SOC 2 auditor interview prep sheet
For each Trust Services criterion: the questions to rehearse for the auditor’s walkthrough, the sentences in your own generated policies that answer them, the evidence to have open on screen, and the policy owners who should be in the room. Built in your browser from your generator answers; print it or download it as a Word prep pack.
How the sheet is built
- Questions. From the Policyseed question bank behind what auditors ask about SOC 2 policies, where a question’s answer points at the criterion. Where the bank has nothing for a criterion, a neutral “walk me through” question is built from the criterion’s own text and labeled as derived.
- Your sentences. The 22 policies are rendered from your saved generator answers, and the best-matching sentences in the policies mapped to the criterion are quoted word for word, with the policy, the section number and a link.
- Evidence and owners. From the Policyseed Trust Services Criteria crosswalk (each criterion has its own page, for example CC6.1): the evidence items for the criterion and the owner role of each policy it maps to.
Your prep sheet
This is the sample company, Northwind Cloud Inc. The quotes below come from its generated policies, not yours. Answer the generator questions (about ten minutes, in your browser) and come back: this page then quotes your own policies.
36 criteria, 76 questions. Built in your browser; nothing is uploaded.
Northwind Cloud Inc: auditor interview prep sheet. Sample company: these quotes come from the policies generated for Northwind Cloud Inc, a fictional company, not from yours. Answer the generator questions to quote your own policies.
CC1 Control environment
CC1.1 Integrity and ethical values
Leadership sets a clear expectation of honest, ethical behavior, writes it down, and holds everyone (including contractors) to it.
In the room: Security Owner (Dana Whitfield, CTO): P01, P02; People/HR Lead: P18
How do you make sure employees have read and agreed to the policies?
From the auditor question bank
Your policy says:
“People Operations retains evidence of screening, signed agreements, policy acknowledgments, training completion, role-change reviews and offboarding checklists for seven years after employment ends, or longer where law requires, in line with the Data Retention and Disposal Policy, so that these controls can be evidenced for any audit period.”
“All personnel acknowledge this policy and the Acceptable Use Policy in writing at hire, at each annual review and after each material revision.”
Walk me through how Northwind Cloud Inc handles integrity and ethical values, and show me signed Acceptable Use Policy acknowledgment forms for every employee and contractor.
Derived from the criterion text
Your policy says:
“No account, device or system access is issued until the individual has signed an employment or services agreement with confidentiality and intellectual property terms and has acknowledged the Information Security Policy and the Acceptable Use Policy in writing.”
“Contractors, agency workers and vendor staff with access to Northwind Cloud Inc systems meet the same screening, acknowledgment, training and offboarding requirements as employees.”
Evidence to have open
- Signed Acceptable Use Policy acknowledgment forms for every employee and contractor
- Employee handbook or code of conduct section referenced by the Information Security Policy
- Approved Information Security Policy with the approver's name and effective date on the revision table
Policies: P01 Information Security Policy (section 4), P02 Acceptable Use Policy (section 4), P18 Human Resources Security Policy (section 4)
CC1.2 Oversight by leadership
Someone independent of day-to-day operations (a board, an advisor, or the CEO for a small company) oversees how security controls are designed and whether they keep working.
In the room: Security Owner (Dana Whitfield, CTO): P01, P17
Walk me through how Northwind Cloud Inc handles oversight by leadership, and show me leadership or board meeting notes showing a security update as an agenda item at least twice a year.
Derived from the criterion text
Your policy says:
“The Security Owner updates the register within ten business days of any change: new risks from incidents, audits, vendor reviews or personnel reports; completed actions with evidence; re-scored risks; and expired acceptances.”
“At least annually, someone who does not own the register (a member of Executive Management, an internal reviewer or an external advisor) evaluates whether the assessment was complete, scores consistent and treatments carried out, and records the outcome.”
Show me memo or email appointing the Security Owner, signed by the approver, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“Where acceptance is proposed, the Security Owner summarizes the risk, controls, residual score and reason, obtains written approval from the authority in 4.5, records approver and expiry in the register and adds the expiry to the review calendar.”
“Engineering, People Operations and the Security Owner collect the evidence in 4.9 on the cadence set in the relevant policy, record results against mapped risks and raise a ticket for any control not operating; the Security Owner reviews results monthly and re-scores affected risks.”
Evidence to have open
- Leadership or board meeting notes showing a security update as an agenda item at least twice a year
- Memo or email appointing the Security Owner, signed by the approver
- Annual risk assessment summary presented to leadership with the date it was reviewed
Policies: P01 Information Security Policy (section 3), P17 Risk Assessment and Management Policy (section 5)
CC1.3 Structure, reporting lines and authority
The company defines who owns security, who reports to whom, and who is allowed to make security decisions, and keeps that picture current as the company grows.
In the room: Security Owner (Dana Whitfield, CTO): P01, P17
Walk me through how Northwind Cloud Inc handles structure, reporting lines and authority, and show me current org chart exported from the HR or identity system.
Derived from the criterion text
Your policy says:
“Every risk has one named owner accountable for the treatment decision and for keeping the entry current; treatment actions have owners and due dates in the ticketing system, and overdue actions on High and Critical risks are reported to Executive Management at the next quarterly review.”
“Engineering designs, builds and operates Northwind and its infrastructure on AWS and Vercel in line with these policies.”
Show me Roles and Responsibilities section of the Information Security Policy naming the Security Owner, Engineering Lead and approver, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“Engineering leads own the technical policies assigned to them in the policy register.”
“People Operations identifies risks from hiring, role changes, departures, training gaps and insider threat, owns treatment actions in those areas, and reflects risk responsibilities in role descriptions.”
Evidence to have open
- Current org chart exported from the HR or identity system
- Roles and Responsibilities section of the Information Security Policy naming the Security Owner, Engineering Lead and approver
- Policy owner column of the Policy Review Calendar showing one named owner per policy
Policies: P01 Information Security Policy (section 3), P17 Risk Assessment and Management Policy (section 3)
CC1.4 Competent people
The company hires people who can do the job, checks their background where the law allows, and trains them so they can carry out their security responsibilities.
In the room: People/HR Lead: P18
Do you conduct background checks and security training?
From the auditor question bank
Your policy says:
“All personnel complete security awareness training within 30 days of starting and at least every 12 months thereafter.”
“The Security Owner maintains the curriculum and assigns it to new starters within 30 days and to all personnel annually.”
Walk me through how Northwind Cloud Inc handles competent people, and show me job descriptions for security-relevant roles listing security responsibilities.
Derived from the criterion text
Your policy says:
“Every job description for a role with system access states its security responsibilities.”
“Every candidate for a role with access to Northwind systems, customer data or company finances is screened before their start date: identity verification, right-to-work confirmation, verification of the most recent employment, and a criminal record check where legally permitted in the candidate's jurisdiction and proportionate to the role.”
Evidence to have open
- Job descriptions for security-relevant roles listing security responsibilities
- Background check completion records or vendor confirmations for hires in the audit period
- Security awareness training completion report for all staff within 30 days of hire and annually
Policies: P18 Human Resources Security Policy (section 4, 5)
CC1.5 Accountability
People are held responsible for their security duties: expectations are written down, performance is reviewed, and policy violations have consequences.
In the room: Security Owner (Dana Whitfield, CTO): P01, P02; People/HR Lead: P18
How do you make sure employees have read and agreed to the policies?
From the auditor question bank
Your policy says:
“People Operations retains evidence of screening, signed agreements, policy acknowledgments, training completion, role-change reviews and offboarding checklists for seven years after employment ends, or longer where law requires, in line with the Data Retention and Disposal Policy, so that these controls can be evidenced for any audit period.”
“No account, device or system access is issued until the individual has signed an employment or services agreement with confidentiality and intellectual property terms and has acknowledged the Information Security Policy and the Acceptable Use Policy in writing.”
