I am not going to tell you that hackers have your holiday calendar. The reality is more mundane and more useful to know: attacks that start when nobody is looking get more time to work. That is the entire mechanism. It is arithmetic, not menace.
An intrusion that begins on a Tuesday at 10 a.m. runs into somebody noticing something odd — a login prompt that should not be there, a file that will not open, a colleague asking why they got a strange email. The same intrusion starting at 4 p.m. on the Friday before a holiday has days of quiet, spent moving between systems, locating backups, and getting everything in place before anything visible happens. Same attack, different amount of undisturbed room.
So the December question is not “are we a target.” It is how long would something run before a person noticed, and does that person have what they need to act. Everything below is preparation for those two questions, and most of it takes an afternoon.
Decide Who Is Actually on Call — Then Check They Can Get In
Every business I ask has an on-call answer. A lot of them turn out not to work when we check.
Write out the holiday period day by day and put a named person against each block, plus a named backup. Not a team. Not “whoever picks up.” A person, with a phone number that is on the list.
Then verify the part that fails most often — access:
- Have that person log in now, from home, on the device they will have with them. Otherwise, expired passwords and lapsed VPN certificates get discovered at 11 p.m. on December 26.
- Confirm their multi-factor device travels with them. A hardware key in a desk drawer at the office is no use to someone at their in-laws’ house.
- Keep administrative credentials in the shared password manager, not in one person’s head or on the laptop of the one person who is out of the country.
- Check that at least two people can approve or elevate. Single-person dependencies are fine right up until that person is on a plane.
- Collect vendor support and account numbers in advance — internet, phone, line-of-business software, IT provider. Hunting for an account number during an outage is a bad use of the hour.
Build an Out-of-Band Contact List
Here is the scenario that catches people. Your email is the thing that is compromised. Now you cannot use email to coordinate the response, and you should not be discussing containment in a system an attacker may be reading.
The fix is boring and cheap: one page, on paper, printed and carried. Personal cell numbers for on-call staff and leadership. Your IT provider’s after-hours number. Your internet provider. Your cyber insurance carrier’s incident line and policy number — most policies require prompt notice and many direct you to their panel responders. Your bank’s fraud line. Outside counsel. And in healthcare, whoever owns breach notification decisions.
Set up a group text on personal phones before you need it, and agree in advance that this is the channel if company systems are suspect. Ten minutes now, an hour of scrambling saved later.
Set the Money Rules Before Anyone Leaves
Business email compromise does not need malware or a holiday. It needs a plausible request, a distracted approver, and an executive who cannot be reached to confirm. December supplies the last two for free.
The version we see most often is a vendor emailing updated banking details, or a message that appears to come from an owner traveling and needing a payment handled quietly. We covered how those accounts get taken over in the first place in why hackers love small business email. The controls are procedural, not technical:
- Any change to banking details gets verified by voice, using a phone number you already have on file — never a number in the request itself. No exceptions, including for vendors you have paid for a decade.
- Set a dollar limit above which a second named approver is required, and make sure the second approver is available during the period. A control that everyone routes around because the approver is in Hawaii is not a control.
- Say out loud, in writing, that leadership will never ask for an exception because they are traveling. This one sentence, sent to the finance team in advance, defuses the most common pretext there is.
- No approvals on email or text alone for new payees or changed details. A voice call to a known number, or nothing.
- Tell the person who processes payments that they will never be in trouble for slowing a payment down to verify it. Most of these succeed because someone junior did not want to question a request that looked like it came from the top.
Freeze the Risky Changes, Not the Maintenance
A change freeze during a skeleton-crew period is sensible. It is also frequently done backwards — people pause the security work and keep the migrations.
Pause: migrations, firewall rule rewrites, mail routing and DNS changes, tenant or licensing restructures, major version upgrades, and anything with “consolidation” in the name. Not because they are dangerous by nature, but because the person who can unwind them at 9 p.m. on a Sunday is on vacation.
Keep running: security patching — critical and actively-exploited vulnerabilities inside 72 hours, everything else weekly. Backups, with the job results actually read. Monitoring and alerting. Endpoint protection updates.
