Access Control Policy Template and Guide

Below is a complete, free document your team can adopt this week, with the reasoning behind every section. It maps to CMMC 2.0 and NIST SP 800-171, HIPAA, SOC 2, PCI DSS and ISO/IEC 27001. Written by Petronella Technology Group, Inc. for regulated businesses in Raleigh and beyond.

  • CyberAB RPO #1449
  • BBB A+ Since 2003
  • Serving Raleigh Since 2002

What Is an Access Control Policy?

An access control policy is a written rule set that states who may access which systems and data, and under what conditions. It names who approves each grant of access. It also says how access is reviewed and how it is removed when someone changes jobs or leaves.

Most small companies already have access rules. They just live in one person's head. The office manager knows who has the accounting login. The IT contact remembers which laptop has admin rights. That works until an auditor asks for proof, or until that one person takes a vacation.

Writing the rules down changes the conversation. A new hire gets access because a manager approved a request, not because someone copied the last hire's account. A departure triggers a checklist instead of a guess. If you're working toward CMMC 2.0, HIPAA compliance or a SOC 2 report, the assessor will ask for this document early. We've written this page so you can hand one over.

Key Takeaways

  • An access control policy answers five questions: who gets access, to what, on whose approval, how it is checked, and how it ends.
  • The free template below has 14 numbered sections. Every number in it is a bracketed default for you to confirm.
  • Least privilege is the core idea. People get the access their job needs and nothing extra.
  • Access reviews are where most policies fail. Put them on a calendar and keep the evidence.
  • Shared logins break accountability. One person, one account, every time.
  • A policy only counts when your systems enforce it. That means MFA, role groups, session locks and logs that someone reads.

What an Access Control Policy Must Cover

A one-page statement that "access is granted on a need-to-know basis" won't survive an assessment. Nobody can configure that sentence. Nobody can test it either. A working policy says specific things that a manager can follow and a system can enforce.

The table below lists the sections we expect to see. It's the same outline we use when we write security policies and procedures for clients. Skim the right-hand column if you need to explain to a partner why the document runs longer than a page.

SectionWhat It Must StateWhy It Matters
PurposeWhy the policy exists and which obligations it servesGives every later rule a reason you can point to
ScopeThe people, systems and data covered, including cloud apps and contractorsGaps hide in whatever the scope forgot. SaaS is the usual one.
Roles and ResponsibilitiesA named policy owner, system owners, managers who approve, and IT who provisionsWithout names, every approval becomes "I thought you handled that"
Account TypesRules for standard, privileged, service, shared, guest and vendor accountsEach type fails differently, so each needs its own rule
Access Request and ApprovalHow access is requested, who approves, and where the record is keptThe approval record is the evidence an assessor samples first
Least Privilege and Role-Based AccessAccess tied to job roles, with the minimum rights each role needsLimits how far a stolen password can reach
Separation of DutiesWhich tasks one person may not perform aloneStops a single account from both creating and approving a payment
Authentication and MFAPassword standards and where multifactor authentication is requiredA password alone is the weakest door in the building
Privileged AccessSeparate admin accounts, who may hold them, and how their use is loggedAdmin rights are what an attacker wants most
Remote, Wireless and Third-Party AccessApproved methods for remote work, Wi-Fi rules, and time limits on vendor accountsThese paths start outside your walls, so they deserve tighter terms
Session ControlsScreen lock, idle timeout and failed logon limitsAn unlocked screen at the front desk is access with no login at all
Access Reviews and RecertificationHow often access is reviewed, by whom, and what evidence is keptAccess piles up over time. Reviews are how you clear it.
Joiner, Mover, LeaverProvisioning for new hires, changes on a role change, and removal at terminationOrphaned accounts come from skipped offboarding steps
Logging, Exceptions, Enforcement and Review CycleWhat is logged, how exceptions are approved and expire, consequences, and when the policy is revisitedCloses the loop so the policy stays true next year

Not sure who can reach what on your network today?

Petronella Technology Group can pull the account lists from your directory and from Microsoft 365, as well as the other apps your business runs on. We compare them to your current staff roster and show you the gaps. The first report usually answers questions nobody had thought to ask.

Call 919-348-4912

Free Access Control Policy Template

This access control policy template is written to be adopted. Replace each bracketed placeholder with your own details. Delete any clause that truly does not apply, and keep a note of why.

