The Blueprint

Logging, detection and incident response.

Noticing when something goes wrong and dealing with it well: logging, monitoring and alerts, an easy way to raise a concern, handling and reporting incidents, and learning from each one.

DET 29 requirements

Foundation

RW-DET-01Security logging switched on

What
Switch on logging of sign-ins, admin changes and other security events in your important systems and cloud services.
Why
Without logs there is no way to tell whether an attacker got in, what they touched or how long they were there, so switching logging on gives you the record you'll need to spot an attack and work out what happened.
Protects
Confidentiality · Integrity · Availability

RW-DET-02One easy way to raise a concern

What
Give staff one simple, well-known way to report anything suspicious, any security weakness and any mistake of their own straight away, such as a phishing email, a lost laptop or data sent to the wrong person, and make clear that reporting promptly will not count against them.
Why
Staff are often the first to notice that something is wrong, and an easy way to tell someone, without fear of blame, means problems reach the people who can deal with them while there's still time to limit the harm.
Protects
Confidentiality

Read the guide: The memory sticks nobody can account for

RW-DET-03A plan for handling incidents

What
Write an incident response plan naming who leads the response, when senior leadership takes charge, who needs to be contacted and the first steps to take for the incidents you're most likely to face, such as ransomware, a taken-over email account or a lost laptop, and review it every year and after each significant incident.
Why
In the first hours of an incident, people who already know who is in charge and what to do first can limit the damage, whereas a team working it out as it goes loses time and makes mistakes under pressure.
Protects
Confidentiality · Integrity · Availability

RW-DET-04A log of every incident

What
Record every incident and suspected incident in one log, noting what happened, who is dealing with it and what was done, and keep each entry open until the matter is resolved.
Why
A single record means nothing that's been reported gets forgotten, makes clear who is responsible for each problem, and builds up a picture over time of where your weak points are.
Protects
Confidentiality · Integrity · Availability

Read the guide: The memory sticks nobody can account for

RW-DET-05Incidents reported on time

What
Tell the regulator, and the customers and people affected, about a data breach or other significant incident whenever the law or your contracts require it, making each required notice and follow-up report, such as an early warning within 24 hours and a final report, within the time allowed.
Why
Notifying on time gives those affected the chance to protect themselves, for example by changing passwords or watching for fraud, and avoids the fines and loss of trust that follow when an organisation is found to have kept quiet.
Protects
Confidentiality

RW-DET-06Urgent security warnings acted on

What
Sign up to receive security warnings from your national cyber security agency and your main suppliers, and act quickly on any that affect the systems you use.
Why
Attackers move fast once a weakness becomes public, so hearing about it promptly and acting on it quickly, rather than waiting for the next routine update, closes the gap before it can be used against you.
Protects
Confidentiality · Integrity · Availability

Established

RW-DET-07Logs kept long enough to look back

What
Keep security logs for a set period that is long enough to investigate an attack which went unnoticed for months.
Why
Attackers are often inside a network for weeks or months before anyone notices, so if the logs have already been deleted by then, you can't tell how they got in, what they took or whether they've really gone.
Protects
Confidentiality · Integrity

RW-DET-08Logs that can't be altered

What
Protect your security logs so that nobody, including your own admins, can change or delete them before their retention period ends.
Why
An attacker who gets in will often try to erase the record of what they did, so logs that can't be altered keep the evidence trustworthy when you need it to understand and recover from an incident.
Protects
Integrity

RW-DET-09Every clock telling the same time

What
Keep the clocks on all your systems, devices and cloud services in step by setting them from reliable time sources that agree with each other.
Why
When every system records the same time, events from different logs can be put in the right order, which is often the only way to piece together how an attack unfolded and to have that timeline accepted as evidence.
Protects
Integrity

RW-DET-10Alerts looked into promptly

What
Have someone check security alerts every working day and investigate any that could be a sign of an attack.
Why
Alerts only protect you if someone acts on them, and many serious attacks were flagged by a tool days before the damage was done, but nobody looked.
Protects
Confidentiality · Integrity · Availability

RW-DET-11Out-of-the-ordinary activity flagged

What
Set alerts for activity that is out of the ordinary for your organisation, such as large downloads, mass deletions, new forwarding rules on mailboxes or sign-ins at odd hours.
Why
Attackers using stolen accounts, and insiders misusing their access, often do nothing a security tool would class as an attack, so watching for behaviour that differs from normal is how they get noticed.
Protects
Confidentiality · Integrity

RW-DET-12Who looked at sensitive records

What
Check regularly who has opened your most sensitive records, such as personnel files or customers' financial details, and follow up any access that has no clear business reason.
Why
People with legitimate access to sensitive records can still look at them out of curiosity or for gain, and knowing that access is checked puts them off as well as catching those who do it anyway.
Protects
Confidentiality

RW-DET-13Incidents rated by severity

What
Agree in advance how you'll judge whether an event counts as an incident and how serious it is, and use those levels to decide who must be told and how quickly the response must start.
Why
Agreed levels stop serious incidents being treated as routine faults, and stop minor ones pulling in senior people for no reason, so the right people are involved at the right speed.
Protects
Confidentiality · Integrity · Availability

RW-DET-14Incidents stopped from spreading

What
Contain each incident as early as possible, using steps worked out beforehand for the incidents you're most likely to face, such as cutting an infected computer off the network, disabling a taken-over account or blocking a malicious sender.
Why
Ransomware and attackers using stolen accounts spread quickly, so cutting them off early, with steps that have already been thought through, limits how many systems and how much data they can reach.
Protects
Confidentiality · Integrity · Availability

