Seed-stage B2B SaaS startup
Sample for a fictional organisation · 3,282 words[Company] Vulnerability and Patch Management Policy
- Version: 1.0
- Owner: CTO
- Approved by: CEO
- Effective date: [Effective date]
- Next review date: [Review date]
4. Severity and Time to Fix
Every vulnerability is given one of four severity ratings, and its rating sets the time within which it must be fixed.
| Severity | What it means | Fixed within |
|---|---|---|
| Critical | A CVSS base score of 9.0 to 10.0, or a rating of critical from the supplier | 14 days |
| High | A CVSS base score of 7.0 to 8.9, or a rating of high from the supplier | 30 days |
| Medium | A CVSS base score of 4.0 to 6.9, or a rating of medium or moderate from the supplier | 90 days |
| Low | A CVSS base score of 0.1 to 3.9, or a rating of low from the supplier | 180 days |
CVSS is the Common Vulnerability Scoring System, a published way of scoring how severe a vulnerability is, from 0 to 10. [Company] uses the base score published for the vulnerability. Where the base score and the supplier's rating give different ratings, the higher rating applies. A vulnerability that has neither, such as a setting that is not secure, is rated by the CTO or a person the CTO has named in writing, and is treated as High until it is rated.
A vulnerability that is known to be used in real attacks is rated Critical where the affected system can be reached from the internet, and no lower than High in every other case.
The CTO, or a person the CTO has named in writing, may raise a rating, and may lower it by one level where the affected system cannot be reached from the internet and holds no confidential information. The reason goes in the vulnerability record. A rating that the paragraph above sets is never lowered.
A Critical vulnerability in a system that can be reached from the internet is fixed, or protected by a temporary measure, within 2 working days of the day its time starts. A temporary measure does not stop the time, and the vulnerability is still fixed within the time the table gives.
5. Counting the Time
The time that the table in section 4 gives is counted in this way.
- For a vulnerability that a supplier's security update fixes, the time starts on the day the supplier releases the update or on the day [Company] finds the vulnerability, whichever is later.
- For a vulnerability in the software that [Company] builds, the time starts on the day [Company] finds it. Where the vulnerability is in a component that the software uses, the time starts on the day a version of the component that fixes it is released or on the day [Company] finds the vulnerability, whichever is later. Until that version is released, the component is treated as software whose supplier has not yet released a security update.
- For a setting that is not secure, and for any other vulnerability that [Company] can fix without a security update from a supplier, the time starts on the day [Company] finds it.
- Where a supplier has not yet released a security update for a Critical or High vulnerability, the CTO, or a person the CTO has named in writing, decides within 2 working days of the day it was found whether a temporary measure is put in place until the update is released, and the decision goes in the vulnerability record. Where the vulnerability is Critical and the affected system can be reached from the internet, a temporary measure is always put in place within those 2 working days.
- Where a rating is raised, the time for the new rating starts on the day it was raised, but the vulnerability is never due later than it was before. Where a rating is lowered, the time for the new rating is counted from the day the time first started.
- Where two times in this policy apply to the same vulnerability, the earlier one applies. Where a contract sets a shorter time than this policy for a system, the contract's time applies to that system.
- A time given in days counts every day of the week. A time given in working days leaves out weekends and public holidays.
- A vulnerability is fixed on the day the fix is made. Its vulnerability record is closed only when a check, such as a repeat scan, has confirmed that the vulnerability is gone. Where the check shows that the vulnerability is still there, it has not been fixed and its time carries on from the day it first started.
9. Exceptions to the Time to Fix
Where a vulnerability cannot be fixed within its time, the member of staff assigned it asks for an exception before the time runs out, and for a renewal before the exception ends. An exception does not fix the vulnerability. It sets a later date by which the vulnerability must be fixed and the temporary measure that protects [Company] until then.