Walk me through how Northwind Cloud Inc handles accountability, and show me enforcement sections of the approved policies describing the disciplinary process.
Derived from the criterion text
Your policy says:
“Violations are handled by People Operations with the Security Owner and may result in retraining, loss of access, disciplinary action up to and including termination of employment or contract, and, where the law requires, referral to authorities.”
“People Operations and the Security Owner handle violations together.”
Evidence to have open
- Enforcement sections of the approved policies describing the disciplinary process
- Performance review template that includes adherence to company policies
- Signed policy acknowledgment forms collected at onboarding and after each annual review
Policies: P01 Information Security Policy (section 7), P02 Acceptable Use Policy (section 7), P18 Human Resources Security Policy (section 4)
CC2 Communication and information
CC2.1 Quality information
The company collects and keeps the information it needs to run its controls, such as an accurate asset inventory, logs and a risk register, and that information is reliable.
In the room: IT/Operations Lead: P05; Engineering Lead: P12; Security Owner (Dana Whitfield, CTO): P17
Walk me through how Northwind Cloud Inc handles quality information, and show me asset inventory export (MDM device list plus cloud resource inventory) with owner and classification columns.
Derived from the criterion text
Your policy says:
“Northwind Cloud Inc maintains an asset register covering hardware, cloud resources, software and services, and data assets.”
“Every resource in AWS and Vercel is created through infrastructure-as-code or a documented change and carries tags for owner, environment, service and data classification.”
Show me log retention configuration screenshot from the logging tool showing retention of at least 12 months for security logs, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“Security and audit logs are retained for at least 12 months with the most recent 90 days searchable without restoration; application debug logs are retained for 30 days.”
“Company devices are procured through the IT/Operations Lead, configured to the baseline in the Endpoint and Workstation Security Policy before handover, enrolled in Kandji so that encryption, screen lock, update and inventory policies apply automatically, and recorded in the register with the assigned user on the day they are issued.”
Evidence to have open
- Asset inventory export (MDM device list plus cloud resource inventory) with owner and classification columns
- Log retention configuration screenshot from the logging tool showing retention of at least 12 months for security logs
- Risk register spreadsheet with last-updated date and Security Owner sign-off
Policies: P05 Asset Management Policy (section 4), P12 Logging and Monitoring Policy (section 4), P17 Risk Assessment and Management Policy (section 5)
CC2.2 Internal communication
Staff know the security policies that apply to them, how to report a problem, and what their responsibilities are, because the company tells them at hire and on an ongoing basis.
In the room: Security Owner (Dana Whitfield, CTO): P01, P13; People/HR Lead: P18
How do you make sure employees have read and agreed to the policies?
From the auditor question bank
Your policy says:
“All personnel acknowledge the Information Security Policy, the Acceptable Use Policy and this policy within five business days of starting and within 30 days of each revision.”
“Three business days before the start date, People Operations opens an onboarding ticket listing role, manager, start date, device requirement and access profile.”
Walk me through how Northwind Cloud Inc handles internal communication, and show me onboarding checklist showing policy distribution and security training as required steps.
Derived from the criterion text
Your policy says:
“People Operations collects acknowledgment from every new hire before access is granted and from all personnel at each annual review, assigns awareness training within thirty days of hire and annually thereafter, and reports outstanding items to the Security Owner monthly.”
“Security Owner (Dana Whitfield, CTO) owns the security content of the onboarding and offboarding checklists, defines the training curriculum, reviews access reconciliations and handles escalations of suspected policy violations.”
Evidence to have open
- Onboarding checklist showing policy distribution and security training as required steps
- Signed policy acknowledgment forms for every active employee
- Screenshot of the internal incident reporting channel or address referenced in the Incident Response Policy
- Announcement to staff when a policy is updated (email or chat message with date)
Policies: P01 Information Security Policy (section 5), P13 Incident Response Policy (section 5), P18 Human Resources Security Policy (section 5)
CC2.3 External communication
Customers, vendors and other outside parties can learn how the company handles security and privacy, and know how to reach it to report a concern or receive an incident notice.
In the room: Security Owner (Dana Whitfield, CTO): P13, P16, P22
How would customers be notified of a security incident affecting their data?
From the auditor question bank
Your policy says:
“The Security Owner assesses the risk to individuals, notifies supervisory authorities where the law requires (within 72 hours where the GDPR or a similar law applies), notifies affected customers within their contractual period and no later than 72 hours after confirmation, and notifies affected individuals where the risk to them is high.”
“Every vendor that processes personal data for Northwind Cloud Inc is bound by a written data processing agreement covering the scope and purpose of processing, security measures, onward sub-processing, and assistance with data subject requests and breach notification.”
Walk me through how Northwind Cloud Inc handles external communication, and show me public security page or trust page URL and a dated screenshot.
Derived from the criterion text
Your policy says:
“Northwind Cloud Inc monitors critical vendors between reviews through status pages, security advisories and published sub-processor changes, and treats any vendor breach notification as an incident under the Incident Response Policy.”
“Incidents arrive from Datadog alerts, AWS and Vercel security notifications, secret-scanning and dependency alerts from GitHub, vendor notices, customer reports and personnel reports to security@northwindcloud.example.”
Evidence to have open
- Public security page or trust page URL and a dated screenshot
- Published privacy notice with the date it was last updated
- Customer incident notification template from the Incident Response Policy procedures
- Contract or DPA clause stating customer breach notification timelines
Policies: P13 Incident Response Policy (section 5), P16 Vendor and Third-Party Risk Management Policy (section 4), P22 Privacy and Data Protection Policy (section 4)
CC3 Risk assessment
CC3.1 Objectives for risk assessment
The company states clearly what it is trying to protect and which commitments (availability, confidentiality, contractual, legal) its risk assessment must address, so risks can be measured against something concrete.
In the room: Security Owner (Dana Whitfield, CTO): P17, P01
Have you performed a risk assessment, and how is it documented?
From the auditor question bank
Your policy says:
“Control operation is monitored through defined evidence: access reviews, vulnerability scans, backup and restore tests, incident metrics, vendor reviews and training completion.”
“At least annually, someone who does not own the register (a member of Executive Management, an internal reviewer or an external advisor) evaluates whether the assessment was complete, scores consistent and treatments carried out, and records the outcome.”
Walk me through how Northwind Cloud Inc handles objectives for risk assessment, and show me risk assessment scope statement naming the in-scope product, environments and Trust Services Categories.
Derived from the criterion text
Your policy says:
“The program exists to protect the confidentiality, integrity and availability of customer data and company systems; to meet the commitments Northwind Cloud Inc makes in contracts, its terms of service and its privacy notice; to satisfy the Security and Availability criteria selected for the SOC 2 examination; and to keep risk within the tolerance set by Executive Management.”
“The Security Owner maintains a mapping from each risk to the policies and controls that treat it and to the applicable Trust Services Criteria, so that the control set can be shown to follow from assessed risks and gaps are visible.”
Evidence to have open
- Risk assessment scope statement naming the in-scope product, environments and Trust Services Categories
- Information Security Policy section listing the company's security objectives and commitments
- System description or architecture diagram used as the input to the annual risk assessment
Policies: P17 Risk Assessment and Management Policy (section 4), P01 Information Security Policy (section 4)
CC3.2 Identifying and analyzing risks
At least once a year the company lists the threats to its systems and data (including from vendors), rates their likelihood and impact, and decides how to treat each one.
In the room: Security Owner (Dana Whitfield, CTO): P17, P16
Walk me through how Northwind Cloud Inc handles identifying and analyzing risks, and show me completed annual risk register with likelihood, impact, treatment and owner columns.
