to its risk appetite, including one of $900 million.4
The bank had limits. The business treated them as negotiable, and the risk function lacked the standing to hold the line. When one unit’s “high” becomes another unit’s “manageable,” the limit stops working.
Only 11% of executives in the 2026 study said risk management gives their organization a unique strategic or competitive advantage.1 Without a shared definition of high, risk information never reaches the strategy table in a usable form. In the Devensa vC-Suite model, building the shared definition belongs to the vCRO, the AI-powered virtual officer for enterprise risk governance and executive risk reporting.
Devensa AI scores Risk Management as one of six domains on the TruMaturity™ framework. The AI-mediated assessment produces a Risk Management Domain Maturity Score on the CMMI scale, 1 Initial, 2 Developing, 3 Defined, 4 Managed, 5 Optimized, plus an Overall Effectiveness Score across all six domains.
Because all six domains share one scale, a Cybersecurity score and a Compliance score mean the same thing at the same level. The tenant sets a target per domain. The gap becomes a roadmap, prioritized by risk, with each item assigned to the domain whose score the item improves.
A single impact scale in business terms. Define each level across dollars, customers affected, regulatory consequence, and operational disruption. A “4” means the same thing in every department.
Likelihood with a time horizon. “Likely” within 12 months, not “likely” at some point.
Tolerance thresholds per category. Written limits for financial, operational, compliance, and reputational risk, approved by the board.
Escalation rules. A breach of tolerance goes to a named executive within a set number of days. Exceptions carry an expiration date and a signature.
One register. Every department feeds the same register using the same scale. Departmental views become filters on shared data.
Devensa AI scores all six governance domains on one TruMaturity™ scale, so a high rating means the same thing in every department and every report. A virtual Chief Risk Officer (vCRO) on the platform keeps the register, owners, and review cadence in one place.
Risk appetite is the broad amount and type of risk an organization accepts to pursue its objectives. Risk tolerance sets measurable thresholds around specific metrics, such as a maximum acceptable loss or a minimum system availability. Appetite sets direction. Tolerance makes direction enforceable.
Heat maps compress different departmental scales into the same colors. A red rating from security and a red rating from finance often represent far different exposures. Without a common impact scale tied to risk appetite, the map ranks departmental emphasis instead of enterprise exposure.
COSO’s 2017 ERM framework places defining risk appetite inside its Strategy and Objective-Setting component.2 ISO 31000 offers a parallel approach built on establishing risk criteria. Either works when leadership commits to one scale across the organization.
Less common than most boards assume. The 2026 NC State and AICPA study found 29% of organizations have formally articulated risk appetite in strategic planning.1
The platform scores all six governance domains on one CMMI-based TruMaturity™ scale, so maturity levels mean the same thing across Cybersecurity, AI Governance, Risk Management, Compliance, Privacy, and Data Governance. AI-generated recommendations route through human approval before anything gets published or acted on. If five of your department heads rated the same risk tomorrow, how many different answers would you get?
Devensa AI scores all six governance domains on TruMaturity™ and keeps the evidence behind each score current. Early access is open now.
Join the early access list: devensa.ai/early-access