Join our Newsletter — 33% off our NHI Course
Home› Guides› Threat Intelligence Programme Metrics Guide: KPIs and KRIs…
Guide Threat Intelligence & Threat Actors

Threat Intelligence Programme Metrics Guide: KPIs and KRIs to Report CTI Value to the Board

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 14 November 2026 14 min read
On this page

Boards keep funding cyber threat intelligence (CTI) on trust, because most programmes report what they produced, feeds, reports and indicators, rather than what changed in the organisation's risk because of it. The CTI-CMM project makes the same point in its metrics material: many organisations measure level of effort or throughput rather than maturity growth or the support given to each stakeholder. This guide sets out a small set of outcome KPIs and key risk indicators (KRIs) that a CTI lead or CISO can put in front of an executive committee or board, with formulas, targets, trend rules and a one-page board pack. It is written for CTI leads, security leaders and the risk and audit staff who challenge them. The operational measures that tell the CTI team whether its own machinery works (feed fidelity, indicator ageing, request turnaround) live in the Threat Intelligence Programme Health Check; programme capability levels are scored in the Threat Intelligence Maturity Model. This page shares only two measures with the health check, and says which.

Key takeaways

  • Report outcomes, not output. Board metrics count decisions, mitigations and exposure closed because of intelligence, never feeds ingested or reports published.
  • Pair KPIs with KRIs. KPIs show the programme delivering; KRIs show where threat-driven risk sits above appetite and needs a decision.
  • Six to eight measures is the ceiling. A board page with four KPIs and three or four KRIs, each with a target, a trend and an owner, is read; a dashboard of 25 is not.
  • Tie every metric to a priority intelligence requirement. If a measure cannot be traced to a requirement a stakeholder agreed, it measures activity, not value.
  • Trend beats snapshot. Show four quarters, set targets after two quarters of baseline, and explain every amber or red in one sentence with the decision you need.

Why CTI is hard to measure, and what to measure instead

Intelligence rarely acts on its own. A detection is written by the detection engineering team, a patch is applied by infrastructure, a supplier is challenged by procurement. The CTI team's contribution is the warning, the context and the priority, which disappears from view once someone else acts. The CTI-CMM metrics guidance notes that as programmes mature, measuring impact needs close collaboration with partner teams. In practice that means three disciplines:

  • Record attribution at the point of action: add an "intelligence source" field to detection changes, vulnerability exceptions, emergency changes and risk decisions, so the link exists when you count later.
  • Measure through stakeholders: each CTI-CMM domain represents a stakeholder function, such as security operations, vulnerability management or third-party risk. Measure what intelligence changed for each one, not what the CTI team did.
  • Separate three layers: operational health (is the machinery working), programme outcomes (did intelligence change decisions and exposure) and threat risk (where is risk above appetite). Only the last two go to the board.
LayerAudienceExample measuresWhere on nhimg.org
Operational healthCTI lead, SOC managerFeed fidelity, indicator ageing, request turnaround, report consumptionThreat Intelligence Programme Health Check
Programme outcome KPIsCISO, executive committee, boardIntelligence-driven decisions, advisory-to-mitigation time, priority technique coverage, forewarned incident rateThis guide
Threat KRIsRisk committee, boardPriority threats without tested controls, unanticipated material incidents, overdue intelligence-prioritised exposuresThis guide; see also Key Risk Indicators for Cyber Risk Guide
Capability maturityCISO, audit, annuallyLevel per domain from Ad hoc to OptimisedThreat Intelligence Maturity Model

For the general distinction between the two indicator types, see KRI vs KPI in Cyber Security.

Start from requirements and appetite

Two inputs must exist before any metric is meaningful. First, written priority intelligence requirements (PIRs), each with a named stakeholder; the Intelligence Requirements Template covers how to write them. Second, risk appetite statements that the KRIs can be measured against. Draft them with the risk function in plain language, for example:

  • "We do not tolerate a known exploited vulnerability on an internet-facing system remaining open more than 14 days after it becomes relevant to us."
  • "Every threat group we rate as a priority must have tested detection or prevention for its main techniques within one quarter of being prioritised."
  • "Material incidents should rarely involve a threat our intelligence had not already reported."

The figures in these statements are examples: set your own with the risk owner. NIST CSF 2.0 GV.RM-03 expects cyber risk outcomes to feed enterprise risk management, and appetite statements are the bridge that lets threat KRIs sit in the enterprise risk register beside other risks.

The programme outcome KPIs

Choose four to six of these. KPI 1 and KPI 2 are the two measures shared with the health check (its T1 PIR coverage and T7 actioned intel share), reported here as quarterly trends rather than as operational thresholds. The rest are board-level only.