Derived from the criterion text
Your policy says:
“Once a year the Security Owner arranges the evaluation required by 4.13, provides the register, treatment evidence and monitoring results, records findings as risks or actions and presents the evaluation to Executive Management.”
“Every assessment identifies assets and owners, threats and vulnerabilities, existing controls, and likelihood and impact on the Section 2 scales, considering external attack, internal misuse, error, vendor failure, environmental events, legal change and fraud, including misuse of privileged access and manipulation of financial or customer records.”
Show me vendor risk assessment records for critical vendors (SOC 2 report review notes or security questionnaire), and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“For Tier 1 vendors, the Security Owner obtains the current SOC 2 Type II report or ISO 27001 certificate, reviews scope, opinion, exceptions and complementary user entity controls, sends the Northwind Cloud Inc questionnaire where gaps remain and documents the data flow with Engineering; Tier 2 vendors need an assurance report or completed questionnaire.”
“Participants agree inherent likelihood and impact on the Section 2 scales, list existing controls and agree residual scores; the Security Owner challenges scores inconsistent with comparable risks or incident history and records the rationale.”
Evidence to have open
- Completed annual risk register with likelihood, impact, treatment and owner columns
- Vendor risk assessment records for critical vendors (SOC 2 report review notes or security questionnaire)
- Meeting notes or ticket showing leadership approval of the risk treatment plan
Policies: P17 Risk Assessment and Management Policy (section 5), P16 Vendor and Third-Party Risk Management Policy (section 5)
CC3.3 Fraud risk
The risk assessment specifically considers how someone inside or outside the company could commit fraud, including misuse of privileged access or manipulation of data.
In the room: Security Owner (Dana Whitfield, CTO): P17, P03; People/HR Lead: P18
Walk me through how Northwind Cloud Inc handles fraud risk, and show me risk register entries tagged as fraud or insider-threat risks with their treatments.
Derived from the criterion text
Your policy says:
“Every assessment identifies assets and owners, threats and vulnerabilities, existing controls, and likelihood and impact on the Section 2 scales, considering external attack, internal misuse, error, vendor failure, environmental events, legal change and fraud, including misuse of privileged access and manipulation of financial or customer records.”
“At least annually, someone who does not own the register (a member of Executive Management, an internal reviewer or an external advisor) evaluates whether the assessment was complete, scores consistent and treatments carried out, and records the outcome.”
Show me segregation-of-duties statement in the Access Control Policy and a screenshot showing production deploys require a second approver, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“When monitoring shows a control has failed, the related risk is re-scored and, if it reaches High or Critical, a triggered assessment is opened.”
“Access to GitHub is granted through organization membership enforced by single sign-on with Okta, repository permissions are assigned by team, and protected branches require a review from someone other than the author before code is merged and deployed by GitHub Actions.”
Evidence to have open
- Risk register entries tagged as fraud or insider-threat risks with their treatments
- Segregation-of-duties statement in the Access Control Policy and a screenshot showing production deploys require a second approver
- Background check policy statement and completion records for roles with privileged access
Policies: P17 Risk Assessment and Management Policy (section 4), P03 Access Control Policy (section 4), P18 Human Resources Security Policy (section 4)
CC3.4 Changes that affect risk
When something significant changes, such as a new cloud region, a major feature, a new vendor, key staff leaving or a new regulation, the company re-examines its risks rather than waiting for the next annual cycle.
In the room: Security Owner (Dana Whitfield, CTO): P17; Engineering Lead: P09
Have you performed a risk assessment, and how is it documented?
From the auditor question bank
Your policy says:
“Participants agree inherent likelihood and impact on the Section 2 scales, list existing controls and agree residual scores; the Security Owner challenges scores inconsistent with comparable risks or incident history and records the rationale.”
“The Security Owner updates the register within ten business days of any change: new risks from incidents, audits, vendor reviews or personnel reports; completed actions with evidence; re-scored risks; and expired acceptances.”
Walk me through how Northwind Cloud Inc handles changes that affect risk, and show me risk register revision history showing an update tied to a major change in the period.
Derived from the criterion text
Your policy says:
“The Security Owner reviews this policy on the annual review cycle and after any significant change to Northwind Cloud Inc's source control, pipeline tooling or cloud platforms, and after any incident caused by a change.”
“The Security Owner reviews every High and Critical risk with its owner, confirms progress, re-scores where circumstances have changed, chases overdue actions, checks acceptances expiring next quarter, reviews monitoring evidence and prepares the report required by 4.11.”
Evidence to have open
- Risk register revision history showing an update tied to a major change in the period
- Change ticket for a significant architecture change with a risk review field completed
- Policy Review Calendar entry showing an out-of-cycle review triggered by a change
Policies: P17 Risk Assessment and Management Policy (section 5), P09 Change Management Policy (section 4)
CC4 Monitoring activities
CC4.1 Ongoing and separate evaluations
The company checks that its controls actually work, through routine monitoring (scans, reviews) and periodic separate checks such as a penetration test or internal audit.
In the room: Security Owner (Dana Whitfield, CTO): P17, P01; Engineering Lead: P11
Do you perform periodic access reviews?
From the auditor question bank
Your policy says:
“Security Owner (Dana Whitfield, CTO) owns this policy, maintains the role-to-access matrix, runs the periodic access reviews, approves privileged and production access, and holds the evidence of every grant, review and removal.”
“Where acceptance is proposed, the Security Owner summarizes the risk, controls, residual score and reason, obtains written approval from the authority in 4.5, records approver and expiry in the register and adds the expiry to the review calendar.”
Walk me through how Northwind Cloud Inc handles ongoing and separate evaluations, and show me annual internal control self-assessment using the Evidence Checklist, signed by the Security Owner.
Derived from the criterion text
Your policy says:
“Once a year the Security Owner arranges the evaluation required by 4.13, provides the register, treatment evidence and monitoring results, records findings as risks or actions and presents the evaluation to Executive Management.”
“Control operation is monitored through defined evidence: access reviews, vulnerability scans, backup and restore tests, incident metrics, vendor reviews and training completion.”
Evidence to have open
- Annual internal control self-assessment using the Evidence Checklist, signed by the Security Owner
- Most recent penetration test report and remediation tracker
- Vulnerability scan reports from the last quarter showing scan dates and findings
- Quarterly access review spreadsheet signed by the Security Owner
Policies: P17 Risk Assessment and Management Policy (section 5), P11 Vulnerability and Patch Management Policy (section 5), P01 Information Security Policy (section 8)
CC4.2 Communicating deficiencies
When a control gap or weakness is found, it is reported to the people who can fix it and to leadership, tracked to closure, and not quietly dropped.
In the room: Security Owner (Dana Whitfield, CTO): P17, P13, P01
Walk me through how Northwind Cloud Inc handles communicating deficiencies, and show me remediation tracker (spreadsheet or issue board) with owner, due date and status for each open finding.
Derived from the criterion text
Your policy says:
“Findings are tracked to closure in the risk register with an owner and due date.”
“Every SEV1 and SEV2 incident receives a blameless post-incident review within five business days of closure that produces a root cause, impact statement and corrective actions with owners and due dates, tracked to closure and reported to Executive Management.”
Show me incident post-mortem documents with action items and their completion dates, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“Within five business days of closing a SEV1 or SEV2 incident, the Incident Commander convenes a blameless review covering the timeline, root cause, response timings, impact, what went well and badly, and corrective actions with owners and due dates; the Security Owner approves it and presents SEV1 reviews to Executive Management.”
“The Security Owner updates the register within ten business days of any change: new risks from incidents, audits, vendor reviews or personnel reports; completed actions with evidence; re-scored risks; and expired acceptances.”
