Every acquisition comes with a list of things you’re buying: customers, products, people, revenue. It also comes with things nobody puts on the list. Old applications. Accounts for contractors who left years ago. Logins for drivers, field techs, and partners that no one on your current team has ever seen.
That second list is where attackers are starting to shop. Last week, one of the world’s largest shipping companies learned that the hard way.
What happened at Maersk
On November 11, Maersk disclosed that an attacker had accessed one of its local legacy systems and pulled data from it. The system came to Maersk through a previous acquisition. A ransomware group called WhiteArmor had already listed Maersk on its leak site, claiming about 100 GB of data and demanding $5 million in bitcoin.
The two sides tell different stories. WhiteArmor claims it took driver records, staff accounts, and customer delivery details. Maersk says the incident is isolated, its core platforms and partner connections are unaffected, and its findings so far suggest no sensitive data was exposed. How the attacker got in hasn’t been disclosed.
I’m not here to second-guess Maersk’s response. By its own account, it caught the problem, contained it, and kept operations running. What interests me is the part of the story that has nothing to do with Maersk specifically: the door was a system the company didn’t build. It came with a deal.
Why inherited access goes unwatched
Nobody decides to leave an acquired system exposed. It happens because integration has a natural order, and security usually isn’t first in line.
The first months after a deal are about keeping the business running: payroll, customers, revenue, retention. Systems that still work get left alone, because migrating them is expensive and breaking them is worse. “We’ll retire it next year” turns into three years.
Meanwhile, ownership quietly dissolves. The people who built and ran the system leave or move on. The new security team knows it exists in a spreadsheet somewhere, but doesn’t know who logs into it, from where, or why. And because it sits off to the side of the core environment, it often never gets the same identity controls, monitoring, or access reviews as everything else.
The result is a system that’s still connected, still holds data, and still accepts logins, but has no one actively deciding who should be allowed in. That’s not a technology failure. It’s an access decision nobody is making.
Where inherited access hides
If you’ve been through an acquisition, four places deserve a hard look.
- Field and contractor accounts. Drivers, technicians, temps, and partners often log in to operational systems that sit outside the corporate identity stack. They frequently use shared or long-lived credentials, and when someone leaves, their account rarely does.
- Service and machine accounts. Integrations, scripts, and scheduled jobs log in without a person behind them. They tend to have broad permissions, passwords that never rotate, and no clear owner once the original engineer is gone.
- Legacy applications still on the network. The old order system or customer portal that “only a few people use” is often reachable far more widely than anyone realizes, because no one has gone back to narrow who can connect to it.
- Devices nobody inventoried. Handhelds, printers, scanners, and IoT gear from the acquired company keep connecting long after the deal closes, often with default settings and no record of who manages them.
None of these show up as a red alert. They show up as normal logins, which is exactly the problem.
Treat access as a day-one integration task
The fix isn’t to migrate everything faster. That rarely happens, and rushing it creates its own risk. The fix is to stop treating access to acquired systems as something you’ll clean up later, and start controlling it from the day the deal closes.
In practice, that means four things.
- Put every login behind your own policy. Whatever the acquired system is, the decision about who reaches it should be made by your access controls, not by whatever the previous owner configured. If a system can’t meet your standard yet, limit who can connect to it until it can.
- Replace passwords where you can. Shared and long-lived passwords are the easiest thing for an attacker to steal and the hardest thing to clean up after. Certificate-based, passwordless authentication removes that target for users and devices alike.
- Check every connection, not just the first one. A login that was fine at 9 a.m. isn’t automatically fine at 2 a.m. from a new location. Access should be re-evaluated continuously, so you can cut it off the moment something changes.
- Give every identity an owner. Every human, device, and service account inherited in a deal should have someone accountable for it, or it should be disabled. If no one will claim it, that tells you something.
None of this requires the acquired system to be modern. It requires the access around it to be.
Questions to ask before your next deal closes
If your company is buying, or still absorbing something it bought, these questions are worth putting to your security and IT leaders this quarter:
- Which systems did we inherit, and which of them are still connected to our network?
- Who can log in to each one today, and how do they authenticate?
- Which accounts belong to people, which to devices, and which to services? Who owns each?
- If one of those systems were breached tomorrow, how quickly could we cut off access to it without taking the business down?
If the answers are fuzzy, you’re not alone. But the deal is done, and so is the inheritance. The only question is whether you decide who gets in, or whether someone else decides for you.
At Portnox, this is the problem we focus on: continuous access control for every identity, human and non-human, across the network and the applications on it, whether you built them or bought them.