Monthly Security Operations
Report: October 2026
Document control and scope of service
Contents
- 1. Executive dashboard3
- 2. Alert volume and triage funnel4
- 3. Incident summaries5
- 4. Detection coverage against MITRE ATT&CK7
- 5. Log source health and retention9
- 6. SLA performance10
- 7. Threat intelligence and exposure11
- 8. Tuning actions12
- 9. Recommendations and next month12
- Annex. Definitions and conventions13
1. Executive dashboard
Alerts per week
Incidents by severity
Median time to detect, last six months (minutes)
Top detection categories
| Category | Alerts reviewed | Share | True positive rate |
|---|---|---|---|
| Authentication anomalies (spray, impossible travel) | 1,104 | 34.4% | 23% |
| Endpoint malware and suspicious execution | 688 | 21.4% | 41% |
| Web application attacks (WAF, application logs) | 602 | 18.7% | 19% |
| Network scanning and reconnaissance | 410 | 12.8% | 12% |
| Data movement and DLP | 238 | 7.4% | 37% |
| Other | 170 | 5.3% | 30% |
2. Alert volume and triage funnel
Raw correlation matches are enriched and de-duplicated by SOAR playbooks before an analyst sees them. The funnel below shows how volume reduces to the incidents that needed human response.
| Stage | Count | Notes |
|---|---|---|
| Events ingested | 412,600,000 | 58 log sources, EDR telemetry, cloud audit logs |
| Correlation matches (raw alerts) | 18,640 | 142 rules |
| Auto-closed by SOAR (enrichment, allow-list, known benign) | 15,428 | 82.8% of raw alerts; each closure auditable |
| Analyst-reviewed alerts | 3,212 | Tier 1 and Tier 2 |
| True positive alerts | 918 | 28.6% of reviewed |
| False positive alerts | 2,294 | 71.4% of reviewed (September: 76.0%) |
| Incidents opened | 27 | Grouped from related true positives |
Funnel view
Alerts by log source family
| Source family | Raw alerts | Reviewed | Incidents |
|---|---|---|---|
| Identity and email | 5,230 | 1,104 | 9 |
| Endpoint (EDR) | 4,110 | 688 | 8 |
| Network and perimeter | 3,870 | 410 | 3 |
| Web application and WAF | 3,420 | 602 | 5 |
| Cloud and data | 2,010 | 408 | 2 |
Observations
- False positive share fell by 4.6 points after the tuning actions in section 8, freeing about 35 analyst hours.
- Identity and email telemetry now produces the largest share of incidents. This matches the industry pattern and the priorities in section 9.
- No alert backlog was carried into November.
3. Incident summaries
Six of the 27 incidents are summarised here: the P1, the P2 incidents and one P3 that carried a lesson. Technical detail for all 27 is in the case management system, available to named client users.
INC-2610-004 Credential-harvesting phishing and mailbox compromise (finance user) Critical
INC-2610-009 Password spraying against the VPN gateway High
INC-2610-012 Suspicious script execution on a branch workstation High
INC-2610-017 Large upload to personal cloud storage by a contractor account High
INC-2610-021 Scanning followed by exploit attempts on a public web server High
INC-2610-023 Privileged account used outside the change window Medium
4. Detection coverage against MITRE ATT&CK
Coverage is measured against techniques that the joint threat model has marked as relevant for Example Finserv (a lending business with retail and agent channels). A technique is Covered when at least one tested rule produces a usable alert, Partial when coverage depends on a single data source or is untested, and a Gap when nothing detects it.
4 covered, 1 partial, 3 gap
56%
2 covered, 1 partial, 3 gap
42%
9 covered, 1 partial, 1 gap
86%
10 covered, 1 partial, 1 gap
88%
11 covered, 3 partial, 2 gap
78%
8 covered, 2 partial, 2 gap
75%
11 covered, 4 partial, 3 gap
72%
5 covered, 1 partial, 2 gap
69%
10 covered, 2 partial, 2 gap
79%
7 covered, 3 partial, 2 gap
71%
6 covered, 1 partial, 1 gap
81%
5 covered, 2 partial, 2 gap
67%
7 covered, 2 partial, 1 gap
80%
5 covered, 1 partial, 1 gap
79%
7 covered, 1 partial, 1 gap
83%
Weighted coverage counts Partial as one half. Tactics follow the current ATT&CK Enterprise matrix as retrieved in November 2026. Overall: 160 prioritised techniques, 107 covered, 26 partial, 27 gaps, 75% weighted.