- A request for an exception gives the reason the fix cannot be made in time, the temporary measure that is or will be in place, and the date by which the vulnerability will be fixed. A request made after the time has run out also says why it was not made before then.
- An exception is given only where the fix is not yet possible: no fix exists, the fix would break something that [Company] depends on, the system cannot be interrupted before a known date, or unsupported software or an unsupported device cannot be replaced before a known date. Other work taking priority is not a reason for an exception.
- The CTO approves each exception in writing. For a Medium or Low vulnerability, a person the CTO has named in writing may approve it instead.
- An exception for a Critical vulnerability also needs the written agreement of the CEO, or of a person the CEO has named in writing.
- Nobody approves an exception that they asked for. Where the CTO is the one who asks for an exception, the CEO, or a person the CEO has named in writing, approves it instead.
- An exception ends on the date it gives. That date is no later than 30 days after the exception is approved for a Critical vulnerability, 3 months for a High vulnerability and 12 months for a Medium or Low vulnerability. An exception is renewed only in the same way, and the request to renew it says why the vulnerability cannot be fixed by the earlier date.
- An exception for a Critical or High vulnerability is recorded as a risk in [Company]'s risk register for as long as it lasts.
- When an exception ends and the vulnerability has not been fixed, the vulnerability is overdue.
Read the full example
[Company] Vulnerability and Patch Management Policy
- Version: 1.0
- Owner: CTO
- Approved by: CEO
- Effective date: [Effective date]
- Next review date: [Review date]
1. Purpose and Scope
This policy sets how [Company] finds vulnerabilities in the systems, software and devices it relies on, how it rates them, how quickly each must be fixed, how security updates are installed and how an exception is approved when a fix cannot be made in time. It sits beneath [Company]'s information security policy.
This policy covers every system, service and device that [Company] uses for its work or that holds its information, and the software on them: the laptops, phones, tablets and other devices that staff use; the servers, cloud accounts and other infrastructure that [Company] runs; the software that [Company] builds and the components that software uses; and the services that suppliers run for [Company]. Where a supplier runs a service, [Company] cannot itself fix the parts that the supplier runs, and section 7 sets what [Company] does about such a service.
This policy applies to everyone who is given access to [Company]'s systems, accounts or information: employees, contractors and anyone else working on its behalf. This policy calls them staff.
- Vulnerability: a weakness in software, in a device or in the way a system is set up that could be used to harm [Company]'s systems or information.
- Security update: a change that a supplier releases to its software or device to fix a vulnerability, often called a patch.
- Fix: a security update, a correction to software or to a setting, or the removal of the affected software or system, that removes the vulnerability.
- Temporary measure: a step that makes a vulnerability harder to use until it is fixed, such as turning off the affected feature or blocking access to it. A temporary measure is not a fix.
- Vulnerability record: the record of a vulnerability, which section 10 describes.
- Overdue: not fixed within the time this policy gives for fixing it and not covered by an exception approved under section 9.
2. Roles and Responsibilities
- The CTO: keeps this policy and reviews it under section 12; makes sure vulnerabilities are found under section 3, rated under section 4 and assigned to a member of staff to fix; approves exceptions under sections 9 and 12; and makes the review and the report that section 11 describes.
- The CEO: approves this policy and each change to it under section 12; gives the agreement that section 9 requires for an exception for a Critical vulnerability; approves an exception that the CTO asks for, under sections 9 and 12; and receives the report that section 11 describes.
- Staff assigned a vulnerability: fix it within the time that section 4 gives, record what was done in its vulnerability record under section 10, and tell the CTO as soon as it is clear that the time cannot be met.
- All staff: install security updates on the devices they use for [Company]'s work as section 6 requires, and report anything that looks like a vulnerability to the CTO as soon as they notice it.
3. Finding Vulnerabilities
- The CTO makes sure that [Company] receives the security notices published by the suppliers of the systems, software and services it relies on, and that each notice is read within 2 working days to see whether it affects [Company].
