If this week were a techno-thriller, you'd expect the villain to be a hacker in a dark room with three monitors. The real villain was less dramatic.
On Friday, November 13, the FAA issued a nationwide ground stop, and operators canceled about 12,462 flights (FAA). The FBI was called. The NSA, the Defense Department and CISA joined the investigation (Nextgov). The culprit turned out to be a router. Crews in El Dorado, Arkansas, were moving an FAA radio site from copper to fiber. The new equipment flooded the network with about 1,000 connection attempts every four to five seconds, and the FAA's legacy equipment couldn't keep up (FAA).
The Transportation Secretary put it best: "We think it was the installation or the router that was the bad actor" (Fortune). Every IT department in America now has a new defense. It wasn't me. It was the router.
The rest of the week followed the same pattern:
Microsoft went down globally on November 12–13, taking Outlook, Teams and Copilot with it (IBTimes). The explanation was "an issue during an earlier network capacity upgrade that surfaced following a recent network change" (Fortune). In plain English, they changed something, and then they changed something else.
JPMorgan Chase delayed transactions because of a software bug, Apple blamed an internal update, and United held departures for about 100 minutes. None reported a cyberattack (IBTimes).
The NSA, the agency whose job is knowing everything, had its classified internal messaging network go down. The agency called it "temporary degradation during scheduled routine maintenance" (The Telegraph).
There were real attacks, too. An attacker exploited a vulnerability in Optimizely, a website-testing platform. Malicious code ended up on the sites of 178 organizations, including the BBC, CNN and Zoom. The code was built to steal payment cards and redirect cryptocurrency (IBTimes). The traced haul by Monday evening was about $7,040 in crypto (IBTimes). That's 178 global brands for the price of a decent used car, which may be the worst return on investment in criminal history. It was still a very bad week for the 178. Fortune counted hundreds of companies and institutions hit by a month of cyberattacks the press has called "Black October" (Fortune).
So the threat is real. But the bigger lesson of the week is less cinematic. For now it appears that most of the damage came from inside the house: upgrades, updates, vendors and old equipment nobody wanted to touch.
That is a CFO problem as much as an IT problem. Here is what it means for founder-led and mid-market companies.
Five lessons from a very bad week
1. Change management is a financial control. Microsoft, Apple, Chase and the FAA all traced their problems to a change or a defect in their own systems, not to an outsider (IBTimes). For most companies, the riskiest day of the year is the day of a system cutover, not the day a hacker shows up. Treat major changes the way you treat a large wire transfer: a named approver, a test, a rollback plan and a time that isn't 3:30 on a Friday afternoon.
2. Your vendor's vulnerability is your vulnerability. None of the 178 Optimizely customers wrote the bad code. It came in through a trusted tool already on their websites (IBTimes). The FAA's network is operated for it by a contractor (Fortune). Expedia reported that attackers reached its systems through a third-party technology provider (IBTimes). You can outsource a function. You can't outsource the risk.
3. Legacy systems are debt, and the interest compounds. The FAA's ground stop happened because modern equipment met legacy equipment that couldn't handle it (FAA). Every company has its own version: the server under someone's desk, the spreadsheet with 40 tabs, the ERP that "works fine as long as Linda is here." It doesn't appear on the balance sheet, but it's a liability, and it comes due at the worst possible moment. Ask Colonial Pipeline. In 2021, hackers got in through a single password on a legacy VPN account that was no longer in use but still worked, and it had no multifactor authentication (Reuters). The password later turned up in a batch of leaked passwords on the dark web (Governing). The company paid a $4.4 million ransom (Governing). In fairness, the CEO told senators it was a complicated password: "It was not a Colonial123-type password" (Reuters). It didn't matter. A strong password on an account nobody remembered was still an open door.
4. Diligence finds the risk. Courage fixes it. This is the uncomfortable part. The FAA's outage happened during a modernization project. It was the upgrade, not the old system sitting quietly, that set off the failure. That is exactly why so many leadership teams keep postponing the upgrade. After Friday, the FAA paused its router installations while it investigates (ABC News), and that's the right call. The danger for the rest of us is that a sensible pause becomes a permanent one. Diligence that never ends in a decision isn't prudence. It's paralysis with a good filing system.
5. It takes courage to say "we don't know yet." On the day of the ground stop, officials said there was no sign of a cyberattack but declined to rule one out (ABC News). That's the right posture for any company in an incident. Say what you know and what you don't, and say when you'll update. Customers, lenders and boards forgive an outage far faster than they forgive a confident statement that turns out to be wrong.