One rule covers every number below. Review frequencies appear in brackets, as do lockout thresholds and timeouts. Every bracketed deadline is an editable default. They're common starting points for a small or midsize business. Confirm each one against your contracts and the frameworks you answer to. This template is a starting point and not legal advice.

The text is framework-neutral on purpose. A defense contractor and a dental office can start from the same words. The framework mapping further down tells you which clauses to tighten.

1. Purpose

This Access Control Policy sets the rules for granting access to [Company Name] systems and data. It covers how that access gets used. It also sets out the review cycle and the conditions under which access ends. Its goal is that only authorized people and processes reach company information, and only to the extent their work requires. The policy supports [Company Name]'s obligations under [applicable frameworks, for example NIST SP 800-171, HIPAA, SOC 2, PCI DSS or ISO/IEC 27001] and under its customer contracts. Where a law or contract sets a stricter rule than this policy, the stricter rule applies.

2. Scope

This policy applies to every person who uses [Company Name] systems. That includes employees and contractors, as well as interns. Temporary staff and vendors fall under it too. It covers all systems that store or process company data, wherever they run:

  • On-premises servers, workstations and network equipment.
  • Cloud services and software-as-a-service applications, including [list key applications].
  • Company-owned laptops and mobile devices.
  • Personal devices approved for work under the BYOD Policy.
  • Service accounts, application accounts and automated processes.

Physical access to offices and server rooms is governed by the [Physical Security Policy]. Where the two overlap, both apply.

3. Roles and Responsibilities

  • Policy Owner: [title, for example IT Director or Security Officer] maintains this policy, approves exceptions and reports on access reviews to management.
  • System Owners: the named owner of each system decides which roles may access it and confirms access during reviews.
  • Managers: request access for their staff, approve only what the job requires, and report role changes and departures to IT and Human Resources [the same business day].
  • IT Administrators: create, change and remove accounts only on an approved request. Administrators do not approve their own access.
  • Human Resources: notifies IT of hires, transfers and separations before the effective date whenever possible.
  • All Users: protect their credentials, use only the access they were given, and report suspected misuse to [contact].

4. Account Types and Unique Identification

Every user receives a unique account tied to one named person. Users must not share credentials. [Company Name] recognizes these account types:

  • Standard accounts for daily work, with no administrative rights.
  • Privileged accounts for administration, issued separately from the holder's standard account.
  • Service accounts for applications and automated tasks. Each has a named human owner and may not be used for interactive logon unless the Policy Owner approves.
  • Shared accounts are prohibited by default. Where a system cannot support individual accounts, the Policy Owner may approve a shared account as a documented exception with a named custodian and a usage log.
  • Guest accounts are limited to the guest wireless network or to specific shared files. They expire after [30 days].
  • Vendor accounts follow Section 10.

Default accounts supplied with hardware or software are disabled or renamed, and their default passwords are changed before the system goes into use. User identifiers are not reissued to a different person for at least [one year].

5. Access Request, Approval and Provisioning

All access starts with a request in [ticketing system or form]. The request names the user and the system. It states the requested role or permission level, and it states the business reason. The user's manager approves it. For systems holding [sensitive data categories, for example CUI, ePHI or cardholder data], the System Owner also approves.

IT provisions access only after approval is recorded. Verbal or hallway approvals are not valid. New accounts are created from the role definition in the Access Matrix, never by copying another user's account. Initial passwords are temporary and must be changed at first logon. Request and approval records are kept for [three years].

6. Least Privilege, Role-Based Access and Separation of Duties

Access is granted by role. [Company Name] maintains an Access Matrix that lists each role and the systems and permission levels it receives. Users get the minimum access their role requires. Access beyond the role requires a separate request with a stated reason and an end date.

No single person may complete a high-risk process alone. At a minimum, the following duties are separated:

  • Requesting access and approving that same access.
  • Creating a vendor or payee and approving payment to it.
  • Administering a system and reviewing that system's audit logs, where staffing allows.
  • Developing a change and approving its release to production.

Where the company is too small to separate a duty, the Policy Owner documents a compensating control, such as a monthly review by an owner or outside party.