- At least once a month, and after any significant change to them, [Company] scans the servers, cloud accounts and other infrastructure that it runs with an automated tool that looks for known vulnerabilities and for settings that are not secure. At least once a month, [Company] also scans from outside its own network each system that can be reached from the internet. Where the tool is able to sign in to a system, the scan is made signed in.
- [Company] must check each change to the software it builds, with an automated tool, for known vulnerabilities in its code and in the components it uses, before the change is released. The source control system that holds [Company]'s software must be set to alert [Company] when a vulnerability is published in a component that the software uses.
- The CTO makes sure that the security settings of each service that a supplier runs for [Company], and that holds its confidential information, are checked when the service is first set up and at least every 12 months.
- Where [Company] has a penetration test or another security test carried out, each weakness the test reports is recorded and rated as a vulnerability under this policy.
- Anyone who finds or is told of a possible vulnerability, including one reported by a person outside [Company], passes it to the CTO as soon as they can. The CTO makes sure each report is looked into and, where it is a vulnerability, recorded.
- The day a vulnerability was found is the day a scan, a test or a report first showed that it affects [Company], or the day [Company] received the supplier's notice of it, whichever is earlier.
4. Severity and Time to Fix
Every vulnerability is given one of four severity ratings, and its rating sets the time within which it must be fixed.
| Severity | What it means | Fixed within |
|---|---|---|
| Critical | A CVSS base score of 9.0 to 10.0, or a rating of critical from the supplier | 14 days |
| High | A CVSS base score of 7.0 to 8.9, or a rating of high from the supplier | 30 days |
| Medium | A CVSS base score of 4.0 to 6.9, or a rating of medium or moderate from the supplier | 90 days |
| Low | A CVSS base score of 0.1 to 3.9, or a rating of low from the supplier | 180 days |
CVSS is the Common Vulnerability Scoring System, a published way of scoring how severe a vulnerability is, from 0 to 10. [Company] uses the base score published for the vulnerability. Where the base score and the supplier's rating give different ratings, the higher rating applies. A vulnerability that has neither, such as a setting that is not secure, is rated by the CTO or a person the CTO has named in writing, and is treated as High until it is rated.
A vulnerability that is known to be used in real attacks is rated Critical where the affected system can be reached from the internet, and no lower than High in every other case.
The CTO, or a person the CTO has named in writing, may raise a rating, and may lower it by one level where the affected system cannot be reached from the internet and holds no confidential information. The reason goes in the vulnerability record. A rating that the paragraph above sets is never lowered.
A Critical vulnerability in a system that can be reached from the internet is fixed, or protected by a temporary measure, within 2 working days of the day its time starts. A temporary measure does not stop the time, and the vulnerability is still fixed within the time the table gives.
5. Counting the Time
The time that the table in section 4 gives is counted in this way.
- For a vulnerability that a supplier's security update fixes, the time starts on the day the supplier releases the update or on the day [Company] finds the vulnerability, whichever is later.
- For a vulnerability in the software that [Company] builds, the time starts on the day [Company] finds it. Where the vulnerability is in a component that the software uses, the time starts on the day a version of the component that fixes it is released or on the day [Company] finds the vulnerability, whichever is later. Until that version is released, the component is treated as software whose supplier has not yet released a security update.
- For a setting that is not secure, and for any other vulnerability that [Company] can fix without a security update from a supplier, the time starts on the day [Company] finds it.
- Where a supplier has not yet released a security update for a Critical or High vulnerability, the CTO, or a person the CTO has named in writing, decides within 2 working days of the day it was found whether a temporary measure is put in place until the update is released, and the decision goes in the vulnerability record. Where the vulnerability is Critical and the affected system can be reached from the internet, a temporary measure is always put in place within those 2 working days.
- Where a rating is raised, the time for the new rating starts on the day it was raised, but the vulnerability is never due later than it was before. Where a rating is lowered, the time for the new rating is counted from the day the time first started.