Evidence to have open
- Remediation tracker (spreadsheet or issue board) with owner, due date and status for each open finding
- Incident post-mortem documents with action items and their completion dates
- Leadership update summarizing open security findings and their status
Policies: P17 Risk Assessment and Management Policy (section 5), P13 Incident Response Policy (section 5), P01 Information Security Policy (section 5)
CC5 Control activities
CC5.1 Selecting control activities
The company chooses controls that match the risks it identified, covering people, process and technology, and documents which control addresses which risk.
In the room: Security Owner (Dana Whitfield, CTO): P01, P17
Have you performed a risk assessment, and how is it documented?
From the auditor question bank
Your policy says:
“Participants agree inherent likelihood and impact on the Section 2 scales, list existing controls and agree residual scores; the Security Owner challenges scores inconsistent with comparable risks or incident history and records the rationale.”
“The Security Owner updates the register within ten business days of any change: new risks from incidents, audits, vendor reviews or personnel reports; completed actions with evidence; re-scored risks; and expired acceptances.”
Walk me through how Northwind Cloud Inc handles selecting control activities, and show me TSC Crosswalk mapping each criterion to the policy section that addresses it.
Derived from the criterion text
Your policy says:
“Engineering, People Operations and the Security Owner collect the evidence in 4.9 on the cadence set in the relevant policy, record results against mapped risks and raise a ticket for any control not operating; the Security Owner reviews results monthly and re-scores affected risks.”
“Controls are selected and prioritized according to the risks they reduce, as recorded in the risk assessment maintained under the Risk Assessment and Management Policy, not solely because a framework lists them.”
Evidence to have open
- TSC Crosswalk mapping each criterion to the policy section that addresses it
- Risk register treatment column linking each risk to a control or policy
- Approved policy set with version numbers and effective dates
Policies: P01 Information Security Policy (section 4), P17 Risk Assessment and Management Policy (section 5)
CC5.2 Technology controls
General controls over technology are in place: access to systems is restricted, changes are managed, and infrastructure is configured and maintained securely.
In the room: Security Owner (Dana Whitfield, CTO): P03; Engineering Lead: P09, P20
Walk me through how Northwind Cloud Inc handles technology controls, and show me cloud IAM configuration export showing role-based groups rather than individual grants.
Derived from the criterion text
Your policy says:
“Production access to AWS and Vercel is granted only to engineers whose role requires it, through federated identity and short-lived credentials rather than long-lived access keys.”
“Access is reviewed at least quarterly for production, source code, identity administration and any system holding Restricted data, and at each annual review for all other systems.”
Show me branch protection settings export for production repositories, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“All production code and infrastructure-as-code lives in GitHub, and production reflects only what has been merged to the protected default or release branch.”
“Access to GitHub is granted through organization membership enforced by single sign-on with Okta, repository permissions are assigned by team, and protected branches require a review from someone other than the author before code is merged and deployed by GitHub Actions.”
Evidence to have open
- Cloud IAM configuration export showing role-based groups rather than individual grants
- Branch protection settings export for production repositories
- Infrastructure-as-code repository with pull request history for infrastructure changes
Policies: P03 Access Control Policy (section 4), P09 Change Management Policy (section 4), P20 Network and Infrastructure Security Policy (section 4)
CC5.3 Policies and procedures
Controls are written into approved policies and step-by-step procedures, staff know about them, owners are assigned, and the documents are reviewed on a schedule.
In the room: Security Owner (Dana Whitfield, CTO): P01; People/HR Lead: P18
How do you make sure employees have read and agreed to the policies?
From the auditor question bank
Your policy says:
“All personnel acknowledge the Information Security Policy, the Acceptable Use Policy and this policy within five business days of starting and within 30 days of each revision.”
“Three business days before the start date, People Operations opens an onboarding ticket listing role, manager, start date, device requirement and access profile.”
How often are policies reviewed, and can you show the last review?
From the auditor question bank
Your policy says:
“Each review is recorded in section 9 even where nothing changes; revisions are approved by Priya Natarajan, CEO before they take effect.”
“The Security Owner maintains a control matrix mapping each criterion in scope (Security and Availability) to the policy statements and evidence artifacts that satisfy it.”
Evidence to have open
- Complete policy set with approver, version and effective date on every revision history table
- Policy Review Calendar with owner and next review date for each policy
- Signed policy acknowledgment forms for all staff
- Record of the last annual policy review (meeting notes or approval email)
Policies: P01 Information Security Policy (section 5, 8), P18 Human Resources Security Policy (section 5)
CC6 Logical and physical access controls
CC6.1 Logical access security
The company protects its systems and data with identity, authentication, authorization, encryption and network controls that are designed around what needs protecting.
In the room: Security Owner (Dana Whitfield, CTO): P03, P04; Engineering Lead: P08, P20
How is access to production systems granted, and who approves it?
From the auditor question bank
Your policy says:
“Access requests are made in the ticketing system, name the system and role required, and are approved by the requester's manager and the system owner before provisioning.”
“Access is reviewed at least quarterly for production, source code, identity administration and any system holding Restricted data, and at each annual review for all other systems.”
Is multi-factor authentication enforced, and where?
From the auditor question bank
Your policy says:
“Multi-factor authentication is enforced for every workforce account, with phishing-resistant methods for administrative roles, under the Authentication and Password Policy.”
“Northwind protects customer accounts to at least the same standard: customer passwords are stored with a slow adaptive hash such as Argon2id or bcrypt and a per-user salt, are never logged or transmitted in plain text, are checked against breached-password lists, and customers are offered multi-factor authentication.”
Is data encrypted at rest and in transit, and how are keys managed?
From the auditor question bank
Your policy says:
“All production data at rest on AWS and Vercel, including databases, object storage, volumes, snapshots, backups and log storage, is encrypted with AES-256 or a provider-equivalent algorithm, enabled before any data is written.”
“Data classified Confidential or Restricted under the Data Classification and Handling Policy is encrypted at rest and in transit without exception, including when exported, emailed, shared with vendors or backed up.”
Evidence to have open
- Identity provider MFA enforcement policy screenshot (for example the Okta or Google Workspace admin console)
- Cloud IAM role and group listing for the production account
- Encryption-at-rest settings screenshot for the production database and object storage
- Password manager admin report showing enrollment of all staff
Policies: P03 Access Control Policy (section 4), P04 Authentication and Password Policy (section 4), P08 Encryption and Key Management Policy (section 4), P20 Network and Infrastructure Security Policy (section 4)
CC6.2 User registration and deprovisioning
New users are created only after approval, credentials are issued to the right person, and access is removed promptly when someone leaves or no longer needs it.
In the room: Security Owner (Dana Whitfield, CTO): P03; People/HR Lead: P18
Do you perform periodic access reviews?
From the auditor question bank
Your policy says:
“Security Owner (Dana Whitfield, CTO) owns this policy, maintains the role-to-access matrix, runs the periodic access reviews, approves privileged and production access, and holds the evidence of every grant, review and removal.”
“Owners respond within ten business days, removals are completed within five business days of the response, and the signed-off review is retained as evidence.”
What happens when an employee or contractor leaves?
From the auditor question bank
Your policy says:
“The device is collected and wiped through Kandji, and the timestamped checklist is stored with the personnel record.”
“When People Operations confirms a departure, Engineering disables the Okta account, revokes sessions and tokens, removes membership of GitHub and AWS and Vercel, and rotates any shared or break-glass credentials the leaver could have known.”
Evidence to have open
- Onboarding access request tickets showing manager approval before account creation
- Offboarding checklist and identity provider deactivation timestamp for each leaver in the period
- Quarterly access review spreadsheet comparing active accounts to the HR roster, signed by the Security Owner
Policies: P03 Access Control Policy (section 5), P18 Human Resources Security Policy (section 5)
CC6.3 Role-based access and least privilege
Access is granted by role and limited to what the job requires, privileged access is restricted, and permissions are reviewed and adjusted on a schedule.
