AnySec
Managed SOC Report: What It Must Contain — AnySec Insights
← InsightsManaged SOC · 8 min read

Managed SOC Report: What It Must Contain

A dashboard of alert counts is not a managed SOC report. The sections to require, the evidence that lets you re-check each case, and the red flags to reject.

By AnySec engineering

The short answer

A useful managed SOC report states what was actually monitored, which log sources and detections were live and which were not, then walks through the alerts that mattered with the evidence behind each verdict, what the provider did, what it asked of you, and what remains open. A dashboard of alert counts shows activity, not coverage, and cannot tell you what the service missed.

Who this is for

This is for a CTO, head of security or compliance owner who runs, or is about to buy, a managed SOC and wants to know what the recurring document must prove. It is also for whoever must show it to an auditor, an insurer, a licensing body or a board. Before the report there is the contract: what a managed SOC quote must say covers scope and price, and where the SLA clock starts covers response commitments. Before either there is the go-live work described in managed SOC onboarding. This article is about the artefact you will read every period afterwards.

Why a dashboard is not a report

Most disappointing SOC reports fail in predictable ways.

  1. They count activity and call it protection. Thousands of alerts triaged sounds reassuring, but the figure rises with noisy detections and falls when a source quietly stops sending logs.
  2. They never state coverage. If the report does not list which sources and detections were live, you cannot tell whether a quiet month meant safety or blindness.
  3. They hide the cases. An alert reported only as "closed, benign" cannot be re-checked. The reasoning is the product.
  4. They skip what the provider asked of you. Many SOC findings need an action by your team, such as disabling an account or patching a host. If those requests, and their status, are not recorded, the loop never closes.
  5. They are the same every month. A template with the numbers swapped is a sign that nobody read the period's data.

The structure of the report belongs in the quote, not in a conversation after a quiet month turns out to have been a blind one.

A printed alert case file with a single lime-green tab lies beside a wax-sealed envelope and a pen on a dark analyst desk
Each closed alert should leave a case file you can re-open: the evidence, the reasoning and the handover.

The report anatomy matrix

Use this as a checklist when you write the quote or review a period's report. It contains no thresholds or timings, because the right ones come from your contract, your licences and your own risk.

SectionWhat it must showRed flag if missingWho relies on it
Executive summaryPosture for the period in plain language, with the confidence level and what changedReassurance with no stated basisBoard, executives
Coverage statementLog sources and detections live during the period, and the ones that were notA silent assumption that everything was coveredSecurity lead, auditor
Source healthSources that stopped, lagged or changed format, and when each was noticed and fixedNo mention of sources, or "all healthy" with no methodOperations, engineering
Significant casesFor each material alert: what fired, the evidence, the verdict and the reasoningVerdicts with no evidence or reasoningSecurity lead
Actions taken and requestedWhat the provider did, what it asked your team to do, owners and statusRequests that disappear after the first mentionRisk owner, engineering
EscalationsWhich cases were escalated under the agreed rules, to whom and how they endedNo record, or escalation by informal message onlyRisk owner, auditor
Detection tuningWhat rules were changed or suppressed, why and who approved itSuppressions with no approval or reasonSecurity lead
Gaps and threatsTechniques or assets the detections do not cover, and the plan for themSilence, which looks like completenessEveryone
Limits and assumptionsWhat the provider could not see and the assumptions the report rests onLimits buried in an appendix or missingEveryone
Next-period planPlanned onboarding, tuning and tests, with ownersA report that looks only backwardsOperations

The two rows that separate a real report from a dashboard export are the coverage statement and the limits section. One shows what was watched; the other shows the author will say where watching stopped.

Coverage first: what was actually watched

A report that cannot say what it covered cannot say what it missed. Ask for these properties.

  • A named list of live sources and detections, so the report can be compared with the list agreed at go-live.
  • Source health with dates. When a source stopped, when the provider noticed, who was told and how it was fixed. A silent source is one of the commonest ways monitoring fails.
  • Changes in your estate. New systems, new vendors and retired assets should appear, because a SOC that is not told about them cannot cover them.
  • Honest gaps. A detection that was planned but not live should be listed as such, not left out.

The coverage statement is also the part an auditor or licensing reviewer is most likely to ask for, because it shows monitoring was operating and not just contracted.

Cases: the evidence behind the verdict

Alerts are the raw material; cases are the product. For each significant case, the report should give you enough to re-check the judgement.

  1. What fired, in terms of the detection and the event, not just a rule name.
  2. The evidence examined, such as which logs, which accounts, which time window.
  3. The verdict and its reasoning, including what would have changed it.
  4. What was done and by whom, including any containment the provider carried out under your pre-agreed authority.
  5. What was asked of your team, with an owner and a status that survives to the next report.

Ask for at least one true finding and one false positive, with reasoning, in each review. A provider that only shows its wins is not showing its work. The incident-side equivalent, for when a case becomes an incident, is covered in what an incident response report must contain.

Tuning and suppression: the quiet source of blind spots

Detections are tuned to cut noise, and every suppression is a decision to look less. A good report records the following.