7. Authentication and Multifactor Authentication

Passwords must be at least [14] characters. They are checked against a list of known compromised passwords where the system supports it. Users may not reuse their last [10] passwords. Passwords are changed immediately on any sign of compromise. Passwords are never stored or sent in plain text.

Multifactor authentication (MFA) is required for:

  • All remote access to the network and all cloud services, including email.
  • All privileged accounts, both local and remote.
  • Any system holding [sensitive data categories].

Approved factors are [authenticator app with number matching, hardware security key]. Text message codes are permitted only where no stronger option exists and the Policy Owner has approved it. Password managers approved by IT are encouraged. Writing passwords on paper kept at the desk is not permitted.

8. Privileged Access

Administrative rights are limited to staff whose role requires them and are approved by the Policy Owner. Each administrator holds a separate privileged account. That account is used only for administrative tasks. It is not used for email or web browsing.

Users do not hold local administrator rights on their workstations unless an exception is approved. Privileged actions are logged. Where the platform supports it, elevated rights are granted for a limited time and expire on their own. The list of privileged accounts is reviewed [monthly]. Emergency "break glass" accounts are stored in [secure location] and tested [twice a year]. Every use triggers an alert to the Policy Owner.

9. Remote, Wireless and Mobile Access

Remote access is permitted only through an approved method, for example [the company VPN or a zero trust access service]. All remote sessions are encrypted and require MFA. Remote access from a personal device requires enrollment under the BYOD Policy.

Company wireless networks use [WPA3 or WPA2-Enterprise] with individual authentication. Guest wireless is separated from the internal network. Mobile devices that access company data must be enrolled in device management. They must also be encrypted and protected by a passcode. Users may not install personal wireless access points or remote access tools on company systems.

10. Third-Party and Vendor Access

Vendors receive access only after a signed agreement is in place and a [Company Name] sponsor is named. Each vendor technician uses an individual account where the vendor's tools allow it. Vendor accounts are:

  • Limited to the specific systems the vendor supports.
  • Protected by MFA.
  • Set to expire at contract end or after [90 days], whichever comes first, and renewed only by the sponsor.
  • Disabled between engagements where practical and enabled on request.

Vendor remote sessions to systems holding [sensitive data categories] are logged. The sponsor tells IT [the same business day] when a vendor relationship ends.

11. Session Controls

Workstations and mobile devices lock automatically after [15 minutes] of inactivity and hide the screen contents. Users lock their screens whenever they step away. Accounts lock after [5] consecutive failed logon attempts and stay locked for [30 minutes] or until IT verifies the user and unlocks the account. Remote sessions and sessions to sensitive applications end after [30 minutes] of inactivity. A logon banner approved by the Policy Owner is displayed on [systems in scope] before access is granted.

12. Access Reviews and Recertification

System Owners and managers review user access [quarterly] for systems holding [sensitive data categories] and [annually] for all other systems. Privileged and vendor accounts are reviewed [monthly]. For each review:

  • IT exports the current account and permission list from the system itself.
  • The reviewer marks each account as keep, change or remove.
  • IT completes the changes within [5 business days].
  • The export, the reviewer's decisions and the completed tickets are saved as evidence for [three years].

Accounts with no logon for [45 days] are disabled. Disabled accounts are deleted after [90 days] unless a legal hold or retention requirement applies.

13. Role Changes and Termination

When a user changes roles, the manager submits a change request. Access from the old role is removed within [5 business days]. It is not carried into the new role by default.

When employment or a contract ends, Human Resources notifies IT in advance where possible. IT disables all accounts by [the end of the user's last working day]. For an involuntary termination, accounts are disabled [at the time of notification]. IT also revokes active sessions and MFA registrations and removes the user from shared mailboxes and groups. Passwords for any shared or service account the user knew are changed. Company devices are collected. Any keys and badges are too. The completed offboarding checklist is retained as evidence.

14. Logging, Exceptions, Enforcement and Review

Logging. Systems record successful and failed logons. They also record account creation and removal, together with permission changes. Every privileged action is logged as well. Logs are protected from alteration and kept for [one year]. Alerts for [defined events, for example repeated lockouts or a new administrator] go to [IT or the security operations provider].

Exceptions. Any exception to this policy is requested in writing with a reason and a compensating control. The Policy Owner approves it. Each exception has an expiration date no later than [12 months] out and is reviewed before renewal.