IDKPIFormulaData sourceExample target
K1PIR coverage (shared with health check T1)Active PIRs with a finished product in the quarter / active PIRsPIR register, product library90% or more
K2Actioned intelligence share (shared with health check T7)Products that led to a recorded action / products distributedProduct library action field40% or more
K3Intelligence-driven decisionsCount of decisions in the quarter with intelligence recorded as a cause: new or tuned detections, blocks, patch reprioritisations, credential resets, supplier actions, risk acceptances or investment changesChange, detection, vulnerability and risk records with the intelligence source fieldRising trend; no fixed number
K4Advisory-to-mitigation timeMedian days from a relevant external advisory or intelligence product to completed mitigation (patch, detection deployed, configuration change), for items rated high or criticalVulnerability and change tickets linked to the productBelow the appetite figure, for example 14 days
K5Priority technique coverageATT&CK techniques used by priority threat groups that have a tested detection or preventive control / ATT&CK techniques used by priority threat groupsThreat profile for each priority group; detection repository with test results; control validation resultsRising trend, 70% or more over time
K6Forewarned incident rateIncidents of medium severity or above where the threat actor, technique or vulnerability was in an intelligence product before the incident / incidents of medium severity or aboveIncident post-incident reviews with a "forewarned" fieldRising trend
K7Stakeholder value scoreMean rating from a short quarterly survey of PIR owners ("Did intelligence change a decision you made?", "Was it timely?") on a 1 to 5 scaleSurvey of named PIR stakeholders4 or above, with response rate above 70%
K8Cost per actioned outcomeTotal CTI spend in the period (staff, feeds, platform) / recorded intelligence-driven actions in the periodFinance data; K3 countFalling or stable while K3 rises

Notes on the KPIs

  • K3 is a count, not a rate: it can be gamed by splitting one decision into several. Define a decision as one change record or one risk decision, and audit a sample each quarter.
  • K4 uses relevance as the start: start the clock when the CTI team marks the advisory as relevant to your estate, not at publication, and report the gap between publication and relevance separately if it is large. The CISA Known Exploited Vulnerabilities catalogue is a common source of advisories for this measure.
  • K5 depends on a threat profile: the technique list comes from your own prioritised threat groups, not the whole ATT&CK matrix. Count a technique as covered only when the detection or control has been tested, for example through purple team exercises or control validation.
  • K6 measures foresight, not prevention: a forewarned incident still happened. Read K6 with KRI R2 below: rising forewarning with no fall in incident impact means warnings are not being acted on.
  • K8 is for spending discipline: use it to challenge feed and platform renewals, not to compare analysts.

The threat KRIs

KRIs carry thresholds tied to appetite. Report three or four.

IDKRIFormulaExample threshold (green, amber, red)Decision it prompts
R1Priority threats without tested controlsPriority threat groups or campaigns with less than 50% technique coverage more than one quarter after prioritisation0; 1 to 2; 3 or moreFund detection work, accept the risk, or reprioritise
R2Unanticipated material incidentsMaterial incidents in the last four quarters involving a threat that no intelligence product had reported0; 1; 2 or moreReview collection and requirements for the blind spot
R3Overdue intelligence-prioritised exposuresIntelligence-prioritised vulnerabilities, for example those in the CISA KEV catalogue, open on internet-facing systems past the appetite deadline0; 1 to 5; more than 5, or any over 30 daysEscalate remediation or approve a time-limited exception
R4Unremediated exposed credentialsLeaked credentials, API keys or tokens for your domains or code found by intelligence and not reset or revoked within 72 hours0; 1 to 3; more than 3, or any privileged or machine credentialEnforce reset, review the identity owner's controls
R5Single-analyst dependencyPIRs that only one analyst can service, or critical collection sources managed by one personUnder 20%; 20% to 40%; over 40%Cross-training, documentation or managed service support

R4 is where intelligence meets identity. Leaked secrets and tokens for service accounts and integrations are often found by the CTI team first, and the measure only works when the identity or platform team owns the reset. See the Threat Intelligence for Identity and NHI Guide.

Building the board page

One page, refreshed quarterly, structured the same way every time:

  1. Headline (two sentences): the most important change in the threat picture this quarter and what was done about it.
  2. KPI strip: four KPIs, each with the current value, target, four-quarter trend line and an arrow.
  3. KRI table: three or four KRIs with RAG status against appetite, and for each amber or red the owner, the decision needed and the date.
  4. Two examples: one intelligence-driven decision with a short before and after, and one gap the programme found. Concrete cases persuade boards more than percentages.
  5. Asks: any funding, risk acceptance or policy decision required.

Keep raw counts of feeds, indicators and reports off the board page. If a director asks about them, the health check has the answer. The Threat Intelligence Executive Briefing covers the questions boards should ask in return.

Setting targets and reading trends

  • Baseline first: collect two quarters before setting a target. A target set before baseline is either trivially met or never met.
  • Direction targets for counts: K3 and K6 get "rising" targets; absolute numbers depend on the threat environment, not the team.
  • Threshold targets for KRIs: derive them from appetite statements and agree them with the risk owner, not the CTI team alone.
  • Explain inflections: a sudden rise in K4 may follow a burst of advisories in one quarter. Annotate the trend rather than adjusting the target.
  • Review annually: retire metrics that have been green for four quarters and no longer drive a decision, and replace them with the next gap.

