A risk register template is a table with one row per risk and a fixed set of columns: risk ID, asset or process, threat, vulnerability, likelihood, impact, inherent score, treatment, owner, residual score, review date and linked controls. The cybersecurity risk register template below has all 12 columns, a 1 to 5 scoring scale and one filled-in row to copy.
At a glance
- Columns: 12: risk ID, asset or process, threat, vulnerability, likelihood, impact, inherent score, treatment, owner, residual score, review date, linked controls
- Scoring: Likelihood 1 to 5 multiplied by impact 1 to 5, so scores range from 1 to 25
- Treatment options: Mitigate, accept, transfer, avoid
- Typical size: 5 to 15 risk scenarios for a small or mid-sized company, with 2 or 3 fraud scenarios
- Review cadence: Quarterly, plus after any significant change or incident
- Home: One register, usually the GRC platform, never split across tools
- Framework fit: SOC 2 CC3.2 and CC3.3, ISO/IEC 27001:2022 clauses 6.1.2, 6.1.3 and 8.2
What columns does a cybersecurity risk register template need?
A cybersecurity risk register template needs columns that trace each risk from cause to decision: what is at risk, what could go wrong, how bad the risk is before and after treatment, who owns the risk and when the team looks at the risk again. My approach uses these 12 columns:
| Column | What to write | Example entry |
| Risk ID | A permanent identifier that never gets reused | R-007 |
| Asset or process | The system, data set or business process at risk | Production cloud management console |
| Threat | The event or actor that could cause harm | Account takeover through phishing or credential stuffing |
| Vulnerability | The weakness the threat would use | Two break-glass admin accounts without MFA |
| Likelihood | 1 to 5 on the written scale | 3 |
| Impact | 1 to 5 on the written scale | 5 |
| Inherent score | Likelihood multiplied by impact, before treatment | 15 (High) |
| Treatment | Mitigate, accept, transfer or avoid, with the planned action | Mitigate: enforce MFA on all admin accounts |
| Owner | A named person with authority to act | Head of Infrastructure |
| Residual score | The score after the treatment is in place | 5 (Low) |
| Review date | The next date the owner rechecks the risk | 2027-02-20 |
| Linked controls | Policies, configurations, tickets and framework criteria | Access Control Policy, MFA config, ticket SEC-142 |
Optional columns some teams add:
- Status (open, in treatment, accepted, closed)
- Date identified
- Category
- Approver and approval date, required for every accepted risk
Key insight
Separate threat and vulnerability columns point to the right control
A row that says only ransomware names a threat without the weakness, and the team cannot tell which control would lower the score.
Free download · Google Sheet
Risk Register Template
The 12-column register with scores and ratings calculated for you, the 1 to 5 scoring scale, a heat map of open risks and a review log. Enter your email to get it.
How do you score likelihood and impact in a risk register?
Risk register scoring multiplies a likelihood rating by an impact rating, both on a written 1 to 5 scale, so the inherent score ranges from 1 to 25. The written scale keeps two people from rating the same risk differently. My approach uses this scale:
| Rating | Likelihood (within 12 months) | Impact |
| 1 | Rare: no known cases in the industry or the company | Minimal: no customer effect, no data exposure |
| 2 | Unlikely: has happened elsewhere in the industry | Minor: short internal disruption, no customer data |
| 3 | Possible: has happened to peers or nearly happened here | Moderate: customer-facing outage of hours, limited data exposure |
| 4 | Likely: expected at least once a year | Major: outage of a day or more, or customer data exposed |
| 5 | Almost certain: happening now or several times a year | Severe: large-scale data breach, contract or regulatory consequences |
Score bands for the risk register:
- Low (1 to 6): accept or monitor, review annually.
- Medium (8 to 12): treat within the current planning cycle, review quarterly.
- High (15 to 25): treat now, report to management, review monthly until the score drops.
The SOC 2 risk management guide for hybrid and on-prem environments compares qualitative and quantitative scoring.
How do you fill in one risk register row? A worked example
A risk register row is complete when a reader who has never heard of the risk can tell what could happen, how bad the risk is, who owns the fix and when the owner checks again. The example below is illustrative, built for this template:
- Name the asset. Production cloud management console, the control point for every production workload.
- Name the threat. Account takeover of an administrator through phishing or credential stuffing.
- Name the vulnerability. Workforce SSO enforces MFA, but two break-glass administrator accounts sign in with a password only.
- Rate likelihood. 3 (possible). Credential phishing against engineers is common across the industry, and the two accounts are reachable from the internet.
- Rate impact. 5 (severe). An attacker with console access could delete backups and read customer data.