Enforcement. Violations may lead to removal of access and to disciplinary action up to termination of employment or contract. Suspected violations are reported to [contact].

Review. The Policy Owner reviews this policy at least [annually] and after any significant change in systems. Changes in regulations or contracts trigger a review too. Version history is recorded below.

Acknowledgment. I have read and understood the [Company Name] Access Control Policy and agree to follow it. Name: [ ] Signature: [ ] Date: [ ]

Access Control Models Explained

Your policy has to pick a way of deciding who gets what. Security textbooks describe five models. You don't need to memorize them. You do need to know which one your policy is built on, because assessors ask.

ModelHow Access Is DecidedWhere You See ItTrade-Off
Discretionary (DAC)The owner of a file or folder chooses who can open itSharing a document from your own cloud driveEasy to use, hard to audit. Permissions drift fast.
Mandatory (MAC)A central authority labels data and clears users. Owners can't override it.Classified government systemsVery strict. Too heavy for most businesses.
Role-Based (RBAC)Permissions attach to roles. People are placed in roles.Security groups such as "Accounting" or "Front Desk"Clear and reviewable. Needs upkeep as jobs change.
Attribute-Based (ABAC)A decision engine weighs attributes of the user, the device, the data and the momentConditional access that checks device health and locationFlexible. Harder to design and to explain.
Rule-BasedFixed rules set by an administrator apply to everyone alikeFirewall rules, or logon hours limited to the workdaySimple and predictable. Blind to who the person is.

For a small or midsize organization, our recommendation is plain. Build on RBAC. Define a short list of roles that match real jobs, and put people in them. Then add a few attribute conditions through conditional access. Require a managed device for the finance system. Block sign-ins from countries where you have no staff. That mix is what a role-based access control policy looks like in practice today.

Keep the role list short. Twelve roles that people understand beat sixty that nobody maintains. If two roles differ by one permission, merge them. Handle the difference as a documented exception.

How to Roll Out the Policy in Six Steps

Adopting the template takes an afternoon. Making it true takes longer. The order below is the one we follow, and the first step is the one people most want to skip.

  1. Inventory systems, data and accounts

    List every system that holds company data. Include the SaaS apps a department bought on a credit card. Then export the accounts from each one. Good IT asset management makes this step quick. A data classification policy tells you which systems deserve the strictest rules.

  2. Define roles and an access matrix

    Write down the roles in your company. For each role, write down the systems that role touches and the permission level each one needs. A spreadsheet is fine. Ask each manager to confirm their column. This matrix becomes the answer key for every future access request.

  3. Set authentication and MFA standards

    Decide your password length and your approved MFA methods. Our page on NIST digital identity guidance explains the thinking behind longer passwords and fewer forced changes. Turn MFA on for email and remote access first. Those two are where stolen passwords get used.

  4. Build the request, approval and removal workflow

    Pick one place where access requests live. A ticket queue works. Tie it to your employee onboarding checklist so day-one access follows the matrix. Tie removal to your employee offboarding checklist so nothing depends on memory.

  5. Schedule access reviews with evidence

    Put the reviews on the calendar now, with a named reviewer for each system. Save the account export and the reviewer's sign-off every time. Store them with your other IT documentation. An undocumented review didn't happen, as far as an assessor is concerned.

  6. Monitor, log and handle exceptions

    Send logon and permission-change events somewhere a person or a service watches. Keep an exception register with an owner and an end date on every line. Review it when you review the policy.

Framework Mapping

Nearly every security framework has an access control chapter. They use different numbering for the same handful of ideas. The table below shows where each part of the template lines up. Treat it as a planning aid, not an official crosswalk. Your assessor works from the source documents, and so should your final mapping.

A note for defense contractors. CMMC Level 2 uses the same requirements as NIST SP 800-171 Rev 2. The Access Control family runs from 3.1.1 to 3.1.22. Identification and Authentication runs from 3.5.1 to 3.5.11. We've written overviews of the CMMC access control domain and the identification and authentication domain, plus a page for each requirement. If your contracts involve controlled unclassified information, start there. It's also worth reading what changes in Revision 3 before you finalize wording.