RW-DET-15Cause removed before systems return

What
Remove the cause of each incident, such as malware, an attacker's accounts or the weakness they got in through, and confirm the affected systems are clean before putting them back into use.
Why
Bringing a system back while the attacker still has a way in invites a second and often worse incident, so making sure the cause has gone first means the recovery lasts.
Protects
Confidentiality · Integrity · Availability

RW-DET-16Evidence kept intact

What
Collect and keep the evidence from serious incidents, such as logs, copies of affected systems and messages, in a way that shows it hasn't been changed since it was gathered.
Why
Evidence handled with care lets you find out what really happened and can be relied on by insurers, regulators, the police or a court, whereas evidence that may have been altered is of little use when you need it most.
Protects
Integrity

RW-DET-17People kept informed during an incident

What
Agree in advance who speaks for the organisation when an incident happens, and how staff, customers, suppliers and, where needed, the public will be updated.
Why
Clear and timely updates from one agreed voice stop rumours and contradictory messages, keep the trust of customers and partners, and avoid hasty statements that could weaken your legal or regulatory position.
Protects
Confidentiality · Integrity · Availability

RW-DET-18Ransom decisions settled beforehand

What
Decide in advance who will make the decision on any ransom or extortion demand, and which legal, insurance and specialist advisers they will call on.
Why
A ransom demand arrives under intense pressure and often with a deadline, so knowing who decides and who advises means the choice is made calmly, lawfully and with the full facts rather than in a panic.
Protects
Availability

RW-DET-19Response plan rehearsed

What
Rehearse your plan for handling incidents at least once a year with everyone who has a part in it, including senior managers, using a realistic scenario.
Why
Rehearsing shows up gaps in the plan, such as missing contacts or unclear decisions, while there's still time to fix them, and means people have practised their part before they have to play it for real.
Protects
Confidentiality · Integrity · Availability

RW-DET-20Lessons taken from each incident

What
Review each significant incident once it's over to find its root cause, update the risk register with what it showed, and follow the improvements you agree through until they're done.
Why
Most incidents expose a weakness that can be fixed, so finding the real cause and seeing the fixes through stops the same thing happening again.
Protects
Confidentiality · Integrity · Availability

RW-DET-21Silent log sources noticed

What
Raise an alert whenever an important system stops sending logs or a log store is close to full.
Why
Attackers sometimes switch logging off to cover their tracks, and logs can also stop because of a fault, so being told straight away means a gap in your records is noticed within hours rather than discovered days later when you need them.
Protects
Integrity

RW-DET-28A backup way to communicate in an emergency

What
Set up a secure way for the people who respond to incidents to reach each other that doesn't depend on your normal email, chat or phone systems, and check from time to time that it works.
Why
In a serious attack your email and chat may be down, or being read by the attacker. A separate, secure way for the response team to reach each other lets them work together without tipping the attacker off or losing hours finding one another.
Protects
Confidentiality · Availability

Read the guide: When the platform everyone uses stops

Advanced

RW-DET-22Critical systems watched day and night

What
Have your most critical systems monitored around the clock by a team, your own or a provider's, with the skills and authority to act out of hours.
Why
Many attacks, ransomware above all, are launched at night or over a weekend because nobody is watching, so round-the-clock cover means an attack can be stopped in its first hour instead of being found on Monday morning.
Protects
Confidentiality · Integrity · Availability

RW-DET-23Traces of an attack linked together

What
Bring the logs from devices, networks, sign-ins and cloud services together in one place where events from different sources can be linked and searched.
Why
An attack usually leaves a small trace in several places, none of which looks serious alone, so linking them reveals the whole picture and lets you find related activity quickly during an investigation.
Protects
Confidentiality · Integrity · Availability

RW-DET-24Detection shaped by current threats

What
Use information about the attackers and methods currently targeting organisations like yours to add to and adjust your alert rules.
Why
Detection tuned to the threats you actually face picks up the methods attackers are using now, rather than only those that were common when the rules were first written.
Protects
Confidentiality · Integrity · Availability

RW-DET-25Hunting for hidden intruders

What
Search your systems regularly for signs of an attacker who has got past your defences without setting off any alerts.
Why
Skilled attackers deliberately avoid the activity that triggers alerts, so actively looking for them is the only way to find an intruder who has stayed quiet.
Protects
Confidentiality · Integrity · Availability

RW-DET-26Leaked information spotted early

What
Watch for your organisation's documents, staff passwords and customer data appearing on leak sites, criminal forums and public paste sites.
Why
Finding your own data or passwords offered for sale is often the first sign of a breach you hadn't noticed, and it gives you the chance to reset accounts and warn people before the information is used against them.
Protects
Confidentiality

RW-DET-27Attacks simulated against your defences

What
Have a team acting as an attacker carry out a realistic attack on your systems and people, to test whether your monitoring spots it and your response stops it, and fix the gaps it reveals.
Why
Tabletop exercises test the plan on paper and break-in tests find weaknesses, but only a realistic attack shows whether your alerts go off, whether anyone acts on them and how far an attacker could get before being stopped.
Protects
Confidentiality · Integrity · Availability

RW-DET-29Network traffic watched for attacks

What
Watch the traffic crossing your internet edge and the boundaries between your network zones for signs of attack, and alert on what you find.
Why
Some attacks leave little trace in system logs but show up in the traffic they create, such as data being sent out or an attacker moving between machines. Watching the network itself catches what logs alone miss.
Protects
Confidentiality · Integrity · Availability