- Where two times in this policy apply to the same vulnerability, the earlier one applies. Where a contract sets a shorter time than this policy for a system, the contract's time applies to that system.
- A time given in days counts every day of the week. A time given in working days leaves out weekends and public holidays.
- A vulnerability is fixed on the day the fix is made. Its vulnerability record is closed only when a check, such as a repeat scan, has confirmed that the vulnerability is gone. Where the check shows that the vulnerability is still there, it has not been fixed and its time carries on from the day it first started.
6. Installing Security Updates and Making Fixes
- Staff must install security updates for the operating system and applications of each laptop, phone, tablet or other device they use for [Company]'s work within 14 days of their release, whatever the severity of the vulnerabilities they fix.
- [Company] turns on automatic updates wherever a system, an application or a device offers them, unless the CTO has recorded a reason to install its updates in another way.
- [Company] takes a security update only from the supplier of the software or device, or from a source that the supplier names.
- Installing a security update or making another fix is a change. Where [Company]'s rules for change management cover the system, the fix is tested, reviewed, approved and recorded as those rules require, and this policy sets how soon it must be made.
- Where a fix cannot wait for the review and approval that [Company]'s rules for change management would otherwise require, it is made in the way those rules allow for a change that cannot wait.
- Where a security update cannot be installed without breaking something that [Company] depends on, the member of staff assigned the vulnerability tells the CTO, and the vulnerability is fixed in another way or has an exception under section 9 before its time runs out.
- Where there are signs that a vulnerability has been used against [Company], it is a security incident and is handled under [Company]'s incident response plan. The vulnerability is still fixed under this policy.
7. Systems and Services That Suppliers Run
Where a supplier runs a system or service for [Company], the supplier fixes the parts it runs, and the times in section 4 do not apply to the supplier. [Company] fixes the parts it controls within those times: the settings of the service, anything [Company] installs or runs on it, and any application of the service that is installed on [Company]'s devices.
- Before [Company] starts to use a service that will hold its confidential information, the CTO makes sure [Company] knows how the supplier gives notice of vulnerabilities in the service.
- Where [Company] learns of a Critical or High vulnerability in a service that a supplier runs for it, the CTO makes sure within 5 working days that the supplier is asked when it will be fixed, and the CTO, or a person the CTO has named in writing, decides whether [Company] limits its use of the service until then. The vulnerability and the decision go in a vulnerability record.
- Where a supplier does not fix a Critical or High vulnerability in a time that the CTO judges reasonable, the CTO makes sure it is recorded as a risk in [Company]'s risk register.
- Where a supplier maintains systems or devices on [Company]'s behalf, a named member of staff makes sure the supplier meets this policy on them.
8. Unsupported Software and Devices
- Software or a device is unsupported when its supplier no longer releases security updates for it.
- The CTO makes sure [Company] knows the date on which support ends for each operating system and for the other main software and devices it relies on, and that each is upgraded, replaced or removed before that date.
- Software or a device that is still in use after its support ends is recorded as a High vulnerability found on the day support ended. It is upgraded, replaced or removed within the time that section 4 gives a High vulnerability, unless it has an exception under section 9.
- An exception for unsupported software or an unsupported device is approved only where a temporary measure cuts it off from the internet and from every system that does not need to reach it.
- A vulnerability that is later found in unsupported software or an unsupported device is rated under section 4 in its own right, and its time starts on the day it is found.
- A component that [Company]'s software uses and that is no longer maintained is unsupported software under this section.
9. Exceptions to the Time to Fix
Where a vulnerability cannot be fixed within its time, the member of staff assigned it asks for an exception before the time runs out, and for a renewal before the exception ends. An exception does not fix the vulnerability. It sets a later date by which the vulnerability must be fixed and the temporary measure that protects [Company] until then.
