Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
A detection coverage map records which attacker techniques you can actually detect, how reliably, and on which parts of the estate. Built honestly, it is the most useful planning document a SOC owns.

What Is a Detection Coverage Map?
Ask a security team what it can detect and the answer is usually a list of products. A detection coverage map replaces that with something checkable: for each attacker technique, is there a detection, does it work, on how much of the estate, and when was it last proven? It is a map in the sense that it shows where you are, which is why the honest ones have large blank areas.
MITRE ATT&CK is the usual backbone because it gives a shared vocabulary of techniques and sub-techniques, and most detection tooling already maps to it. The ATT&CK Navigator renders the scoring as coloured layers that can be combined and exported.
The decision that shapes everything afterwards is the level of granularity. Scoring at technique level flatters you badly: claiming coverage of T1003, OS Credential Dumping, because you detect LSASS memory access says nothing about NTDS extraction or DCSync. Score at sub-technique level and let a parent inherit the weakest score among its children.
A binary covered or not covered scale is useless, because almost everything ends up amber in reality. A workable scale separates the states that matter:
| Score | Meaning |
|---|---|
| 0 | No telemetry. You could not detect this even if you wrote a rule. |
| 1 | Telemetry exists, no detection written. |
| 2 | Detection written, never validated against a real execution. |
| 3 | Validated by testing within the last quarter, known false positive profile. |
| 4 | Validated and alert-free: the technique is prevented outright. |
Keep the estate dimension separate. A rule that works on managed Windows laptops and not on Linux servers, macOS, or the domain controllers is not coverage; it is partial coverage, and the proportion of assets in scope belongs in the record alongside the score.
The single most common failure is scoring from rule names. A rule called “Suspicious LSASS Access” scores 3 in most organisations and has never been fired by anything except a test the author ran on their own machine two years ago. Score 3 should require evidence: an execution via Atomic Red Team, a simulation platform result, or a purple team exercise, with a date attached. Attach the test identifier to the map entry so the claim can be re-checked by someone else.
The companion discipline worth adopting is DeTT&CT, an open project that scores data source quality before coverage: completeness, timeliness, retention, and field availability. It frequently reveals that a technique scored 2 is really a 0, because the log source nominally in place does not carry the field the detection needs.
Three uses justify the effort. Prioritisation: overlay the techniques used by groups credibly targeting your industry and work the difference, rather than attempting the whole matrix. Assurance: a dated, evidence-backed map answers the board question about what you would catch far better than a product list, and supports NCSC Cyber Assessment Framework principle C1, which asks whether monitoring is informed by the threat the organisation actually faces. Change control: a map updated whenever a log source or rule changes shows the coverage consequences of decisions made for cost reasons elsewhere.
Coverage theatre is the main risk: a map produced for a presentation, coloured generously, never revisited. Guard against it by requiring a date and an evidence reference on every score above 1, by capping how much of the matrix may be scored without testing, and by treating the map as a living artefact in version control rather than a slide.
Chasing completeness is the second trap. Full coverage of ATT&CK is neither achievable nor desirable, since many techniques are irrelevant to your platforms and several are not detectable with any practical telemetry. A map showing deliberate, documented decisions not to cover certain techniques is more mature than one showing uniform mediocrity.
At Privia Security, we deliver coverage maps with the underlying test evidence attached as part of our SOC maturation services, so the scores can be audited rather than believed.
No. Many techniques do not apply to your platforms, some have no practical telemetry, and a few are better handled by prevention than detection. A credible map includes explicit decisions not to cover certain techniques, with the reasoning recorded. Uniform partial coverage across the whole matrix is usually a worse position than deep, tested coverage of the techniques your likely adversaries actually use.
Treat it as living. Update entries whenever a rule is written, retired, or materially changed, and whenever a log source is added or lost. Re-validate scores claiming tested coverage at least quarterly, since environments drift and a rule that worked in March may be silently broken by June. A map with a last-reviewed date older than a quarter should be treated as unverified.
Coverage says a detection exists for a technique. Data source quality asks whether the underlying telemetry is complete, timely, retained long enough, and carries the fields the detection needs. The DeTT&CT project scores the second explicitly, and the exercise commonly downgrades coverage claims, because a rule built on a log source that reaches only half the estate cannot deliver what the map asserts.
Detection engineering, with input from threat intelligence on which techniques matter and from the SOC on how rules behave in production. Give it a single named owner, keep it in version control alongside the rules themselves, and review it in the same forum that approves detection work. Shared ownership without a named individual is how maps quietly stop being updated.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.