A patch management policy is a short governance document that says which systems you patch, who is responsible, how fast each kind of update must be applied and what happens when a patch cannot go on. The patch management process is how you carry that out day to day. For a UK organisation the quickest way to a defensible policy is to anchor your timescales to NCSC guidance and the Cyber Essentials rule of 14 days for critical and high-risk fixes, then write down who does what.
This guide is for IT managers, directors and office managers at small businesses, schools, colleges and charities who need a policy that fits on two pages and survives an audit. It covers the contents, sensible timescales, roles, exceptions and the link to continuous vulnerability management.
Key Takeaways
- A policy sets the rules and accountability. The process is the repeatable cycle of finding, testing, deploying and verifying updates.
- NCSC best practice is 5 days for internet-facing software, 7 days for operating systems and applications, and 14 days for internal or air-gapped systems.
- Cyber Essentials v3.3 requires software in scope to be updated within 14 days of release for critical or high-risk vulnerabilities.
- Every exception needs an owner, a reason, a review date and a compensating control. Temporary workarounds must be time-bounded and tracked.
- Scan at least monthly and review the whole process regularly so it keeps pace with changes in your estate.
Table of Contents
- What a patch management policy is, and how it differs from the process
- What your patch management policy should contain
- Patching timescales: what UK guidance says
- How to write the policy in six steps
- Finding vulnerabilities and prioritising by risk
- Testing, deployment and verification
- Exceptions, workarounds and unsupported systems
- From patching to continuous vulnerability management
- Patching and Cyber Essentials
- A sample policy skeleton you can adapt
- Getting help with the harder parts
- FAQ
- Sources
What a patch management policy is, and how it differs from the process
The policy is the rulebook. It states scope, responsibilities, target timescales, how exceptions are approved and how you report. Think of it as the document an auditor, an insurer or a customer's procurement team reads to decide whether you take updates seriously.
The process is the operating cycle underneath: identify what you own, find out which updates and vulnerabilities apply, prioritise by risk, test, deploy, verify and record. Tanium describes the policy as the governance layer that gives the technical process structure and accountability, and that split is a useful way to keep your own documents tidy. Put the rules in the policy and the step-by-step instructions in a separate procedure, so you can change a tool or a schedule without rewriting the policy.
Small organisations often skip the policy because "updates are automatic". Automatic updates are a good start, and the NCSC itself advises adopting a policy to update by default, applying updates as soon as possible and ideally automatically. But you still need to write down the devices that sit outside that default, such as servers, routers, firewalls, line-of-business applications and anything a supplier manages.

What your patch management policy should contain
A comprehensive policy covers scope, roles, patch classification and risk-based prioritisation, testing and deployment, exceptions and reporting, according to the Tanium guidance on writing one. For vendors, the NCSC says plans should describe proactive vulnerability detection, risk-based prioritisation and a periodically tested response playbook. Here is how that translates into headings you can use.
| Policy section | What to write | Example for a 30-person organisation |
|---|---|---|
| Purpose and scope | Which devices, operating systems, applications, cloud services and network equipment are covered | All company laptops, servers, firewalls, routers, mobile devices and cloud tenants |
| Roles | Who finds, approves, deploys and signs off | IT lead deploys, department head approves downtime, director owns risk |
| Asset inventory | Where the list of assets lives and who keeps it current | A single spreadsheet or management console, reviewed monthly |
| Classification | How you rate urgency | Critical, high, medium, low, plus "actively exploited" as an override |
| Timescales | Maximum time from release to deployment | See the table below |
| Testing and rollout | Pilot group, rollback plan | Ten percent of devices first, then everyone |
| Exceptions | Approval, compensating controls, review date | Director signs, reviewed every 90 days |
| Unsupported software | Removal or isolation rules | Retire, or segregate with documented risk acceptance |
| Reporting | What gets reported, to whom, how often | Monthly one-page summary to the director |
| Review | When the policy itself is reviewed | Annually and after any major incident |
Keep it short. A policy that people read and follow beats a forty-page template that nobody owns. If you start from a university or public sector template, strip it back to what your organisation actually runs.
Patching timescales: what UK guidance says
The most common question about a policy is "how fast?". You do not need to invent numbers. The NCSC gives best-practice targets of 5 days for internet-facing services and software, 7 days for operating systems and applications, and 14 days for internal or air-gapped services and software, with a faster response for actively exploited vulnerabilities.

For comparison, the Department for Work and Pensions' 2024 standard sets 7 days for critical vulnerabilities that meet specified criteria, otherwise 14 days to patch or mitigate, and 45 days for medium or low-risk items. That is a government department standard rather than a requirement for you, but it shows what a tiered policy looks like in practice.