- A request for an exception gives the reason the fix cannot be made in time, the temporary measure that is or will be in place, and the date by which the vulnerability will be fixed. A request made after the time has run out also says why it was not made before then.
- An exception is given only where the fix is not yet possible: no fix exists, the fix would break something that [Company] depends on, the system cannot be interrupted before a known date, or unsupported software or an unsupported device cannot be replaced before a known date. Other work taking priority is not a reason for an exception.
- The CTO approves each exception in writing. For a Medium or Low vulnerability, a person the CTO has named in writing may approve it instead.
- An exception for a Critical vulnerability also needs the written agreement of the CEO, or of a person the CEO has named in writing.
- Nobody approves an exception that they asked for. Where the CTO is the one who asks for an exception, the CEO, or a person the CEO has named in writing, approves it instead.
- An exception ends on the date it gives. That date is no later than 30 days after the exception is approved for a Critical vulnerability, 3 months for a High vulnerability and 12 months for a Medium or Low vulnerability. An exception is renewed only in the same way, and the request to renew it says why the vulnerability cannot be fixed by the earlier date.
- An exception for a Critical or High vulnerability is recorded as a risk in [Company]'s risk register for as long as it lasts.
- When an exception ends and the vulnerability has not been fixed, the vulnerability is overdue.
10. Vulnerability Records
Every vulnerability has a vulnerability record, which holds these seven things.
- What and where: the vulnerability, and the system, software or device it affects.
- Severity: its rating under section 4, with any change to the rating and the reason for it.
- Dates: the day it was found, the day its time started and the day by which it must be fixed.
- Assigned to: the member of staff responsible for fixing it.
- Temporary measure: any temporary measure that was put in place, and the day it was.
- Fix: what was done, the day the fix was made and how it was confirmed.
- Exception: any exception under section 9, with its reason, its temporary measure, its end date and who approved it.
A vulnerability record may be kept in the tool that found the vulnerability, provided that it holds these seven things. [Company] keeps each vulnerability record, and the results of each scan and test, for at least 3 years.
11. Monitoring and Reporting
- At least once a month, the CTO makes sure that every open vulnerability is looked at, and that each one that is overdue, or will be due within the month, has a member of staff acting on it.
- Within 5 working days of a vulnerability becoming overdue, the CTO makes sure that it is fixed or has an exception under section 9.
- The CTO tells the CEO of each Critical vulnerability that becomes overdue, within 5 working days of the day it became overdue.
- At least every 3 months, the CTO reports to the CEO the number of vulnerabilities open at each severity, the number overdue, the number fixed within their time and the number fixed after it since the last report, and each exception in force.
- Where vulnerabilities of the same kind keep coming back, or are often overdue, the CTO looks for the cause and makes sure it is dealt with.
12. Other Exceptions, Breaches and Review
An exception to a rule in this policy other than a time to fix is requested from and approved in writing by the CTO, with the reason and any conditions recorded, and lasts no longer than 12 months unless it is renewed in the same way. Where the CTO is the one who asks for the exception, the CEO approves it instead.
Hiding a vulnerability, turning off automatic updates or a scan without the approval of the CTO, or writing in a vulnerability record something that the person writing it knows to be untrue, is a breach of this policy. A breach may lead to disciplinary action up to dismissal, in line with [Company]'s disciplinary process and the employment law of the country where the person works; for contractors and other non-employees it may lead to the engagement ending. Nobody is penalized for reporting a vulnerability, or a time that has been missed, in good faith.
The CTO reviews this policy at least every 12 months and after any significant change to [Company]'s systems, suppliers or ways of working, and each change to this policy is approved by the CEO.
Disclaimer
This document is provided for informational purposes only and does not constitute legal advice. It is provided "as is", without warranty of any kind, express or implied, and no liability is accepted for any loss or damage arising from its use. It is used at your own discretion. Review it with a qualified adviser before adopting it.