In the room: Security Owner (Dana Whitfield, CTO): P03
How is access to production systems granted, and who approves it?
From the auditor question bank
Your policy says:
“Requests beyond the bundle are raised as tickets, approved by the manager and the system owner (the Engineering lead for production, the Security Owner for identity and security tooling), and closed with a note of the permission granted.”
“Access requests are made in the ticketing system, name the system and role required, and are approved by the requester's manager and the system owner before provisioning.”
Do you perform periodic access reviews?
From the auditor question bank
Your policy says:
“Security Owner (Dana Whitfield, CTO) owns this policy, maintains the role-to-access matrix, runs the periodic access reviews, approves privileged and production access, and holds the evidence of every grant, review and removal.”
“Owners respond within ten business days, removals are completed within five business days of the response, and the signed-off review is retained as evidence.”
Evidence to have open
- Role-to-permission matrix for production systems (identity provider groups, cloud IAM roles, source control teams)
- Quarterly access review spreadsheet with removals noted and Security Owner sign-off
- Identity provider admin role listing showing a small named set of administrators
- Access change tickets for privilege escalations in the period with approval recorded
Policies: P03 Access Control Policy (section 4, 5)
CC6.4 Physical access
Physical access to places where systems and data live (data centers, offices, home offices) is restricted to authorized people and removed when no longer needed.
In the room: IT/Operations Lead: P21
How do you address physical security, for offices and for remote staff?
From the auditor question bank
Your policy says:
“All production systems for Northwind run in facilities operated by AWS and Vercel, and Northwind Cloud Inc relies on those providers' physical and environmental controls (access control, surveillance, power, cooling and fire suppression) as described in their independent assurance reports.”
“Where Northwind Cloud Inc maintains an office, entry is restricted to authorized personnel through keys, badges or a managed access system, with access issued at onboarding and revoked on the last working day.”
Walk me through how Northwind Cloud Inc handles physical access, and show me cloud provider SOC 2 Type II report (or bridge letter) covering data center physical security.
Derived from the criterion text
Your policy says:
“Security Owner (Dana Whitfield, CTO) approves office access arrangements, reviews provider data center assurance, handles reports of physical security incidents and approves exceptions.”
“Company devices are never left unattended in public places, in vehicles or in unsecured shared spaces.”
Evidence to have open
- Cloud provider SOC 2 Type II report (or bridge letter) covering data center physical security
- Office access badge or key holder list reconciled against the HR roster
- Remote work security statement acknowledged by remote staff
Policies: P21 Physical and Remote Work Security Policy (section 4)
CC6.5 Disposal of physical assets
When laptops, drives or other equipment leave the company's control, the data on them is wiped or destroyed first, and the disposal is recorded.
In the room: IT/Operations Lead: P05, P21; Security Owner (Dana Whitfield, CTO): P07
What happens when an employee or contractor leaves?
From the auditor question bank
Your policy says:
“People Operations includes device return in the offboarding checklist and provides a prepaid shipping label for remote staff.”
“When an engagement ends, the Okta account is disabled and sessions revoked before the end of the last working day (immediately for involuntary departures or leave pending investigation), and all remaining access is removed within one business day, as set out in the Access Control Policy.”
How long do you retain data, and how do you dispose of it?
From the auditor question bank
Your policy says:
“Engineering configures lifecycle rules, time-to-live settings or scheduled deletion jobs for each category and documents them alongside the schedule; changes follow the Change Management Policy.”
“When a service, environment or account on AWS and Vercel is retired, Engineering completes a checklist: delete volumes and buckets, expire snapshots and backups, revoke credentials and keys, remove DNS and monitoring, and update the asset inventory.”
How do you address physical security, for offices and for remote staff?
From the auditor question bank
Your policy says:
“All production systems for Northwind run in facilities operated by AWS and Vercel, and Northwind Cloud Inc relies on those providers' physical and environmental controls (access control, surveillance, power, cooling and fire suppression) as described in their independent assurance reports.”
“At onboarding, and whenever a person's primary work location changes, People Operations issues the home-office checklist covering workspace privacy, Wi-Fi security, router configuration and equipment storage.”
Evidence to have open
- MDM remote-wipe confirmation for laptops returned by leavers in the period
- Asset inventory entries marked disposed with the wipe date and method
- Certificate of destruction from the recycling or ITAD vendor, where used
Policies: P05 Asset Management Policy (section 5), P07 Data Retention and Disposal Policy (section 5), P21 Physical and Remote Work Security Policy (section 5)
CC6.6 Protection from external threats
The boundary between the company's systems and the outside world is defended: firewalls, WAF, secure remote access, MFA on external entry points, and hardened endpoints.
In the room: Engineering Lead: P20; Security Owner (Dana Whitfield, CTO): P04; IT/Operations Lead: P19
Is multi-factor authentication enforced, and where?
From the auditor question bank
Your policy says:
“Northwind protects customer accounts to at least the same standard: customer passwords are stored with a slow adaptive hash such as Argon2id or bcrypt and a per-user salt, are never logged or transmitted in plain text, are checked against breached-password lists, and customers are offered multi-factor authentication.”
“Multi-factor authentication is mandatory for every workforce account at Northwind Cloud Inc, without exception for seniority.”
How are laptops and other endpoints secured?
From the auditor question bank
Your policy says:
“The loss, theft or suspected compromise of any endpoint is reported to security@northwindcloud.example within one hour of discovery, as the Incident Response Policy requires.”
“Kandji enforces the configuration in Sections 4.4 to 4.8 and reports compliance to the Security Owner.”
How is your network and cloud infrastructure secured?
From the auditor question bank
Your policy says:
“Vercel: workloads sit in private subnets or equivalent private segments with no direct inbound route from the internet; databases and caches accept connections only from application segments; security groups and firewall rules default to deny and allow only documented ports and sources; and a web application firewall or equivalent edge protection fronts every public HTTP endpoint.”
“Every compute image, container base image and managed service is configured against a documented hardening baseline informed by provider best practices and CIS benchmarks: unnecessary services disabled, default credentials removed, instance metadata access restricted and logging enabled.”
Evidence to have open
- Cloud security group or firewall rule export for production showing no unrestricted inbound admin ports
- WAF or edge protection configuration screenshot
- MFA enforcement screenshot for the identity provider, source control organization and cloud console
- MDM compliance report showing disk encryption and firewall enabled on all managed devices
Policies: P20 Network and Infrastructure Security Policy (section 4), P04 Authentication and Password Policy (section 4), P19 Endpoint and Workstation Security Policy (section 4)
CC6.7 Data in transit and on removable media
Data is protected while it moves: encrypted in transit, restricted from being copied to unmanaged devices or removable media, and shared only through approved channels.
In the room: Engineering Lead: P08; Security Owner (Dana Whitfield, CTO): P06; IT/Operations Lead: P19
How do you classify data, and how is customer data handled?
From the auditor question bank
Your policy says:
“Northwind Cloud Inc classifies information into four levels.”
“Personal data is collected only for the purposes in the privacy notice of Northwind Cloud Inc, limited to what those purposes require, not repurposed without a lawful basis, and handled under the Privacy and Data Protection Policy.”
Is data encrypted at rest and in transit, and how are keys managed?
From the auditor question bank
Your policy says:
“All production data at rest on AWS and Vercel, including databases, object storage, volumes, snapshots, backups and log storage, is encrypted with AES-256 or a provider-equivalent algorithm, enabled before any data is written.”
“Restricted information is encrypted at rest and in transit under the Encryption and Key Management Policy.”
How are laptops and other endpoints secured?
From the auditor question bank
Your policy says:
“All laptops and mobile devices used for company work have full-disk encryption enabled, enforced and reported through Kandji.”