Template SectionNIST SP 800-171 Rev 2 / CMMC Level 2NIST SP 800-53 Rev 5HIPAA Security RuleSOC 2, PCI DSS v4.0, ISO/IEC 27001:2022
1 to 3. Purpose, Scope, RolesFrames the whole 3.1 familyAC-1 policy and procedures45 CFR 164.308(a)(4) information access managementSOC 2 CC6.1; ISO Annex A 5.15 access control
4. Account Types and Unique Identification3.5.1, 3.5.5AC-2 account management164.312(a)(1) access controlPCI DSS Requirement 8; ISO Annex A 5.16 identity management
5. Request, Approval and Provisioning3.1.1, 3.5.9AC-2 account management164.308(a)(4) information access managementSOC 2 CC6.2; ISO Annex A 5.18 access rights
6. Least Privilege, Roles, Separation of Duties3.1.2, 3.1.4, 3.1.5AC-3 access enforcement, AC-5 separation of duties, AC-6 least privilege164.308(a)(3) workforce securitySOC 2 CC6.3; PCI DSS Requirement 7; ISO Annex A 8.3 information access restriction
7. Authentication and MFA3.5.2, 3.5.3, 3.5.7, 3.5.8, 3.5.10AC-3 access enforcement164.312(d) person or entity authenticationPCI DSS Requirement 8; ISO Annex A 5.17 authentication information and 8.5 secure authentication
8. Privileged Access3.1.6, 3.1.7, 3.1.15AC-6 least privilege164.308(a)(4) information access managementPCI DSS Requirement 7; ISO Annex A 8.2 privileged access rights
9. Remote, Wireless and Mobile Access3.1.12 to 3.1.14, 3.1.16 to 3.1.19AC-17 remote access164.312(a)(1) access controlSOC 2 CC6.1; PCI DSS Requirement 8
10. Third-Party and Vendor Access3.1.20, 3.1.21AC-2 account management, AC-17 remote access164.308(a)(4) information access managementSOC 2 CC6.2; PCI DSS Requirement 8
11. Session Controls3.1.8, 3.1.9, 3.1.10, 3.1.11AC-7 unsuccessful logon attempts, AC-11 device lock164.312(a)(1) access controlPCI DSS Requirement 8; ISO Annex A 8.5 secure authentication
12. Access Reviews and Recertification3.1.1, 3.5.6AC-2 account management164.308(a)(4) information access managementSOC 2 CC6.2 and CC6.3; ISO Annex A 5.18 access rights
13. Role Changes and Termination3.1.1, 3.5.6AC-2 account management164.308(a)(3) workforce securitySOC 2 CC6.3; ISO Annex A 5.18 access rights
14. Logging, Exceptions, Enforcement, Review3.1.7, 3.1.12AC-1 policy and procedures164.308(a)(3) workforce securitySOC 2 CC6.1; ISO Annex A 5.15 access control

Healthcare readers should pair this table with our pages on HIPAA access control and Security Rule compliance. Unique user identification is the piece small practices most often miss. For audit-driven companies, our guide to SOC 2 controls shows how an auditor tests the CC6 criteria. Merchants can find the access requirements in our PCI DSS overview. Firms pursuing certification will want the ISO 27001 page.

As Craig Petronella details in CMMC 2.0 Certification Guide, assessors don't grade the policy alone. They look for the policy, then the procedure, then proof that the procedure ran. A mapped policy with no review evidence scores poorly. So write less and do all of it.

Access Control Policy vs Password Policy vs Acceptable Use Policy vs Identity and Access Management

These four terms get mixed up in meetings. Three are documents. One is a discipline with tools behind it. Here's how they differ.

Access Control PolicyPassword PolicyAcceptable Use PolicyIdentity and Access Management (IAM)
What it isRules for granting, reviewing and removing accessRules for creating and protecting passwordsRules for how people behave on company systemsThe processes and tools that carry out access decisions
Main questionWho may reach what, and who said so?How strong must a credential be?What may you do once you're in?How do we make the rules happen at scale?
Primary audienceManagers, system owners, ITEvery userEvery userIT and security staff
Typical ownerSecurity officer or IT directorITHuman Resources with ITIT
RelationshipThe parent document for access decisionsOften a section inside the access control policySits beside it. Users sign both.Implements all three

A small company can fold its password rules into Section 7 of the template and skip a separate document. The acceptable use policy should stay separate, because every employee signs it and it needs to be short. Personal phones and laptops belong in a BYOD policy. If your staff use AI tools with company data, add an AI acceptable use policy as well.