Detail by tactic
| Tactic | ID | Prioritised | Covered | Partial | Gap | Weighted % |
|---|---|---|---|---|---|---|
| Reconnaissance | TA0043 | 8 | 4 | 1 | 3 | 56% |
| Resource Development | TA0042 | 6 | 2 | 1 | 3 | 42% |
| Initial Access | TA0001 | 11 | 9 | 1 | 1 | 86% |
| Execution | TA0002 | 12 | 10 | 1 | 1 | 88% |
| Persistence | TA0003 | 16 | 11 | 3 | 2 | 78% |
| Privilege Escalation | TA0004 | 12 | 8 | 2 | 2 | 75% |
| Stealth | TA0005 | 18 | 11 | 4 | 3 | 72% |
| Defense Impairment | TA0112 | 8 | 5 | 1 | 2 | 69% |
| Credential Access | TA0006 | 14 | 10 | 2 | 2 | 79% |
| Discovery | TA0007 | 12 | 7 | 3 | 2 | 71% |
| Lateral Movement | TA0008 | 8 | 6 | 1 | 1 | 81% |
| Collection | TA0009 | 9 | 5 | 2 | 2 | 67% |
| Command and Control | TA0011 | 10 | 7 | 2 | 1 | 80% |
| Exfiltration | TA0010 | 7 | 5 | 1 | 1 | 79% |
| Impact | TA0040 | 9 | 7 | 1 | 1 | 83% |
Largest remaining gaps and plan
| Area | Why it matters | Plan |
|---|---|---|
| Reconnaissance and resource development | Mostly outside your network. Visibility depends on threat intelligence. | Brand-abuse and lookalike-domain monitoring added in November |
| Stealth (log tampering, masquerading) | Attackers hide after initial access | Add audit-policy change and log-clearing rules for 12 server groups |
| Discovery on endpoints | Early signal of intrusion, but noisy | Behavioural baselines for admin tooling, tested in December |
| Collection (staging of data) | Precursor to exfiltration | Extend DLP and file-access analytics to the lending database share |
5. Log source health and retention
| Source | Type | Avg events per second | Health (month) | Retention (days) | Location | Status |
|---|---|---|---|---|---|---|
| Perimeter firewalls (2) | Network | 2,140 | 99.98% | 365 | India | OK |
| Core and branch switches | Network | 610 | 99.90% | 180 | India | OK |
| Identity provider and directory | Identity | 480 | 99.99% | 365 | India | OK |
| EDR (1,240 endpoints) | Endpoint | 3,020 | 99.70% | 180 | India | OK |
| Web application firewall | Web | 1,330 | 99.95% | 365 | India | OK |
| Cloud audit logs | Cloud | 540 | 99.95% | 180 | India region | OK |
| Email security gateway | 260 | 99.90% | 180 | India | OK | |
| Core lending application | Application | 190 | 99.60% | 180 | India | OK |
| Legacy ERP application | Application | 35 | 98.20% | 90 | India | Action |
| Branch router syslog (14 sites) | Network | 75 | 97.40% | 180 | India | Watch |
Health issues this month
- Branch router syslog: three sites dropped for up to 11 hours because of a link outage. No security-relevant gap identified from other telemetry.
- EDR agents: 14 endpoints out of contact for more than 7 days. IT asked to confirm decommissioning or repair.
- Onboarding: 2 new sources (payments gateway logs, HR system audit logs) are planned for November.
6. SLA performance
Clocks start when the alert is raised. Acknowledgement means an analyst has taken ownership. Response means containment has begun or a documented decision has been made.
| Severity | Acknowledge target | Respond target | Incidents | Median acknowledge | Median respond | SLA met |
|---|---|---|---|---|---|---|
| Critical P1 | 15 min | 60 min | 1 | 7 min | 22 min | 1 of 1 |
| High P2 | 30 min | 4 h | 4 | 7 min | 1 h 12 min | 4 of 4 |
| Medium P3 | 4 h | 24 h | 13 | 31 min | 5 h 20 min | 12 of 13 |
| Low P4 | 1 business day | 3 business days | 9 | 2 h 10 min | 1 business day | 9 of 9 |
Service-level summary
| Measure | Target | Achieved | Trend |
|---|---|---|---|
| Alert acknowledgement within target | 95% | 97.8% | Up 1.1 points |
| Incident response within target | 95% | 96.3% | Up 0.4 points |
| Monthly report issued by working day 5 | Yes | Yes (5 November) | Stable |
| Critical alerts handled by Tier 2 within 15 minutes | 100% | 100% | Stable |
| Platform availability (SIEM) | 99.5% | 99.93% | Stable |
SLA exception
INC-2610-023 (P3) exceeded its resolution target by 3 hours because the confirmation needed from the application owner arrived the next morning. No security impact. We have agreed a named delegate for out-of-hours confirmations.