The test we use for anything in between: if this goes wrong, is the person who can reverse it reachable within the hour? If not, it waits until January.
Make Sure an Alert Reaches a Person
Alerting quietly rots. A distribution list still exists but the two people on it left. The on-call number in the monitoring tool belongs to a former employee. Security alerts go to a mailbox nobody has opened since March. None of it shows up until the night it matters.
Before the holiday, spend twenty minutes verifying end to end:
- Where does a critical alert go, and who receives it on a Saturday?
- Generate a test alert and confirm someone’s phone actually lights up.
- Check the escalation path — what happens if the first person does not acknowledge?
- Confirm backup job failures notify somebody, and that a failed job over the holidays would be seen rather than found in January.
- Confirm your provider knows which of your people to call and in what order over the holiday period. Send them the on-call list.
If you are not sure what your provider is watching or what happens when something fires at 2 a.m., that is a fair thing to ask directly — we wrote about what 24/7 monitoring actually means and the questions worth asking.
Two Things Worth Doing While It Is Quiet
Test a restore. Not a backup report — an actual restore of a real file to a real place. Our standing recommendation is a restore test quarterly and one full restore test a year, and a quiet week is a reasonable time to knock out the quarterly one while the people who would need it are still around.
Check your MFA coverage on the accounts that matter. Microsoft’s research has consistently found that multi-factor authentication blocks more than 99% of automated account-compromise attempts. But MFA is not a force field. Adversary-in-the-middle phishing kits steal the session token after you approve the prompt, which defeats app codes and push approvals alike. For finance, email admin, and anything touching money, use phishing-resistant MFA — passkeys or a hardware security key. If you are still standing up MFA generally, start with the MFA guide for business owners.
If Something Does Happen
Two instructions that are worth having written on the same page as your contact list. First: do not power the machines off. Disconnect them from the network if you need to contain something, but shutting down destroys evidence that responders need and can make recovery harder. Second: call the list. Your IT provider and your insurance carrier both need to hear from you early, and insurers frequently have requirements about who does the response work.
The most common entry points are stolen credentials, exposed remote access, and phishing email — which is worth knowing because it tells you what the first questions will be. We walk through the sequence in what happens in the first 72 hours after a breach.
None of this is dramatic or requires new spending. A list of names, a printed page, a rule about payments, and twenty minutes confirming an alert reaches a phone. Do it in the second week of December and the rest of the month is genuinely quieter.
Related Questions
Why do attacks seem to happen on weekends and holidays?
Mostly because that is when they succeed, not because attackers are unusually strategic. An intrusion that starts on a Tuesday morning runs into someone noticing something odd within hours. The same intrusion starting on the Friday before a holiday can run for days before anyone looks at it, which is more time to move between systems, find backups, and stage everything before doing anything visible. The attack is not different. The amount of undisturbed time is.
What is the single most useful thing to do before the holidays?
Write down who is on call for each block of days, then confirm that person can actually get in. Verify their access works, their multi-factor device is with them, and they know where the credentials live. A named on-call person who cannot log in is the most common failure we see, and it is found at the worst possible moment. Second most useful: a one-page contact list on paper, because if email is what is compromised you cannot use email to coordinate the response.
Should we stop making IT changes over the holidays?
Pause discretionary changes, not maintenance. Migrations, firewall rewrites, mail routing changes, and major upgrades should wait until the people who understand them are back and available to reverse them. Security patching, backups, and monitoring must keep running. The rule we use is simple: if something goes wrong with this change, is the person who can undo it reachable within the hour? If not, it waits.
How do we protect against payment fraud while executives are traveling?
Set the rule before anyone leaves and put it in writing. Any change to banking details, any new payee, and any payment above a limit you choose gets verified by phone using a number already on file, never a number in the request. Require a second approver above that limit, and make clear that no executive will ask for an exception because they are traveling. Urgency, secrecy, and an unavailable approver are the three ingredients in almost every business email compromise attempt, and December supplies all three.
Want a Coverage Plan on Paper Before the Holidays?
We’ll help you put together the on-call schedule, verify the access actually works, and build the one-page contact list you hope you never open. It is a short conversation and it removes the worst hour from your December.
Talk Through Holiday Coverage (888) 735-7701