“The loss, theft or suspected compromise of any endpoint is reported to security@northwindcloud.example within one hour of discovery, as the Incident Response Policy requires.”
Evidence to have open
- TLS configuration scan result for public endpoints (for example an SSL Labs report) showing TLS 1.2 or higher only
- Data handling matrix from the Data Classification Policy listing approved sharing channels per classification
- MDM policy screenshot restricting USB storage or requiring encrypted external drives
Policies: P08 Encryption and Key Management Policy (section 4), P06 Data Classification and Handling Policy (section 4), P19 Endpoint and Workstation Security Policy (section 4)
CC6.8 Malicious software
The company prevents and detects unauthorized or malicious software on endpoints and in its software supply chain, and keeps systems patched.
In the room: IT/Operations Lead: P19; Engineering Lead: P11, P10
How do you find and fix vulnerabilities, and what are your remediation timelines?
From the auditor question bank
Your policy says:
“All findings from every source are tracked in a single vulnerability register with asset, severity, source, owner, detection date, due date, status and closure evidence.”
“Penetration test findings enter the register at the tester's severity, adjusted under 4.2 where warranted, are remediated within the 4.3 timelines, and Critical and High findings are retested by the tester before closure.”
How are laptops and other endpoints secured?
From the auditor question bank
Your policy says:
“The loss, theft or suspected compromise of any endpoint is reported to security@northwindcloud.example within one hour of discovery, as the Incident Response Policy requires.”
“Kandji enforces the configuration in Sections 4.4 to 4.8 and reports compliance to the Security Owner.”
Evidence to have open
- Endpoint protection or EDR console report showing coverage of all managed devices
- MDM configuration restricting software installation to approved sources
- Dependency scanning configuration and recent results from the source control platform
- Patch compliance report from MDM for operating system updates
Policies: P19 Endpoint and Workstation Security Policy (section 4), P11 Vulnerability and Patch Management Policy (section 4), P10 Secure Software Development Policy (section 4)
CC7 System operations
CC7.1 Detecting vulnerabilities and configuration changes
The company watches for new vulnerabilities and for unexpected changes to its infrastructure configuration, and fixes what it finds within defined timeframes.
In the room: Engineering Lead: P11, P12, P20
How do you find and fix vulnerabilities, and what are your remediation timelines?
From the auditor question bank
Your policy says:
“All findings from every source are tracked in a single vulnerability register with asset, severity, source, owner, detection date, due date, status and closure evidence.”
“The Security Owner engages an independent penetration testing firm annually, tracks findings to closure in the vulnerability register and retains the report and remediation evidence.”
How is your network and cloud infrastructure secured?
From the auditor question bank
Your policy says:
“Each quarter Engineering exports every rule permitting inbound traffic from the internet, plus all public IP addresses, buckets and endpoints, and reviews them with the Security Owner.”
“Vercel: workloads sit in private subnets or equivalent private segments with no direct inbound route from the internet; databases and caches accept connections only from application segments; security groups and firewall rules default to deny and allow only documented ports and sources; and a web application firewall or equivalent edge protection fronts every public HTTP endpoint.”
Evidence to have open
- Vulnerability scanner configuration and last quarter's scan reports
- Remediation tracker showing critical and high findings closed within the policy's SLA
- Cloud configuration monitoring alerts (for example AWS Config or GCP Security Command Center) export
- Infrastructure-as-code drift detection output or pull request history for the period
Policies: P11 Vulnerability and Patch Management Policy (section 4, 5), P12 Logging and Monitoring Policy (section 4), P20 Network and Infrastructure Security Policy (section 5)
CC7.2 Monitoring for anomalies
Systems and logs are monitored for unusual activity that could indicate a security event, and alerts reach a person who is expected to act on them.
In the room: Engineering Lead: P12; Security Owner (Dana Whitfield, CTO): P13
What logging and monitoring is in place, and who responds to alerts?
From the auditor question bank
Your policy says:
“Alerts are acknowledged and triaged within defined timelines: Critical within 1 hour at any time; High within 4 business hours; Medium within 2 business days; Low at the weekly review.”
“Before a new service, cloud account, vendor or environment goes live, the change record includes a logging checklist confirming audit logging is enabled, forwarding is verified with a test event, sensitive fields are masked, retention is set and alert rules cover the new source.”
Walk me through how Northwind Cloud Inc handles monitoring for anomalies, and show me logging tool alert rules export (for example Datadog monitors) covering authentication failures, privilege changes and production errors.
Derived from the criterion text
Your policy says:
“Alert rules exist for at least: repeated authentication failures or credential stuffing; multi-factor authentication disabled or reset; a new administrator or privilege escalation; changes to cloud identity policies or roles; use of a root, owner or break-glass account; a security group or firewall rule opened to the internet; a storage bucket made public; a secret detected in GitHub; logging disabled; unusual data export or deletion volume; changes to GitHub Actions configuration or deployment credentials; a device falling out of compliance in Kandji; and any endpoint security detection.”
“At minimum the following are logged: for each platform in AWS and Vercel, the provider's management and API audit log, identity and access changes, network flow or firewall logs where available, and access logs for storage holding customer data; Okta sign-in events, multi-factor enrollment changes and administrator actions; the organization audit log of GitHub and repository security setting changes; GitHub Actions runs and configuration changes; Northwind authentication successes and failures, password and multi-factor changes, authorization failures, privileged actions, API key creation and revocation, data exports and deletions, and security setting changes; Kandji enrollment and compliance changes; endpoint security agent events; and vendor administrator audit logs where available.”
Evidence to have open
- Logging tool alert rules export (for example Datadog monitors) covering authentication failures, privilege changes and production errors
- Sample alert notification with timestamp and the on-call response
- Centralized logging dashboard screenshot showing production log sources
- On-call rotation schedule or escalation configuration
Policies: P12 Logging and Monitoring Policy (section 4, 5), P13 Incident Response Policy (section 4)
CC7.3 Evaluating security events
When something suspicious is detected, the company decides whether it is a real incident, how serious it is, and whether a control failed.
In the room: Security Owner (Dana Whitfield, CTO): P13; Engineering Lead: P12
Do you have an incident response plan, and has it been tested?
From the auditor question bank
Your policy says:
“Each quarter the Security Owner reports to Executive Management the number of incidents by severity, mean time to acknowledge, contain and resolve, notifications made, overdue corrective actions and trends.”
“Within 24 hours of confirming that customer or personal data was affected, the Security Owner prepares a notification assessment listing who must be told and by when; Executive Management approves it and the messages, counsel reviews regulatory notices, and copies of every notice are attached to the incident record.”
Walk me through how Northwind Cloud Inc handles evaluating security events, and show me incident severity classification table from the Incident Response Policy.
Derived from the criterion text
Your policy says:
“Within the acknowledgment target, the Security Owner or on-call responder confirms whether the event is an incident, assigns a severity (choosing the higher when facts are uncertain), opens a record with an identifier in the form INC-YYYY-NNN and, for SEV1 and SEV2, appoints the Incident Commander.”
“The on-call responder acknowledges each within the target for the suspected severity.”
Evidence to have open
- Incident severity classification table from the Incident Response Policy
- Incident log for the period showing triage decisions, including events that were closed as non-incidents
- Ticket or chat thread for a sample security alert showing the evaluation and outcome
Policies: P13 Incident Response Policy (section 5), P12 Logging and Monitoring Policy (section 5)
CC7.4 Responding to incidents
The company has a defined incident response process with roles, containment steps, communication and evidence handling, and follows it when an incident occurs.
In the room: Security Owner (Dana Whitfield, CTO): P13
How would customers be notified of a security incident affecting their data?
From the auditor question bank
Your policy says:
“Where personal data is involved, supervisory authorities are notified within 72 hours where the GDPR or UK GDPR applies, affected individuals and state attorneys general where US state breach-notification laws require it, and customers acting as data controllers in time to meet their own deadlines.”