Two more documents feed this one. Your data classification policy defines which information is sensitive, and the access rules get stricter as the label goes up. Your data retention policy decides how long access records and logs are kept. Some organizations call the whole set a user access management policy. The name matters less than having each rule written in exactly one place.

Common Access Control Mistakes

We see the same problems in offices of every size. Each one maps to a section of the template. If one sounds familiar, fix that section first.

Shared accounts

The front desk has one login and four people who use it. The password is on a sticky note. When something goes wrong, the log says "frontdesk" did it. That tells you nothing. Give each person an account, even if the workstation is shared. Fast user switching or badge tap sign-in keeps it painless.

Standing admin rights

Someone needed admin rights for an install three years ago. They still have them. Now every email attachment that person opens runs with the keys to the machine. Admin rights should be separate from daily accounts where your tools allow. They should also be limited and temporary. That is the heart of a privileged access management policy.

No access reviews

Access only ever grows unless somebody prunes it. People change departments and keep the old folders. A user access review is the pruning. It's dull work, and assessors ask for the evidence. Put it on the calendar and assign a name.

Orphaned accounts after offboarding

The admin who left two years ago still has a mailbox. Maybe it forwards to a personal address. Maybe the password never changed. Orphaned accounts are a favorite way in, because no one is watching them. Tie account removal to the HR separation process so it can't be forgotten.

MFA gaps on remote access and email

Many firms turned on MFA for most users and left a few holes. The owner was exempted because the prompts were annoying. An old mail protocol was left open for the copier. Attackers look for exactly those holes. Insurers ask about them too, as our cyber insurance readiness page explains.

Vendor accounts that never expire

Your software vendor got a remote login during installation. The project ended. The login didn't. Vendor access should carry an end date from the first day. For healthcare vendors, the access terms belong alongside the business associate agreement.

Service accounts with unmanaged credentials

Service accounts run backups and scanners. They also run line-of-business apps. Their passwords are often old and widely known inside IT. Some have domain admin rights they never needed. Give each one an owner. Trim its rights. Store its credential in a vault and rotate it when anyone who knew it leaves.

A policy that ignores SaaS

The policy talks about "the network." Meanwhile the real data sits in a dozen cloud apps, each with its own user list. Marketing's design tool still has last year's intern. Bring every app that holds company data into scope. Connect them to single sign-on where you can, so one disable action closes every door.

Exceptions never revisited

An exception made sense when it was granted. Nobody wrote down an end date. Years later it has become the way things are done. Every exception needs an owner and an expiration, and the list needs a regular read. Staff who understand why the rules exist ask for fewer exceptions, which is one more argument for security awareness training.

Enforcing the Policy: Technical Controls That Back the Words

A policy is a promise on paper. Controls are what keep it. Here is how the main sections of the template turn into settings.

Conditional access and MFA. In a Microsoft 365 shop, identity lives in Microsoft Entra ID. Conditional access policies there require MFA and can block sign-ins from unmanaged devices or unexpected locations. Legacy authentication gets switched off. This is the core of our Microsoft 365 security work.

Role groups. The access matrix becomes a set of security groups. Permissions are assigned to groups and never to individuals. Adding a new hire to "Accounting" gives them what accounting needs. Removing them takes it all away.

Privileged access separation. Administrators get a second account for admin work. Where licensing allows, elevated roles are activated on request and expire after a set time. Local admin rights come off user workstations.

Session lock and lockout settings. Device management pushes the screen lock timer and the failed logon threshold to every machine. The numbers in your policy and the numbers in the configuration should match exactly. Assessors compare them.

Zero trust network access. Older remote access put a user on the whole network once the VPN connected. Zero trust access checks the user and the device on every request and opens one application at a time. Our zero trust security page covers the approach, and zero trust vs VPN compares the two. The underlying model comes from the NIST zero trust architecture.

Logging and alerting. Sign-in logs and permission changes flow to our security operations center through managed XDR. Analysts watch for the patterns that matter: a new global administrator, a burst of lockouts, a logon from two distant places within the hour. The Managed XDR Suite page describes what is monitored.