A workable table for most small organisations:
| Category | Target | Basis |
|---|---|---|
| Actively exploited vulnerability | As fast as possible, treated as an incident | NCSC guidance on active exploitation |
| Internet-facing services and software | 5 days | NCSC best practice |
| Operating systems and applications | 7 days | NCSC best practice |
| Internal or air-gapped systems | 14 days | NCSC best practice |
| Critical or high-risk (CVSS 7 or above) in Cyber Essentials scope | No later than 14 days from release | Cyber Essentials v3.3 |
The NCSC also suggests agreeing in advance with IT incident owners that rapid remediation of an actively exploited flaw gets the same priority as a total system compromise. Write that into your policy as one sentence: it removes the argument on the day.
If your policy sets stricter targets than Cyber Essentials, you will pass that part of the assessment comfortably. If you set looser ones, the assessment is where it shows.
How to write the policy in six steps
- List what you own. You cannot patch what you do not know about. Include laptops, phones, servers, network kit, cloud services and anything running in a cupboard or a supplier's data centre. Name the person who keeps the list current.
- Decide who is accountable. Security or IT identifies and prioritises, operations deploy and verify, application owners test and approve, and a director provides oversight, resources and escalation. In a small team one person may hold several roles; write the names down anyway.
- Set your categories and timescales. Use the table above as a starting point and adjust for your environment.
- Define testing and rollout. Say how you trial updates and how you roll them back.
- Define exceptions. Who can approve one, for how long and with what safeguards.
- Decide on reporting and review. Agree a short monthly report and an annual policy review.
Then get it signed off by someone with authority and tell staff where to find it. A policy nobody has seen will not hold up when a customer asks to read it.
Finding vulnerabilities and prioritising by risk
Patch lists come from two places: vendor release notices and your own scanning. Vulnerability assessment tools automate the detection of security gaps and so identify the specific vulnerabilities that need remediation. The same NCSC guidance advises scanning at least monthly, and again straight after changes made to fix a critical issue.
Not everything can be fixed at once, so triage. The NCSC recommends sorting findings into three groups: fix, acknowledge and investigate.
- Fix: prioritise internet-facing and high-impact issues.
- Acknowledge: accept the risk with a written rationale and a planned review date.
- Investigate: a temporary status for ambiguous cases, not a place to park things.
Those three words map neatly to a spreadsheet column. If you want a practical walk-through of ranking findings beyond a raw severity score, our guide on vulnerability prioritisation for UK teams covers it.
Testing, deployment and verification
Testing should be proportionate. For workstations, a small pilot group that receives updates first is usually enough. The NCSC points to phased or canary deployments as a way to do "live testing", letting you pause or roll back if something breaks while still patching quickly.
A good deployment section of the policy answers these questions:
- Who is in the pilot group, and how long does the pilot last?
- What counts as a failed update, and who can halt the rollout?
- How do you roll back, and where are the backups?
- How do you confirm the patch landed? Re-scan or check the management console; do not assume.
Verification is the step most often skipped. A patch that was approved but failed on a third of the devices is not a patch. Record the result, not the intention.
Exceptions, workarounds and unsupported systems
Some systems cannot take updates automatically: a finance application that breaks on every release, a machine controlling equipment, a supplier-managed appliance. Where an organisation judges automatic updates too risky, the NCSC says the system should be added to an exception list, with updates still put through the required safety testing.
Your policy should require, for every exception:
- a named owner and a business reason;
- the compensating control, for example network segregation or restricted access;
- an expiry or review date;
- sign-off by someone senior enough to accept the risk.
When a vendor offers a temporary workaround before a full fix, treat it as a loan. The NCSC says such mitigations should be time-bounded and actively tracked until the proper patch is available and applied. Add a line to your tracker with the date you will revisit it.
Software that no longer receives security updates is a separate case. Cyber Essentials expects supported software, so the policy should say that unsupported systems are removed, replaced or, where genuinely unavoidable, isolated and documented.
From patching to continuous vulnerability management
Readers often ask about the gap between continuous vulnerability assessment and continuous vulnerability management. Assessment is the detection half: scanning and reviewing to find weaknesses. Management is the whole loop: detect, prioritise, fix, verify and improve. Patching sits inside it as the most common fix.
The NCSC says the process should be regularly reviewed so it keeps pace with organisational changes, new threats and newly discovered vulnerabilities, and that third-party penetration tests are one way to check it is working. That is a useful design rule for your policy: include a review trigger such as a new office, a new cloud platform, an acquisition or a significant incident, not just the calendar.
A vulnerability scan tells you what is missing. A penetration test shows what an attacker could actually do with what is missing. If you are unsure which you need, see our explainer on penetration testing versus vulnerability scanning.