- Calculate the inherent score. 3 × 5 = 15, High.
- Choose the treatment. Mitigate: enforce phishing-resistant MFA on both break-glass accounts, store the credentials in the password vault, and alert the security channel on every break-glass sign-in.
- Assign the owner. Head of Infrastructure, with ticket SEC-142 for the work.
- Rate the residual score. 1 × 5 = 5, Low. Impact stays at 5 because a successful takeover would still be severe. Likelihood drops to 1.
- Set the review date and link the controls. Review on 2027-02-20. Link the Access Control Policy, the MFA configuration, the alert rule and ticket SEC-142, closed with a screenshot of the MFA setting.
Key insight
Most risk register controls lower likelihood and leave impact unchanged
Most controls lower likelihood, so a residual score with a lower impact needs a reason, for example encryption that makes stolen data unreadable.
What do mitigate, accept, transfer and avoid mean in a risk register?
Each risk treatment option in a risk register creates a different evidence trail, and an auditor follows that trail. Mitigate means something was done, so evidence of the work has to exist. Accept means a conscious decision was made, so a written justification has to exist.
| Treatment | Meaning | Evidence to keep |
| Mitigate | Add or improve controls to lower likelihood or impact | Ticket or task, completion date, proof the control works |
| Accept | Keep the risk unchanged, inside the risk appetite | Justification, approver, approval date, expiry or review date |
| Transfer | Shift the financial impact to another party | Cyber insurance policy or contract clause, with coverage limits |
| Avoid | Stop the activity that creates the risk | Decommission record or change record |
Watch out
Accepted risks in a risk register need an expiry date
An acceptance with no review date becomes a permanent exception nobody rechecks. The SOC 2 vulnerability risk acceptance guide covers the fields for accepting a vulnerability with no vendor fix.
Why should every risk live in one risk register?
A single risk register gives the team, management and the auditor one view of what has been identified, accepted, mitigated and remediated. When risks are spread across a GRC platform, a spreadsheet and informal notes, nobody has the full picture.
In one SOC 2 Type 1 preparation review, the security lead and the outside consultant found they had each created an MFA risk entry in different places without knowing about the other. Several risks flagged months earlier had been fully remediated but were still marked open, so the risk posture looked worse than the real state of the controls. The team agreed on one home for every risk, the GRC platform, because the auditor has direct visibility there.
Rules for keeping one risk register:
- One home. Every new risk goes into the same register, whoever finds the risk.
- Search before adding. Check for an existing row on the same asset and threat before creating a new one.
- Close on evidence. Move a mitigated risk to closed when the treatment ticket closes, with the evidence attached.
- Link, do not copy. Vulnerability tickets, vendor assessments and incident records link to the register row.
The SOC 2 CC3.2 guide covers what the criterion requires from the identification and analysis process behind the register.
How should a verbal not-affected call go into the risk register?
A verbal not-affected call becomes a risk decision only when the analysis is written down, dated and approved, either in the risk register or in a ticket linked to the register. Until then, the call is an undocumented risk acceptance, and the finding stays open whether or not the team is right.
In one audit engagement, the final report was due the next day and nearly every finding was remediated. Two items were still open, and neither needed technical work:
- A high-severity vulnerability. The team was confident the vulnerability did not affect them. The only record of the analysis was a verbal assurance on the review call, with closure deferred to a follow-up email.
- A vendor security assessment. The assessment was complete, but the security owner and business owner fields were empty.
A documented not-affected decision records:
- The vulnerability or risk, and the affected asset
- Why the risk does not apply, for example the vulnerable component is not installed or not reachable
- Who did the analysis, and the date
- Who approved the decision
- The review date, or the event that reopens the decision
Evidence items with named dates and owners were the ones that got done on that engagement. For ranking vulnerabilities by exposure and exploitability before they reach the register, see the risk-based vulnerability management guide.
Which risks belong in an information security risk register?
An information security risk register for a small or mid-sized company typically holds 5 to 15 risk scenarios, enough to cover the real threats and few enough for the team to review each risk in a quarterly session. Categories to cover:
- Identity and access: privileged account takeover, orphaned accounts after offboarding.
- Vulnerabilities: exposed services, vulnerabilities with no vendor fix.
- Vendors and third parties: a critical SaaS provider outage or breach. The third party risk management guide covers the vendor inventory and assessments that feed these rows.
- Fraud: 2 or 3 scenarios, such as management override of controls or misuse of customer payment data. The SOC 2 CC3.3 fraud risk guide explains why auditors check for fraud scenarios by name.
