INDEPENDENT SECURITY PERSPECTIVES & ADVISORYBASED IN NORTH CAROLINA · WORKING EVERYWHERE
← All field notesCASE STUDY / M&A / SECURITY LEADERSHIP

Three Deals, No Playbook

What M&A security integration actually looks like, across three deals that had almost nothing else in common.

There's no shortage of M&A security frameworks out there. This isn't one. This is what actually happened across three deals that had almost nothing else in common: a tiny brand acquisition early in my career, a large, well-resourced enterprise acquisition where I worked directly with the acquiring company's CISO, and a merger between peers, where the final calls weren't mine to make.

Where I started, and why it stopped working

I used to run a pretty rigid model. Thirty, sixty, ninety days, and I wanted to see specific things land on schedule: EDR installed, cloud logs shipping, end user awareness training rolled out. Clean, measurable, satisfying to plan.

Three deals later, I don't run that model anymore. You have to be fluid and bend with the business, and that's genuinely frustrating for a security practitioner, we're trained to want the checklist, not the negotiation. Sometimes the team you're integrating is up against deadlines that have nothing to do with you, and they can't install EDR this week no matter how much you want them to. That's not a failure. That's a risk decision, and it belongs to leadership, not to you alone: do we push, or do we accept the exposure for now and revisit it later.

The first move is always a conversation, not a scan

Before anything technical happens, there's a formal introduction, and the framing matters more than people think. This is a business decision, not a security decision, and treating it like the second one from minute one is how you lose the room before you've said anything useful.

After that comes a rapid assessment, and how deep you're allowed to go depends entirely on the relationship you've established, some companies let you dig, some don't. Either way, you're looking at what's visible from the outside: compliance frameworks they claim to follow, SOC 2 reports and other external assessments, an external vulnerability scan, your own version of OSINT on the organization and its key players. You can even ask what they spend on security relative to the rest of the business, which tells you a lot about whether a real program exists or whether security has been an afterthought.

That whole picture is still just a picture. With one acquisition, the target's own lawyers told us their security was solid, good controls, nothing to worry about. We had nothing to contradict that going in, so we took it as a starting point, not a fact. Once we actually had access, the reality was less tidy than the packet suggested, in ways nobody puts in a due diligence package. You only find that once you're inside.

Trust, but verify, isn't a slogan. It's the actual operating model.

That wasn't a one-off. In every deal, what gets said in the room and what's actually running in the environment turned out to be two different documents.

Prioritization follows risk, compliance, and whatever the business is actually up against, not a template

On a merger, the instinct is often to start with visibility: bring the other side's telemetry onto what you've already built instead of standing up something new. Straightforward, in theory.

Regulatory deadlines can reorder a plan like that overnight. The EU's Cyber Resilience Act is a good example: it entered into force on December 10, 2024, its vulnerability and incident reporting obligations apply from September 11, 2026, and its main obligations follow on December 11, 2027. For an organization it covers, building a compliant response and notification process for the reporting obligations can jump to the front of the line, ahead of visibility work you'd already planned, because a regulatory deadline doesn't negotiate the way an internal roadmap does. Only after that lands do you go back to the original plan: finding where your processes overlap, and where it makes sense to bring them onto what you've already built instead of running two of everything in parallel.

There's also a slower, messier layer running underneath all of this: who reports to whom, whose priorities win when they conflict. That's not something security resolves. That's a conversation for the C-suite, sometimes the board, about how the deal itself is meant to work.

Getting something prioritized when it isn't your call

In a merger between peers, you often don't own the final decision, even when your team is arguably the most affected by how it plays out. Getting anything prioritized under those conditions comes down to two things: real relationship-building, and laying out actual tradeoffs instead of dictating a decision.

With visibility, the pitch isn't "use our tool." It's "here's the upside: strong visibility, and we already know how to run it well, so you're not starting from zero." With incident response, the conversation is closer to: here's what we do, does it make sense to you, are you properly staffed for something like this, would folding into what we already have actually help you. Then, honestly, what does staying fragmented cost both of us: duplicated work, no shared knowledge, two cultures that never actually merge. Laying out that contrast tends to do more work than any directive ever would.

A peer relationship is slower than an acquisition, and that's not a complaint

With an acquisition, there's a clearer hierarchy, the acquiring company sets more of the terms. A merger between peers is genuinely different: more give and take, less directing, and no shared history to fall back on for reading each other's behavioral norms. Culture is a variable you have to actually account for, the same way you'd account for a technical constraint.

It's harder. I won't pretend otherwise. It takes more conversations just to clarify what you each mean by the same words. But you go in saying you're there to solve a shared problem together, us against the problem, not us against each other, and you stay honest about what you don't yet know. Building trust depends on that.

Why 70% today can be the right answer, not a compromise

The instinct a lot of security people carry is binary: compliant or not, control in place or not. What three deals have actually taught me is that getting to something like 70 percent of a control baseline early, with the rest landing over the following year, can be exactly the right target, rather than insisting on 100 percent immediately and blowing up the relationship to get there. That number is illustrative, not a measurement from any particular organization. The point is sequencing by risk.

That's risk-based sequencing, and it's different from regulatory compliance. A legally required control or a binding deadline isn't up for negotiation, and nothing here is an argument for treating one as optional. The flexibility applies to everything else. A typical example: rolling EDR out to laptops and desktops is an easy, near-term conversation. Rolling it onto production servers is a different conversation entirely. Service owners are genuinely concerned it'll slow systems down, break databases, demand a lot of tuning, eat resources they don't have slack for, the usual worry about how heavy an agent actually is. That's not stonewalling. It's a legitimate concern from people who understand their own systems, and pushing it through by mandate rather than working through it would cost more trust than the timeline is worth. So it waits, deliberately, while you learn how their environment actually behaves.

In these situations, most of what I get done comes from influence, and influence comes from trust. I can't mandate anything in someone else's environment, so what I have is whether people believe I'll be straight with them and that what I'm asking for is worth their time.

Where the patience actually runs out

Everything above assumes you have room to negotiate the timeline. Sometimes you don't. If a KEV, a known exploited vulnerability, turns up in the acquired or merging team's environment, that's not a negotiation anymore. It needs prompt assessment and risk-based remediation: is it actually exposed, is it being exploited, what's the fix or the compensating control. It isn't automatically an incident, but it should trigger an urgent, risk-based response until you understand the exposure. The same is true for a regulatory deadline like the CRA's reporting obligations: it doesn't move because your integration roadmap would prefer it to.

A few things function as close to non-negotiable, even inside a trust-first approach: multi-factor authentication (and increasingly I'd push for something stronger than SMS, a real passkey or hardware key), some baseline of telemetry and logging so you're not flying blind, and eventually a real vulnerability management program and actual security awareness training if neither exists yet. None of those are hills to die on today necessarily, but they don't get to wait indefinitely either.

Plans, like blueprints, have to change as you learn more about the house you're actually renovating. Not every room needs a vault door. Sometimes a lock that actually works is exactly right, for now.

MORE FROM THE FIELD

Keep reading.