Patching and Cyber Essentials
If you hold, or want, Cyber Essentials, your policy becomes evidence. Under the Cyber Essentials Requirements v3.3, software in scope must be updated within 14 days of release where the fix addresses a critical or high-risk vulnerability, or one with a CVSS v3 base score of 7 or above. Everything in scope counts: operating systems, applications, firmware on routers and firewalls, and cloud services you configure.
Practical points for a smooth assessment:
- Your scope list and your asset inventory should match.
- Write "14 days" or shorter in the policy, and make sure your tooling can show it.
- Keep a record of when updates were released and when they were applied.
- Remove or replace unsupported software before assessment day.
The 2026 changes also matter, so read our summary of the Cyber Essentials April 2026 changes and the note on multi-factor authentication becoming a hard fail. If a customer, tender or insurer has asked for the certificate, this page explains why clients require Cyber Essentials.
A sample policy skeleton you can adapt
Use this as a first draft and replace the bracketed items.
- Purpose: to reduce the risk of compromise by applying security updates promptly and consistently.
- Scope: all [devices, servers, network equipment, cloud services] owned, leased or managed by [organisation].
- Responsibilities: [IT lead] maintains the inventory and deploys updates; [director] approves exceptions and receives the monthly report.
- Standard: updates apply automatically wherever possible. Critical or high-risk updates are applied within [5/7/14] days depending on exposure and never later than 14 days.
- Actively exploited vulnerabilities: treated as incidents and handled immediately.
- Testing: [pilot group] receives updates first; rollback steps are documented.
- Exceptions: recorded with owner, reason, compensating control and review date no more than [90] days ahead.
- Unsupported software: removed or isolated, with documented director approval.
- Verification and reporting: scans at least monthly; patch status reported monthly.
- Review: annually, and after any significant change or incident.
Getting help with the harder parts
Writing the policy is the easy bit. Keeping evidence, spotting what your scans miss and proving the controls work is where small teams stall. At Solusec you deal directly with the testers and assessors, not a sales team, and the advice is designed to be practical and proportionate for SMEs, schools and charities.
If Cyber Essentials is the goal, our Cyber Essentials certification is priced at £320 to £600 depending on organisation size, a gap analysis is £300, and urgent turnaround adds £100. We can also help you with vulnerability management services so your scanning and patching run as a continuous cycle.

To check that your patching holds up against a real attacker, our penetration testing starts at £2,500 for a CMS website test and £3,000 for an external IP test. If you need Cyber Essentials Plus, where the assessor verifies your systems technically, see Cyber Essentials Plus.
FAQ
How often should patch management be performed?
Treat it as continuous rather than periodic. Apply updates automatically where you can, scan at least monthly as the NCSC advises, and meet your target timescales for critical and high-risk items. Many organisations also hold a monthly review to check the report and chase exceptions.
What is the difference between a patch management policy and a patch management process?
The policy defines the rules, roles and timescales and is reviewed rarely. The process is the working cycle of identifying, testing, deploying and verifying updates, and it is carried out constantly. You need both: the policy shows what you have committed to, and the process shows you are doing it.
How quickly must critical patches be applied for Cyber Essentials?
Updates for critical or high-risk vulnerabilities, or those with a CVSS v3 base score of 7 or above, must be applied within 14 days of release for software in scope. The NCSC's wider best practice is tighter for internet-facing systems, at 5 days, so many organisations set a stricter internal target.
Do small organisations really need a written policy?
If you want Cyber Essentials, respond to customer security questionnaires or bid for contracts, a written policy gives you something concrete to show. It can be one or two pages. What matters is that it names owners, sets timescales and records how exceptions are handled.
What should I do when a patch breaks something?
Use your rollback plan, pause the rollout and log the system as a temporary exception with an owner and a review date. Add a compensating control such as restricted access, and track any vendor workaround until the full fix is applied.
Sources
- How to create a patch management policy in 5 steps, Tanium
- NCSC: Software Security Code of Practice, implementation guidance
- NCSC: Vulnerability management, update by default
- NCSC: Vulnerability management, responding to active exploitation
- NCSC: Vulnerability management, carry out assessments
- NCSC: Vulnerability management, verify and review the process
- NCSC: Vulnerability scanning tools and services
- DWP Technical Security Standard: Security Patching
- Cyber Essentials Requirements for IT Infrastructure v3.3
Recommended
- Managed Penetration Testing
- Vulnerability Management Services
- Vulnerability Disclosure Policy & Management
- Annual Penetration Testing
This article was created with assistance of AI. We never use AI with your data unless given explicit permission.