“Incidents at vendors, including Supabase, Stripe, Datadog and Slack, that affect Northwind Cloud Inc data or Northwind are handled under this policy, and vendor contracts must require notification to Northwind Cloud Inc within 72 hours under the Vendor and Third-Party Risk Management Policy.”
Walk me through how Northwind Cloud Inc handles responding to incidents, and show me approved Incident Response Policy with contact details and severity levels.
Derived from the criterion text
Your policy says:
“Every report is triaged and assigned a Section 2 severity within the acknowledgment target for that level; severity may be raised by any responder and lowered only by the Incident Commander with a written rationale.”
“Each quarter the Security Owner reports to Executive Management the number of incidents by severity, mean time to acknowledge, contain and resolve, notifications made, overdue corrective actions and trends.”
Evidence to have open
- Approved Incident Response Policy with contact details and severity levels
- Incident record for at least one real or tabletop incident showing detection, containment, communication and closure
- Annual tabletop exercise notes with attendees, scenario and lessons learned
- Customer notification template and record of any notifications sent
Policies: P13 Incident Response Policy (section 4, 5)
CC7.5 Recovering from incidents
After an incident the company restores normal service, documents what happened, fixes the root cause and updates controls so it is less likely to recur.
In the room: Security Owner (Dana Whitfield, CTO): P13; Engineering Lead: P14, P15
Do you have an incident response plan, and has it been tested?
From the auditor question bank
Your policy says:
“Each quarter the Security Owner reports to Executive Management the number of incidents by severity, mean time to acknowledge, contain and resolve, notifications made, overdue corrective actions and trends.”
“Within 24 hours of confirming that customer or personal data was affected, the Security Owner prepares a notification assessment listing who must be told and by when; Executive Management approves it and the messages, counsel reviews regulatory notices, and copies of every notice are attached to the incident record.”
What are your backup and disaster recovery arrangements, and do you test them?
From the auditor question bank
Your policy says:
“As part of the disaster recovery test under the Business Continuity and Disaster Recovery Policy, Engineering restores the full Tier 1 data set from backups into a rebuilt environment, runs the application test suite against it, and records the achieved recovery time and data age.”
“Engineering restores a Tier 1 environment from infrastructure code and backups into an isolated account or region, measures elapsed time and the age of restored data, and records whether RTO and RPO were achieved, with the runbook used, timestamps, verification evidence and defects found.”
Evidence to have open
- Post-mortem document for each significant incident with root cause and follow-up actions
- Remediation tracker entries created from post-mortem action items and their closure dates
- Backup restore test record used as part of incident recovery or recovery testing
Policies: P13 Incident Response Policy (section 5), P14 Business Continuity and Disaster Recovery Policy (section 5), P15 Backup and Recovery Policy (section 5)
CC8 Change management
CC8.1 Change management
Changes to software and infrastructure are authorized, developed and tested in a controlled way, reviewed by someone other than the author, approved, and deployed with a way to roll back.
In the room: Engineering Lead: P09, P10
How are changes to production reviewed, tested and approved?
From the auditor question bank
Your policy says:
“All code is peer reviewed through a pull request in GitHub under the Change Management Policy, and reviewers apply a security checklist covering authorization, input handling, secrets, logging of sensitive data, error handling and dependency changes.”
“Branch protection on default and release branches requires at least one approving review from someone other than the author, requires all status checks from GitHub Actions to pass, dismisses stale approvals on new commits, blocks force pushes and branch deletion, and applies to administrators without bypass.”
Walk me through how Northwind Cloud Inc handles change management, and show me branch protection settings export requiring pull request review and passing checks before merge.
Derived from the criterion text
Your policy says:
“After approval and passing checks the change is merged to the protected branch; GitHub Actions deploys to staging, runs smoke tests and then deploys to production, automatically or on approval of a named engineer per repository configuration, tagging each deployment with the commit identifier.”
“The Security Owner samples at least ten percent of production changes and no fewer than five, verifies independent approval, passing checks and a deployment record for each, confirms branch protection settings in GitHub remain as required, investigates any bypass or emergency change, and summarizes the results to Executive Management.”
Evidence to have open
- Branch protection settings export requiring pull request review and passing checks before merge
- Sample of merged pull requests from the period showing reviewer approval and CI status
- CI/CD pipeline configuration file showing automated tests and the production deploy step
- Emergency change record showing after-the-fact review, if any occurred
Policies: P09 Change Management Policy (section 4, 5), P10 Secure Software Development Policy (section 4)
CC9 Risk mitigation
CC9.1 Mitigating business disruption risk
The company identifies what could disrupt the business, decides how to reduce or accept each risk, and maintains plans (including backups and recovery) to keep operating or recover.
In the room: Engineering Lead: P14, P15; Security Owner (Dana Whitfield, CTO): P17
What are your backup and disaster recovery arrangements, and do you test them?
From the auditor question bank
Your policy says:
“Tier 1 recovery is tested at least annually by restoring a production-equivalent environment from infrastructure code and backups, and the whole plan is exercised at least annually through a tabletop with Executive Management; achieved recovery times are recorded against the objectives.”
“Restores are tested at least quarterly for a sample of Tier 1 databases and storage, and at least annually for a full Tier 1 environment as part of the disaster recovery test.”
Walk me through how Northwind Cloud Inc handles mitigating business disruption risk, and show me business continuity and disaster recovery plan with recovery time and recovery point objectives per system.
Derived from the criterion text
Your policy says:
“Emergency access granted to recover a system is time-limited, logged and revoked when recovery is complete.”
“After every declared disaster and every test, Engineering completes a review within ten business days recording the timeline, achieved RTO and RPO, deviations from the plan and corrective actions with owners and due dates; unresolved gaps affecting Tier 1 objectives are entered in the risk register.”
Evidence to have open
- Business continuity and disaster recovery plan with recovery time and recovery point objectives per system
- Risk register entries for availability and disruption risks with treatments
- Cyber insurance policy summary or documented decision not to carry it
Policies: P14 Business Continuity and Disaster Recovery Policy (section 4), P17 Risk Assessment and Management Policy (section 4), P15 Backup and Recovery Policy (section 4)
CC9.2 Vendor and business partner risk
The company assesses the security of vendors that touch its systems or data, puts security and confidentiality commitments in contracts, monitors them over time, and handles offboarding.
In the room: Security Owner (Dana Whitfield, CTO): P16, P22
How do you manage vendor and third-party risk?
From the auditor question bank
Your policy says:
“The requester states the business purpose, data to be shared, systems to be accessed, expected users and cost, and any existing alternative in the inventory; the Security Owner assigns a provisional tier within five business days and lists the due diligence required.”
“Each quarter the Security Owner reconciles the inventory against vendor payments, the single sign-on application list and OAuth grants found by Engineering, adds vendors found outside the process and has the responsible person complete intake or stop use.”
Walk me through how Northwind Cloud Inc handles vendor and business partner risk, and show me vendor inventory with data access level, criticality tier and review date for each vendor.
Derived from the criterion text
Your policy says:
“Northwind Cloud Inc maintains a vendor inventory recording, for every vendor, the business owner, service provided, data shared and its classification, systems accessed, tier, contract and renewal date, last review date and outcome, and assurance evidence held, and reviews it for completeness at least annually.”
“Contracts with Tier 1 and Tier 2 vendors include confidentiality obligations, security controls appropriate to the data, notification of incidents affecting Northwind Cloud Inc data within 72 hours of the vendor becoming aware, restrictions on sub-processors, return or deletion of data on termination, and a right to assurance evidence.”
