Skip to content
CTEM Vendors

Home / Guides / The five stages of CTEM and what each hands to the next

Guide, read 12 Nov 2026

Each CTEM stage needs one input and hands one output to the next

Stage Coverage MapAll guides

Scoping turns a business priority into a testable boundary

Scoping needs one input: a business owner willing to say which service matters most. Gartner's 21 Aug 2023 article tells programs to scope beyond the focus of typical vulnerability management, including social media accounts, code repositories and supply chain systems, and suggests external attack surface or SaaS security posture as first pilots.

The output is a scope statement: the business service, the systems and identities that support it, the third parties it depends on, the owner of each, and what counts as a bad outcome. Without that statement every later stage drifts back to scanning everything. The scoping stage page lists vendors that help with this stage, and the first cycle guide covers how to choose a first scope.

Discovery fills the scope with assets and exposures

Discovery needs the scope statement and access to the data sources that describe it: scanners, cloud accounts, identity providers, SaaS admin consoles, code platforms. Gartner says discovery should identify visible and hidden assets, vulnerabilities, misconfiguration and other risks, and warns that confusing scoping with discovery is often the first failure.

The output is an inventory of in-scope assets with the exposures found on each, plus the gaps: systems the scope names that no tool could see. A useful inventory record carries an owner and a criticality. New York's Part 500, as amended, already asks covered entities to track owner, location, classification or sensitivity, support expiration date and recovery time objective for each asset, which makes a sensible field list for any program. See the discovery stage page.

Prioritization cuts the list to what an attacker would use

Prioritization needs the exposure inventory and threat context. Gartner lists the factors: urgency, security, availability of compensating controls, tolerance for residual attack surface and level of risk to the organization. Public threat context includes the CISA KEV catalog (1,779 entries in version 2026.11.10) and MITRE ATT&CK, whose v20.1 release, current since 13 Oct 2026, lists 227 Enterprise techniques and 479 sub-techniques.

The output is a short list, sized to what the validation team can test inside the cycle, each with a stated reason: on a path to a crown-jewel asset, exploited in the wild, no compensating control. Gartner says the goal is not to fix every single security issue. If prioritization hands validation a thousand items, it has not done its job. See the prioritization stage page.

Validation proves or disproves each item on the short list

Validation needs the short list and permission to test in the scoped environment. Gartner's validation step asks whether attackers could actually exploit the vulnerability, what paths lead to the asset, and whether the current response plan is fast and substantial enough. Methods include control testing with safe simulations, automated penetration testing and human-led testing.

The output is each item marked proven, not proven or blocked by a named control, with evidence: the test run, the date, the technique, and what the control did. That evidence is what makes the mobilization stage credible to owners outside security. Gartner adds that stakeholders should agree on what triggers lead to remediation. See the validation stage page and the validation guide.

Mobilization moves proven exposures to the people who fix them

Mobilization needs proven exposures, an owner for each, and an agreed path for approvals. Gartner says programs cannot wholly rely on automated remediation, though it may suit obvious and unobtrusive issues, and that the aim is to remove obstacles to approvals, implementation and mitigation deployment, with cross-team approval workflows documented.

The output is a set of closed items: patched, mitigated in an existing control, or formally accepted by a named risk owner, each confirmed by a re-test. That record feeds the next cycle's scope. The mobilization guide covers deduplication, SLAs and re-validation, and the mobilization stage page lists vendors.

The handoffs in one table

Each row is a contract between two stages. When a cycle stalls, the stalled row usually has a missing input.

StageNeedsHands onTypical owner
ScopingA business owner and a named serviceScope statement with systems, identities, third partiesCISO with the business owner
DiscoveryScope statement and data accessAsset and exposure inventory, plus visibility gapsExposure management team
PrioritizationInventory and threat contextShort list with a reason per itemExposure management team
ValidationShort list and test permissionProven, not proven or blocked, with evidenceValidation or red team
MobilizationProven items and ownersFixes, mitigations, accepted risks, re-test resultsAsset owners in IT, cloud and applications

The stages repeat as a cycle

The C in CTEM is continuous. Gartner's analyst Jeremy D'Hoinne describes the approach as a way to continuously refine priorities, since organizations cannot fix everything nor be sure which remediation they can safely postpone. In practice that means one cycle per scope, then the next scope, with discovery and prioritization data refreshed between runs.

This site's method assumes a cycle of 4 to 6 weeks per scope. Faster is possible once data feeds are automated; slower usually means validation or approvals are blocking. The roles guide assigns each stage, and the Stage Coverage Map shows how many vendors sell each one.

Each cycle also feeds the next one. Mobilization records show which kinds of exposure kept coming back, which owners closed items fastest and which accepted risks expire soon. Those facts belong in the next scoping meeting: they tell the business owner whether to keep the same scope for a second cycle or move on, and they tell the exposure team which data sources discovery missed.

When a program breaks, it usually breaks at a handoff, not inside a stage. Discovery produces an inventory that prioritization cannot rank because owners and criticality are missing. Prioritization produces a list too long to test. Validation produces results that owners never see. Checking each handoff against the table above, once per cycle, catches most of these before they cost a quarter.

Sources

  1. https://web.archive.org/web/20251230022449/https://www.gartner.com/en/articles/how-to-manage-cybersecurity-threats-not-episodes, read 12 Nov 2026
  2. https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json, read 12 Nov 2026
  3. https://attack.mitre.org/techniques/enterprise/, read 12 Nov 2026
  4. https://attack.mitre.org/resources/versions/, read 12 Nov 2026
  5. https://www.dfs.ny.gov/system/files/documents/2023/12/rf23_nycrr_part_500_amend02_20231101.pdf, read 12 Nov 2026

Last reviewed 12 Nov 2026.