  • What was changed or suppressed, in terms you can read, not an internal rule identifier alone.
  • Why, with the alert pattern that justified it.
  • Who approved it, by role, because a suppression that your team never saw is a decision made on your behalf.
  • When it will be reviewed, so a temporary exception does not become permanent by neglect.

This section often explains a drop in alerts that looks like good news. If volume fell, the report should say whether the environment improved or the detections were narrowed.

Three layers of reporting

Decide before the first quiet month which information arrives when.

  • Immediate. Notification of incidents under the agreed escalation rules, by a channel and to named roles. This is not a report; it is an alert with a human behind it.
  • Operational summary. A regular written view of coverage, source health, significant cases and open requests, for your security team.
  • Periodic review. A conversation with a person who can answer questions, supported by a management summary for executives and any audit evidence you need.

Ask which layer carries each item, who attends the review, and whether the provider will adapt the format for your auditor or licensing body. Where a case turns into a breach, check whether a PCI forensic investigator is required, as that role has its own reporting format.

Red flags when you review a report

  • Alert counts with no coverage statement.
  • "All sources healthy" with no method or dates.
  • Cases closed with a verdict but no evidence or reasoning.
  • Actions requested of your team that vanish from later reports.
  • Suppressions with no reason or approval.
  • A report identical in structure and wording to last month's, with only the numbers changed.
  • No limits section, or limits in an appendix.
  • Metrics that reward closing alerts quickly rather than getting them right.
  • A provider unwilling to let you share the report with an auditor or licensing body.

Ask for the report before you sign

Treat the deliverable as a quote line. Before you sign, ask for a redacted sample from a current client, the headings the provider uses, the cadence of each reporting layer, who attends the review, how source health is detected, how suppressions are approved, whether you may share the report with auditors and in what form, and whether reporting time is inside the price. If no sample exists, treat that as information. A report also does not replace the detections themselves: the log-source scoping for your own sector is covered in SOC vs MDR vs SIEM for iGaming, and a recurring finding is a signal for a penetration test or a security hardening review.

Limitations

This article is not legal or regulatory advice and does not state what any regulator, insurer, licensing authority or payment scheme requires in monitoring evidence or by when. It uses no external statistics and no case studies. The matrix is a review checklist to adapt to your own contract, jurisdiction and estate, and it contains no thresholds or timings. A managed SOC report describes what was monitored within the limits of the sources and detections in place; it does not prove that nothing happened, that an attacker was never present, or that coverage is complete. Reports reflect the data available at a point in time, so the period and the version matter.

Decision and next step

If you can read a period's report and, for the coverage claim, see the named sources and detections, for each significant case find the evidence and reasoning, and for the limits find a clear statement of what was not watched, the report is doing its job. If not, ask for the evidence behind the summary or reject the format. To have the report, its three layers and its evidence rules written into the service before the first quiet month, get a managed SOC quote from AnySec and we will map your estate, your log sources and your audit needs to the sections you receive.

Sources and review

The framing follows public guidance: NIST SP 800-92, Guide to Computer Security Log Management, which covers log sources, retention and review, and NIST SP 800-61 Rev. 3, Incident Response Recommendations and Considerations for Cybersecurity Risk Management, which covers detection, analysis and lessons learned, read 2026-11-20; check the current versions, and read your own regulator's and insurer's requirements. AnySec prepared the report matrix from the public structure of managed monitoring engagements; no AnySec-original statistics are used. Author: AnySec Engineering. Published 2026-11-20; reviewed by the AnySec security team.


Related reading

To have the deliverable specified up front, get a managed SOC quote.

Frequently asked questions

What should a managed SOC report contain?
Coverage first: which log sources and detections were live during the period and which were not. Then the alerts that mattered, each with the evidence behind the verdict, what the provider did, what it asked you to do, and what is still open. Add tuning changes, detection gaps found, and the limits of what was monitored. A count of alerts with no coverage statement tells you activity, not protection.
How often should a managed SOC provider report?
Agree three layers in the contract: immediate notification for incidents under the escalation rules, a regular operational summary, and a periodic review with a human who can answer questions. The cadence should come from your risk and licence obligations, not a template. Ask what is delivered at each layer and who attends the review.
Are alert counts and closed tickets a good measure of a SOC?
On their own, no. Volume can rise because detections are noisy or fall because a log source silently stopped. Ask for counts alongside coverage, the share of alerts that led to an action, and examples of both a true finding and a false positive with the reasoning. Numbers are useful when you can see what produced them.
What should we do if a log source stopped sending data?
The report should show it as a coverage gap, with the date it was noticed, who was told and what was done. A silent source is one of the commonest reasons a SOC misses something. Make source health a standing section of every report, and agree who owns fixing the source on your side.
Can we use the SOC report for an audit or a licensing review?
It can support one, but check that the provider will let you share it, in what form, and whether sensitive details can be redacted. Evidence your own auditor or licensing body asks for should be written into the contract before you need it. This article does not say what any regulator requires.

Rather not learn this in production.

Talk to the engineers behind these write-ups — no sales script, a straight read on where you stand.

Get a fixed quote