Resolution time by severity, median (hours)
7. Threat intelligence and exposure
Relevant advisories and campaigns
| Theme | Relevance to the client | SOC action |
|---|---|---|
| Phishing kits targeting financial services sign-in pages | Directly relevant (see INC-2610-004) | Lookalike-domain watchlist, mail gateway rule, user alert issued |
| Password spraying on remote access | Seen this month (INC-2610-009) | Conditional access review, weak-password audit |
| Exploitation of internet-facing web components | Scanning observed (INC-2610-021) | Patch verification and WAF virtual patch review |
| CERT-In advisories received and reviewed | 17 advisories this month | 9 applicable, 9 actions closed, 0 pending |
External exposure snapshot
| Measure | September | October | Comment |
|---|---|---|---|
| Internet-facing hosts | 46 | 47 | One new host, owner confirmed |
| Critical or High exposures open | 5 | 3 | Two patched; one in change window |
| Expired or expiring certificates (30 days) | 2 | 1 | Renewal booked |
| Leaked credentials found in monitored sources | 4 | 7 | All forced to reset; none linked to a successful login |
| Lookalike domains registered | 2 | 3 | Takedown requests filed |
Phishing simulation and reporting
Users reported 214 suspicious emails this month, up from 168. Median time from user report to SOC verdict was 14 minutes.
8. Tuning actions
Tuning keeps analysts focused on real activity. Each change is tested in a staging mode against seven days of history before release, and rolled back if true positives are lost.
| Rule or use case | Issue | Change | False positives before | After |
|---|---|---|---|---|
| Impossible travel (identity) | VPN exit nodes caused false alerts | Excluded corporate VPN ranges, added device trust check | 310 | 64 |
| Brute-force on VPN | Threshold too low for password spraying | Added per-source and per-account windows | 40 | 12 |
| PowerShell encoded command | Admin scripts flagged | Allow-list signed scripts by hash | 188 | 41 |
| Large outbound transfer | Backup jobs flagged | Excluded backup service accounts with time window | 96 | 18 |
| WAF scanning noise | Duplicate alerts | Grouped by source over 15 minutes | 420 | 95 |
| New admin account | Provisioning flow flagged | Correlated with approved change tickets | 52 | 7 |
| Rare process on server | Patch cycle noise | Maintenance window suppression | 126 | 30 |
| Email forwarding rule (new) | No rule existed | Added after INC-2610-004; 0 false positives to date | - | 0 |
New detections released
- Inbox rule creation with external forwarding (T1114.003).
- Sign-in from a new device after impossible-travel signal, combined with mailbox rule change.
- Cloud storage upload from non-corporate devices to three consumer services.
- Audit policy change and log-clearing on domain controllers.
- Legacy authentication attempts, by protocol.
- Rare parent-child process for Office applications (two variants).
9. Recommendations and next month
| # | Recommendation | Why | Owner (suggested) | Priority | Target |
|---|---|---|---|---|---|
| 1 | Disable legacy authentication tenant-wide and enforce phishing-resistant MFA for finance and privileged users | INC-2610-004 root cause | IT and security | High | 30 days |
| 2 | Raise ERP log retention to 180 days, confirm India-resident storage | CERT-In direction (iv) | Application owner | High | 30 days |
| 3 | Enforce a password blocklist and review service-desk accounts | INC-2610-009 | IT | Medium | 30 days |
| 4 | Complete EDR coverage on 14 silent endpoints | Coverage | IT | Medium | 15 days |
| 5 | Add payments gateway and HR audit logs | Coverage of critical applications | SOC and application owners | Medium | 45 days |
| 6 | Run a tabletop exercise on data breach notification (CERT-In 6 hours, regulator and DPDP process) | Readiness | CISO, Compliance, SOC | Medium | Q1 |
| 7 | Name an out-of-hours delegate for change confirmations | SLA exception | Head of IT | Low | 15 days |
Planned for November
- Purple-team exercise on the six highest-priority ATT&CK techniques.
- Two new log sources onboarded.
- Quarterly service review with the CISO and a refresh of the threat model.
Decisions we need from you
- Approval for conditional access changes on the VPN gateway (recommendation 1).
- Confirmation of the named delegate for out-of-hours change confirmation.
Annex. Definitions and conventions
| Term | Definition |
|---|---|
| Event | A single log record |
| Alert | A correlation match raised by a detection rule |
| Incident | One or more related true-positive alerts that require response |
| MTTD | Median time from first malicious activity visible in logs to alert (measured on incidents with a known start) |
| P1 to P4 | Critical, High, Medium, Low by impact and urgency, agreed in the service schedule |
| False positive | Alert reviewed and judged not to be malicious or a policy violation |
| Covered, Partial, Gap | Detection coverage categories used in section 4 |
| SOAR | Security orchestration, automation and response |
| EDR | Endpoint detection and response |
| DLP | Data loss prevention |
| NTP | Network time protocol |
All names, systems, incidents and figures are fictional. MITRE ATT&CK is a registered trademark of The MITRE Corporation.