Policy acknowledgment and evidence. Compliance Armor® is Petronella Technology Group's compliance documentation platform. It includes an SSP generator, a gap analysis module and evidence collection. Continuous monitoring and policy acknowledgment are part of the platform too. For access control, that means you can show who signed the policy and where each review's evidence sits.

Have a policy that your systems don't actually enforce?

That gap is common, and it's fixable. We compare your written rules to your real conditional access settings, to your group structure and to your lockout configuration, then close the differences one by one. Clients in Durham, Cary, Chapel Hill and across the Triangle work with the same team.

Call 919-348-4912

Digital Forensics and Access Logs: Why They Matter After an Incident

Access control is usually sold as prevention. It has a second job that owners rarely think about until the bad day arrives. It makes investigation possible.

After an incident, the first questions are simple. Which account was used? What did it touch? When? Unique user IDs and intact logs let an examiner rebuild that sequence. The work is slow, but it can be done. With good records, you can tell a regulator or an insurer what happened and what did not.

Shared accounts destroy that. If five people know the password, the log proves only that one of five people, or an outsider, did something. Attribution is gone. The same holds when logs were never kept or when an administrator could quietly edit them.

Craig Petronella is a North Carolina licensed Digital Forensic Examiner (License 604180-DFE). That background shapes how we write access policies. We ask what an examiner would need a year from now and work backward. You can read about our digital forensics services and our incident response guide for the wider picture.

Not every incident starts outside. A departing employee with broad access can copy a client list on the way out. Tight access and good logs are the base of any insider threat program. As Craig Petronella details in How Hackers Can Crush Your Business, small companies tend to find out how much access people had only after it has been misused.

Three Ways to Get an Access Control Policy: An Honest Comparison

You have options, and the free one may be enough. Here's a fair look at each.

Free Template (This Page)Generic Template VendorPetronella Technology Group Engagement
What you getA 14-section policy you edit yourselfA policy pack, often covering many topics at onceA policy written around your systems, plus the configuration to enforce it
Effort on your sideYou fill in every bracket and build the access matrixSimilar. You still adapt each document.You answer questions and approve decisions. We do the drafting.
Fit to your environmentAs good as the time you put inVaries. Many packs are written for larger companies.Based on an inventory of your actual accounts and apps
Framework mappingPlanning table includedUsually included, at varying depthMapped to the frameworks in your contracts, with evidence tracked
Technical enforcementNot included. Your IT team configures it.Not includedIncluded: conditional access, groups, lockout and logging
Best forA company with capable IT staff and timeA team that needs many policies quickly and can edit themA regulated business facing an assessment, or one without spare IT hours
LimitsNo one checks your workNo one checks your work, and the fit may be looseCosts more than a template. We don't perform CMMC certification assessments.

If you have a sharp IT person and a free week, the template will serve you well. Some companies want ongoing ownership of the policy without hiring a security executive. A vCISO arrangement suits them. Our virtual CISO page explains how that works.

Our Approach

Petronella Technology Group has worked from Raleigh since 2002. We are a CyberAB Registered Provider Organization (RPO #1449) and have held a BBB A+ rating since 2003. When a client asks us for an access control policy, the work follows the same path each time.

  1. Discover

    We export the accounts and groups (admin roles included) from the directory and the applications your company runs. Then we compare them to your staff list. This often sits inside a broader risk assessment.

  2. Decide

    We sit down with you and your managers to define roles. You make the calls about who needs what. We bring the questions and the framework requirements from your compliance risk assessment.

  3. Draft

    We write the policy and the access matrix in plain language. Every bracket gets a real value that you approved.

  4. Configure

    Our engineers set up the groups and the conditional access rules, and they tune every session setting to match the document. Stale accounts get cleaned out with your sign-off.

  5. Prove

    Staff acknowledge the policy in Compliance Armor®. The first access review runs, and its evidence is filed where an assessor can find it.

  6. Maintain

    Reviews recur on schedule. For clients on our managed IT services in Raleigh, the same workflow handles onboarding and offboarding, so the policy stays true between audits.

"Craig takes the time to understand our business model, not just our technology stack. It makes his recommendations more strategic and tailored to our actual goals."

Daniel Lee, TrustIndex verified review

Petronella Technology Group is rated 4.7 across 92 verified TrustIndex reviews and 5.0 across 15 Google reviews.