What I tell clients at $10 million
Once a client reaches critical mass, roughly $10 million in revenue, 50 or more employees, and electronic banking and accounts payable, I suggest two things: a cyber audit and a cyber insurance policy.
At that size, the company is big enough to be worth stealing from and usually too small to have a full-time security team. Its money also moves electronically, which is exactly where criminals look. One spoofed vendor email asking AP to "update our bank details" can cost more than a year's profit.
Neither one is cheap anymore. For a company that size, here's a realistic budget:
Cyber audit: A risk assessment for a company under 100 employees runs about $6,000 to $16,000, and adding penetration testing, where specialists actually try to break in, adds $5,000 to $15,000 (SecurityBudget). Plan on roughly $11,000 to $30,000 for the first year.
Cyber insurance: The median annual premium for companies with $10 million or more in revenue is about $9,200 at one large broker (Vouch). Companies in the $10 million to $100 million range commonly pay $5,000 to $50,000, depending on the limits and the industry (PlayCISO).
All in, that's roughly $20,000 to $40,000 in the first year for a typical $10 million company. That's real money. It's also less than one bad wire transfer, and the two pay off together. Insurers now ask detailed questions about your controls before they quote. The audit gives you the answers, and better answers usually mean better terms.
Stuff happens
Here's the part no audit or policy fully covers: sometimes stuff just happens.
I once ran a plant with $50 million in equipment that needed to be productive 80% of every 24-hour day. It came to a screeching halt because a squirrel decided the lead from the main transformer into the plant would make good carpeting for his nest. It was a fatal decision for the squirrel and a very expensive one for us. No cyber audit would have caught him. He didn't hack anything. He just chewed.
And in fairness to the squirrel, I've been the bad actor myself. In one of my first jobs, I shut down the can filler on the Rheingold Brewery line in Orange, New Jersey, by feeding a sleeve of can tops into the machine upside down. One small, honest mistake, and a major beer line stopped cold. The Orange plant closed for good in 1977, when Rheingold was sold to Christian Schmidt Brewing (Wikipedia). For the record, I maintain that had nothing to do with me.
The FAA's router, Microsoft's network change, a squirrel in a transformer, a young man with a sleeve of can tops: none of them were sophisticated, and all of them stopped the operation. You can't prevent every one of them. The point is this: when stuff happens, how quick and efficient is your response? Who knows what to do, who has the authority to do it and how many hours pass before you're running again? That's the number that shows up in your results.
The five-question outage test
Ask your leadership team these this quarter:
Which planned system changes in the next 90 days could stop revenue, and does each one have a tested rollback?
Which vendors run code on our website or touch customer and payment data, and when did we last review their access?
What is the oldest system we can't operate without, and what does keeping it alive cost compared with replacing it?
If a core system, or the power, went down at 3:30 on a Friday, who decides, who tells customers and lenders, and how many hours until we're running again?
Which decision have we been studying for more than a year?
If question five got a laugh, you've found the place where diligence needs to become courage.
The bottom line
In six weeks, we mark the 250th anniversary of Washington crossing the Delaware to attack Trenton on December 26, 1776. He knew his army's enlistments were running out and that waiting for better conditions meant losing the army. His intelligence told him where the enemy was. Courage put him in the boat on a freezing Christmas night.
Diligence and courage aren't opposites. Diligence tells you where the ice is. Courage gets you across the river anyway, with a plan for what happens if the boat leaks.
There's a third ingredient, and it may be the hardest. It takes faith to focus on the future. After a week like this one, it's tempting to spend every meeting on what broke and who to blame. Diligence looks backward at what went wrong. Courage acts on what's in front of you. Faith keeps your eyes on where you're going: the belief that the modernized system, the new vendor, the hard decision will leave the company stronger than the one you have today. Washington's army had no guarantee of victory on the far bank of the Delaware. It had faith that the cause was worth the crossing. Companies that only look in the rearview mirror stop investing, and companies that stop investing eventually become the legacy system.
So check your routers, read your vendor contracts and know your rollback plan. Then make the decision. The router won't make it for you. Apparently, it has other plans.
If you want an outside view of your company's operational risk, vendor exposure or systems roadmap, let's talk.
— John Mortenson, Fractional CFO · Straiteis Consulting