Rolling out the metric set

  1. Quarter 1, foundations: confirm PIRs and owners, agree three appetite statements with the risk function, and add the intelligence source field to detection, vulnerability, change and risk records.
  2. Quarter 1, pilot: report K1, K2, K3 and R3 internally to the CISO only, and check the data by sampling ten records per measure.
  3. Quarter 2, extend: add K4 and K5 once the threat profile for priority groups is built, add the "forewarned" field to post-incident reviews for K6, and run the first stakeholder survey for K7.
  4. Quarter 3, first board page: present KPIs with two quarters of trend and KRIs with thresholds; mark targets as provisional.
  5. Quarter 4 onwards: set final targets, add R1, R2 and R4 as data permits, and include the measures in internal audit's scope once a year.
  6. Rollback: if a measure's data quality fails sampling two quarters running, drop it from the board page and report it internally until fixed, rather than presenting numbers nobody trusts.

Choosing measures by size and maturity

SituationRecommended setReason
One part-time analyst or a consumed serviceK2, K4, R3, R4All four come from tickets you already have; no threat profiling needed
Small dedicated team (two to five)K1, K2, K3, K4, R1, R3Adds requirement coverage and decision counts once PIRs are written
Mature team with detection engineering partnersK3, K4, K5, K6, R1, R2, R4Technique coverage and foresight need a threat profile and incident review discipline
Regulated financial entityAdd evidence of threat-led testing inputs to the board pageUnder TIBER-EU, bespoke threat intelligence is the input for red team tests, and the framework supports DORA threat-led penetration testing

Organisations that buy intelligence rather than produce it can still measure outcomes: K2, K4 and R3 apply equally to a managed service, and they are a fair basis for the service review. See Open-Source vs Commercial Threat Intelligence.

Data sources and measurement quality

  • Intelligence source field: the single most important data change. Without it, K2, K3 and K4 are estimates.
  • Threat profile: a list of priority groups and their techniques, kept in the threat intelligence platform and reviewed each quarter. The Strategic vs Operational vs Tactical Threat Intelligence guide explains which products feed which layer.
  • Detection test results: K5 depends on tests, not on the existence of a rule. Link detection IDs to test runs.
  • Post-incident review template: add "forewarned: yes, partly, no" and the product reference.
  • Sampling: each quarter, sample records behind each metric and confirm the attribution is genuine. NIST SP 800-55 Vol. 1 covers selecting and evaluating measures, which is a useful structure for the annual metric review.

Framework mapping

AreaNIST CSF 2.0NIST SP 800-53 Rev. 5Other
Receiving and using CTIID.RA-02, DE.AE-07PM-16 Threat Awareness Program, SI-5 Security Alerts, Advisories, and DirectivesISO/IEC 27001:2022 Annex A 5.7 Threat intelligence; MITRE ATT&CK M1019 Threat Intelligence Program
Threats informing risk prioritisation (R1 to R3)ID.RA-03, ID.RA-05RA-3 Risk AssessmentCISA KEV catalogue for exposure prioritisation
Measuring and reporting performanceGV.OV-03, ID.IM-01PM-6 Measures of PerformanceNIST SP 800-55 Vol. 1
Feeding enterprise riskGV.RM-03RA-3Appetite statements and enterprise risk register
Hunting driven by intelligenceDE.AE-07RA-10 Threat HuntingSee the Threat Hunting Guide

Common failure patterns

  • Volume dressed as value: "2.3 million indicators ingested" on a board slide invites the question "so what?" and erodes credibility.
  • No attribution at the point of action: trying to reconstruct, at quarter end, which detections came from intelligence produces guesses. Add the field first.
  • Coverage counted on the whole matrix: reporting coverage against every ATT&CK technique hides gaps against the groups that actually target you.
  • KRIs without appetite: a red status no one has agreed a threshold for starts an argument instead of a decision.
  • Metric sprawl: adding every health check measure to the board page buries the two or three numbers that need attention.
  • Targets owned only by the CTI team: K4 and R3 depend on infrastructure and application owners. Agree targets with them or the metric reports their backlog as a CTI failure.

Practitioner checklist

  • Confirm written PIRs with named stakeholders before choosing any metric.
  • Agree three to five threat appetite statements with the risk function.
  • Add an intelligence source field to detection, vulnerability, change, risk and incident records.
  • Pick four KPIs and three or four KRIs from the tables, matched to your size.
  • Keep the shared measures to K1 and K2 and leave operational measures in the health check.
  • Build the threat profile for priority groups before reporting technique coverage.
  • Collect two quarters of baseline before fixing targets.
  • Present the one-page board pack quarterly, with two concrete examples and explicit asks.
  • Sample the records behind each metric every quarter and drop measures that fail.
  • Review the metric set annually alongside the maturity assessment.

Standards and references

Related NHI Mgmt Group resources: Threat Intelligence Maturity Model · Threat Intelligence Programme Health Check · Threat Intelligence Executive Briefing · Security Metrics and KPIs Programme Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 14 November 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.