Frequently Asked Questions

What is an access control policy?

An access control policy is a written rule set that states who may access which systems and data, under what conditions, and on whose approval. It also explains how access is reviewed and how it is removed when a person changes roles or leaves. The policy turns informal habits into rules a manager can follow and a system can enforce. Assessors for most security frameworks ask to see it early, along with proof that its procedures are followed.

What should an access control policy include?

At a minimum it should state its purpose and scope. It should name the people responsible and define each account type. It needs a request and approval process, plus rules covering least privilege and the separation of duties. Authentication standards, including multifactor authentication, belong in it too. Add sections covering privileged access and remote access, plus a third one for vendor access. Then cover session lock and lockout settings. Add scheduled access reviews, and removal of access on a role change or at termination. Finish with logging, exceptions, enforcement and a review cycle.

What is the principle of least privilege?

Least privilege means each person and each process gets only the access needed to do the job, and no more. A bookkeeper can reach the accounting system but not the engineering drawings. An IT administrator uses a standard account for email and a separate one for admin tasks. The idea limits damage. If a password is stolen, the thief gets a small slice of the company and not the whole thing. It also makes mistakes less costly.

How often should user access be reviewed?

A common starting point is quarterly for systems holding sensitive data and annually for everything else. Privileged and vendor accounts deserve a closer look, often monthly. Your contracts and frameworks may set a different schedule, so confirm before you commit to one. Whatever you choose, write it into the policy. Keep evidence of each review. Save the account export and the reviewer's decisions, along with the tickets that show changes were made. A review with no record does not count in an assessment.

What is the difference between RBAC and ABAC?

Role-based access control (RBAC) grants permissions to roles, and people are placed in those roles. It is simple to explain and easy to review. Attribute-based access control (ABAC) makes each decision by weighing attributes such as the user's department, the health of their device, the sensitivity of the data in question and even the time of day. ABAC is more flexible but harder to design. Most small and midsize organizations do well with RBAC plus a few attribute conditions applied through conditional access.

Can a small business use this access control policy template as written?

Yes, after you fill in the brackets. The template is written so a small business can adopt it without a rewrite. Replace the placeholders with your company name and your systems, and name the people responsible. Check every bracketed number against your contracts and the frameworks you follow, because those values are defaults. Remove any clause that truly does not apply and note why. Then have staff sign the acknowledgment and schedule your first access review, so the policy starts with evidence behind it.

Are shared accounts ever acceptable?

Rarely, and only as a documented exception. Some older equipment and a few applications cannot support individual accounts. In those cases, name one custodian for the shared account and keep a log of who uses it and when. Limit what the account can reach, and change the password whenever someone who knew it leaves. Set an expiration date on the exception and look for a replacement system. For everything else, each person should have a unique account so actions can be traced to an individual.

Does Petronella Technology Group help write and enforce access control policies?

Yes. Petronella Technology Group writes access control policies around your real systems and then configures the controls that enforce them. That work includes an account inventory and role definitions, plus conditional access and MFA settings. It also covers session and lockout configuration, and a recurring access review with evidence. We have served Raleigh since 2002 and are a CyberAB Registered Provider Organization (RPO #1449) with a BBB A+ rating since 2003. Call 919-348-4912 or use the contact form on our website to start with a policy review.

Make Your Access Control Policy Real

A policy in a binder protects nobody. We can inventory your accounts and write the rules with you. Then we configure your systems to match and set up the reviews that keep it honest. Our promise is a policy that describes what your systems actually do.

Sources

Craig Petronella

Craig Petronella founded Petronella Technology Group, Inc. in 2002. He is a CMMC Registered Practitioner (CMMC-RP), Cisco CCNA, Certified Wireless Network Expert (CWNE) and North Carolina Digital Forensic Examiner (License 604180-DFE), and the author of 16 books. He leads the firm's compliance and security policy work for regulated organizations in Raleigh and across North Carolina.

Learn more about Petronella Technology Group. Questions about this template can be sent to info@petronellatech.com.

Last Updated: November 18, 2026 | Petronella Technology Group, Inc. | 5540 Centerview Dr., Suite 200, Raleigh, NC 27606 | 919-348-4912 | info@petronellatech.com