Evidence to have open
- Vendor inventory with data access level, criticality tier and review date for each vendor
- SOC 2 report review notes or completed security questionnaire for each critical vendor
- Signed vendor DPA or contract with security and confidentiality clauses
- Annual vendor review record signed by the Security Owner
Policies: P16 Vendor and Third-Party Risk Management Policy (section 4, 5), P22 Privacy and Data Protection Policy (section 4)
A1 Availability
A1.1 Capacity management
The company tracks how much of its infrastructure capacity is being used, forecasts demand, and adds capacity before it runs out.
In the room: Engineering Lead: P14, P12, P20
Walk me through how Northwind Cloud Inc handles capacity management, and show me capacity or utilization dashboard screenshot from the monitoring tool (CPU, memory, storage, database connections).
Derived from the criterion text
Your policy says:
“Each month Engineering reviews availability against the service level objectives, documents any breach and its cause, updates the capacity forecast and reports the results to Executive Management.”
“Vercel: workloads sit in private subnets or equivalent private segments with no direct inbound route from the internet; databases and caches accept connections only from application segments; security groups and firewall rules default to deny and allow only documented ports and sources; and a web application firewall or equivalent edge protection fronts every public HTTP endpoint.”
Show me alert rules for capacity thresholds with the person or channel they notify, and tell me who keeps it current.
Derived from the criterion text
Your policy says:
“Alert rules are tuned to keep false positives manageable; changes to a rule's logic, threshold or severity are recorded with the reason, and rules that produced only false positives for a quarter are reviewed for relevance.”
“When the criteria in 4.5 are met, the declaring person records the declaration, names the recovery lead, opens the alternative channel and notifies Executive Management.”
Evidence to have open
- Capacity or utilization dashboard screenshot from the monitoring tool (CPU, memory, storage, database connections)
- Alert rules for capacity thresholds with the person or channel they notify
- Autoscaling configuration export or documented capacity review notes
Policies: P14 Business Continuity and Disaster Recovery Policy (section 4), P12 Logging and Monitoring Policy (section 4), P20 Network and Infrastructure Security Policy (section 4)
A1.2 Environmental protections, backups and recovery infrastructure
Systems are protected from environmental threats, data is backed up, and the infrastructure needed to recover exists and is designed to meet recovery objectives.
In the room: Engineering Lead: P15, P14; IT/Operations Lead: P21
What are your backup and disaster recovery arrangements, and do you test them?
From the auditor question bank
Your policy says:
“Tier 1 recovery is tested at least annually by restoring a production-equivalent environment from infrastructure code and backups, and the whole plan is exercised at least annually through a tabletop with Executive Management; achieved recovery times are recorded against the objectives.”
“Restores are tested at least quarterly for a sample of Tier 1 databases and storage, and at least annually for a full Tier 1 environment as part of the disaster recovery test.”
How do you address physical security, for offices and for remote staff?
From the auditor question bank
Your policy says:
“All production systems for Northwind run in facilities operated by AWS and Vercel, and Northwind Cloud Inc relies on those providers' physical and environmental controls (access control, surveillance, power, cooling and fire suppression) as described in their independent assurance reports.”
“Where Northwind Cloud Inc maintains an office, entry is restricted to authorized personnel through keys, badges or a managed access system, with access issued at onboarding and revoked on the last working day.”
Evidence to have open
- Backup configuration screenshot showing schedule, retention and encryption for production data
- Cloud provider SOC 2 report covering environmental controls for the regions in use
- Architecture diagram showing multi-availability-zone or multi-region deployment for critical services
- Backup job success report for the last 90 days
Policies: P15 Backup and Recovery Policy (section 4), P14 Business Continuity and Disaster Recovery Policy (section 4), P21 Physical and Remote Work Security Policy (section 4)
A1.3 Recovery testing
The company actually tests its recovery plan and backup restores on a schedule, records the results, and fixes anything that did not work.
In the room: Engineering Lead: P14, P15
What are your backup and disaster recovery arrangements, and do you test them?
From the auditor question bank
Your policy says:
“As part of the disaster recovery test under the Business Continuity and Disaster Recovery Policy, Engineering restores the full Tier 1 data set from backups into a rebuilt environment, runs the application test suite against it, and records the achieved recovery time and data age.”
“Engineering restores a Tier 1 environment from infrastructure code and backups into an isolated account or region, measures elapsed time and the age of restored data, and records whether RTO and RPO were achieved, with the runbook used, timestamps, verification evidence and defects found.”
Walk me through how Northwind Cloud Inc handles recovery testing, and show me backup restore test record with date, dataset restored, time taken and who performed it.
Derived from the criterion text
Your policy says:
“Each quarter Engineering restores a backup of at least one Tier 1 database and one storage bucket, no more than one day old, into an isolated environment, verifies row counts, checksums or sample records against production, measures elapsed time against the RTO, deletes the copy and files the test record for the Security Owner's review.”
“Each test records what was restored, the backup age, the elapsed time, the verification performed and any defects, and is retained as evidence.”
Evidence to have open
- Backup restore test record with date, dataset restored, time taken and who performed it
- Annual disaster recovery exercise report with scenario, results and follow-up actions
- Ticket showing remediation of any issue found during recovery testing
Policies: P14 Business Continuity and Disaster Recovery Policy (section 5), P15 Backup and Recovery Policy (section 5)
Using it in the walkthrough
- Send the sheet to everyone named under “In the room” a few days before the session.
- Read every quote aloud once. If it is not what happens, fix the policy or the practice now; the auditor will compare the two.
- Open each evidence item in a tab or a shared folder, with dates visible, before the call starts.
- Answer from the procedure you follow and point to the section. “Section 5.3 of the Access Control Policy; here is last quarter’s review” is the kind of answer that closes a question.
For the evidence itself, see the SOC 2 evidence checklist; for what the walkthrough tests in each type of report, Type I vs Type II.
Frequently asked questions
- Will my auditor ask exactly these questions?
- Probably not word for word. Every CPA firm runs its walkthroughs its own way. The questions marked as from the question bank are the 25 Policyseed questions about policies, written in our own words, matched to a criterion only where the bank's own answer points at it. The others are built from the criterion's own description and evidence list and are labeled as derived. Use them to rehearse, not as a script.
- Where do the quoted sentences come from?
- From the 22 policies the free generator renders from your answers, rendered again in your browser with the same templates. A keyword ranking (the same one the questionnaire tool uses) picks the sentences that best match each question within the policies mapped to that criterion. It is a search, not a judgment: read each quote and make sure it really answers the question. If you edited or tailored your policies after downloading them, quote your edited version instead, because this page cannot see those edits.
- Is anything uploaded?
- No. The generator keeps your answers in this browser's local storage, the policies are rendered here, and the Word file is built here. With no saved answers the page shows the sample company, Northwind Cloud Inc, and says so at the top of the sheet.
- Who should be in the room?
- The sheet lists the owner role of every policy involved in a criterion, such as the Security Owner, Engineering Lead, IT/Operations Lead or People/HR Lead. In a small company one person often holds several roles. The auditor wants to hear from whoever actually runs the control, so bring that person even if the policy names a different owner, and fix the policy afterward.
- What if a quoted sentence is not true of how we work today?
- Change the policy or change the practice before the walkthrough, and never create evidence after the fact. A policy that promises a quarterly review that has never happened is a likely finding; a policy that states when the first review will run, with the calendar entry to show it, is a plan the auditor can assess.
- Does this replace the evidence checklist?
- No. The sheet lists the evidence items the crosswalk maps to each criterion so you know what to have open during the conversation. Collecting that evidence over the period is a separate job; the SOC 2 evidence checklist guide covers it, and the Audit Kit includes it as a spreadsheet.
Policyseed provides governance policy templates and AI tailoring. It is not legal advice and not a compliance guarantee. Management adopts the policies; the CPA firm performs the SOC 2 examination.