- Availability: loss of a cloud region, failed backups, ransomware.
- People: key-person dependency in a small engineering or operations team.
At one fintech company, the platform team tuned fraud detection rules continuously, but the risk assessment had no fraud scenarios. Once asked, the team named several scenarios handled informally for months.
How often should a risk register be reviewed?
A risk register should be reviewed on a fixed cadence and after significant events. My approach:
- Quarterly: the owners walk through every open row, update scores, close remediated risks and recheck accepted risks against their expiry.
- Event-triggered: a new product, a new critical vendor, a major architecture change or a security incident adds or rescores rows within weeks.
- Annually: management reviews the full register, the scoring scale and the risk appetite, with minutes as the record.
ISO/IEC 27001:2022 clause 8.2 calls for risk assessments at planned intervals and when significant changes occur, with the results kept as documented information.
What evidence does a risk register produce for SOC 2 and ISO 27001?
A risk register kept this way produces the evidence for SOC 2 and ISO 27001 risk assessment as a byproduct. The same rows, scores, treatment tickets and review minutes cover both frameworks:
| Framework | Requirement | What the risk register provides |
| SOC 2 CC3.2 (AICPA Trust Services Criteria) | Identify risks to objectives and analyze them to decide how to manage each risk | Risk rows with likelihood, impact, owner and treatment |
| SOC 2 CC3.3 | Consider the potential for fraud in the risk assessment | Fraud scenarios in the same register |
| ISO/IEC 27001:2022 clause 6.1.2 | Information security risk assessment, with risk owners | Scoring scale, risk rows, owner column |
| ISO/IEC 27001:2022 clause 6.1.3 | Risk treatment, Statement of Applicability, risk owner approval of residual risk | Treatment column, linked controls, approvals for accepted risks |
| ISO/IEC 27001:2022 clauses 8.2 and 8.3 | Risk assessment at planned intervals, treatment plan carried out | Review dates, quarterly minutes, closed treatment tickets |
For method guidance outside the frameworks, NIST SP 800-30 Rev. 1, Guide for Conducting Risk Assessments, describes threat, vulnerability, likelihood and impact analysis, and NIST IR 8286, Integrating Cybersecurity and Enterprise Risk Management, focuses on using risk registers to report cybersecurity risk to enterprise risk management.
How does a risk register fit into an effective security program?
A risk register is the working record of the risk management capability, one of the roughly 19 security capabilities in an effective security program. A cybersecurity risk assessment from Truvo Cyber builds the first register with the team and sets the scoring scale. For teams that want register updates tied to tickets and scanner output, GRC engineering services connect those sources to the GRC platform.
Build a risk register the team uses
Truvo builds risk management into an effective security program, with one register and dated review records.
Frequently asked questions
What is a risk register in cybersecurity?
A cybersecurity risk register is a table with one row per security risk, recording the asset at risk, the threat, the vulnerability, likelihood and impact scores, the treatment decision, the owner, the residual score and the next review date. The register is the record management and auditors use to see which risks the company has found and what was decided about each.
What should a risk register template include?
A risk register template should include 12 columns: risk ID, asset or process, threat, vulnerability, likelihood, impact, inherent score, treatment, owner, residual score, review date and linked controls. Many teams add status, date identified, category, and approver with approval date for accepted risks. Separate threat and vulnerability columns make the right control easier to choose.
What is the difference between inherent risk and residual risk?
Inherent risk is the score before treatment, likelihood multiplied by impact with current gaps in place. Residual risk is the score after the chosen treatment is in place. In a risk register, most controls lower likelihood, so impact usually stays the same. A residual score inside the risk appetite can be accepted by the risk owner with a review date.
Does SOC 2 require a risk register?
SOC 2 CC3.2 requires an entity to identify and analyze risks to the entity's objectives, and CC3.3 requires fraud to be considered. The criteria do not name a format, but auditors expect a risk register or equivalent record with scores, owners and treatment decisions. For a Type 1 audit, the risk assessment is one of the few evidence items auditors ask to see.
How many risks should a risk register have?
A small or mid-sized company typically keeps 5 to 15 risk scenarios in the register, with 2 or 3 fraud-specific scenarios. Fifteen rows or fewer cover the real threats and keep the quarterly review manageable. Larger companies or on-prem environments with physical and hardware risks often need more rows, grouped by category.
Ready to Start Your Compliance Journey?
Get a clear, actionable roadmap with our readiness assessment.
Contact Us
Sample SOC 2 Control List
Input your email to download the list.
About the Author
Former security architect for Bank of Canada and Payments Canada. 20+ years building compliance programs for critical infrastructure.