Regional security strategy for a standalone CEE infrastructure group

Context

After the separation closed, the infrastructure side of a CEE telecoms group stood up as an independent neutral-host operator. Regional footprint, multi-country, its own board, its own regulators to talk to. Security had been split and rebuilt through the carve-out. Now it had to become a strategy: a function that belonged to the new company, not a shared-services legacy.

Mandate

Design a multi-year regional security strategy for the newly independent infrastructure group, aligned with the parent group's security direction, and drive the execution across every market.

Three ambitions to land. Security as a condition of growth and resilient operation, the defensive baseline. An agile and efficient operating model, fit for a lean post-carve-out company. And an ambition with no internal template: security offered to external customers as a service, a revenue line on top of the internal cost centre.

Role

I was Regional Security Director for CEE, continuing the mandate I had carried through the separation, now inside the standalone infrastructure group. Reporting into the group Chief Security Officer layer. Remit covered every operating company across the region and the regional coordination layer above them.

I designed the strategy myself. I wrote it up. I carried it through the approval chain. And then I drove the execution: standardization, SOC, service portfolio, reorganization. All of it while keeping day-to-day security running.

Approach

Three strategic ambitions, one five-year roadmap. Designed and executed in parallel.

On the baseline, I ran a regional standardization programme. I went capability by capability: monitoring, identity, endpoint, privileged access, data leakage, vulnerability management, firewall management, web application protection, DNS, PKI. For each one I set criticality tiers and drove the markets to the same floor. A mandatory baseline for every operating company; higher tiers where the risk or the revenue case justified it. Delivered as infrastructure-as-code where the target technology allowed, so deployments scaled across countries rather than being repeated per market.

The SOC came up in iterations. The first one was a pilot, to learn on live problems: borrowed people, ad-hoc infrastructure, a minimum set of playbooks. The second formalized it, with dedicated analysts, defined procedures, signed commercials, proper governance. The third automated it, with orchestration, standard playbooks, and scope widened by the efficiency gains. Three iterations over roughly two years. Each one closed a real capability gap and produced enough operational proof to justify the next.

The service side was the outlier. I designed a security service portfolio for external customers: consultancy, enterprise services, SOC services. A shared-delivery model behind it, so one team and one knowledge base served every customer off a multi-tenant platform. Positioning: specialized service provider for the SME and mid-market segments that the big integrators do not fight over. Sales went through the retail-side B2B channels and a few partner agreements. Unusual angle for a tower-and-fibre operator: security as a product to sell, not only a line in the operating budget.

Organization design came alongside all of this, pulling in two directions on purpose. Engineering and O&M decentralized to local security teams, for local adaptation, local language, and presence in the moments that matter. Monitoring and analysis went the other way, centralized, with local analysts kept for language and context. A regional architect pool sitting horizontally across markets. Working groups for the cross-country projects.

I designed each track and drove it to delivery. Nothing handed off.

Deliverables

The regional security strategy itself. One document tying the three ambitions, the enablers, the initiatives, and the five-year roadmap into a coherent plan.
A regional standardization programme with defined capability tiers and a rolling delivery plan per market.
A regional SOC, progressed from pilot to operational service, with commercial annexes signed, SLA and KPI framework formalized, and a dedicated analyst pool in place.
A security service portfolio and commercial framework for external customers, under a shared-delivery model.
A redesigned regional security organization: decentralized engineering, centralized monitoring, a regional architect pool, and formal governance.
A regional security governance baseline aligned with ISO 27001 and with the parent group's security governance.

What made it hard

The company was still finding its feet. The separation had closed, but the commercial model, the contracts, and the operating rhythm of the new entity were all stabilizing at the same time I was trying to commit it to a five-year security strategy. Every decision I asked the business to make was competing with ten other decisions the business had to make.

The service-provider ambition cut against the culture. Tower and fibre operators do not think of themselves as commercial security vendors. Getting the organization to treat security as something to sell was more organizational change than technical design. Pricing, delivery model, sales channel, commercial terms, service management: none of that machinery existed, and all of it had to be built from zero by people who were also doing their day jobs.

Multi-country execution meant the standardization programme had to respect local regulators, local hiring markets, and local language. I kept the target state common and let the delivery path localize. Per-market strategies would have been faster to sell and impossible to run.

Standardization also had a political face the technical plan didn't capture. Each market came with inheritance: technology choices made earlier by people who were no longer there, or, more awkwardly, by people who still were. And real skills and experience the local teams had built up around those choices. Every market also had its own agenda, its own priorities, its own politics. And because security was now a service provider to those markets, the relationship was commercial: local businesses were customers of the regional security function, not subordinates of it. Pushing a standard was never a purely technical call. Stakeholder management became part of the standardization job, not a side effect.

The org redesign asked for two different motions at once. Decentralizing engineering and O&M meant giving local teams more authority. Centralizing monitoring meant taking authority back in the other direction. Both were right for different reasons, and both had to be sold on their own merit to the same people.

And running all of this on top of business as usual was the constant pressure. Security for the network and IT of an infrastructure group cannot pause while the strategy is being built. Every change to monitoring, access, or endpoint controls had to carry the day-to-day through the transition.

What I took from it

After the separation, I expected the hard part to be behind me. It wasn't. Just different. The separation work rewarded fast, right-now decisions. The strategy work rewarded patience and sequence. The two gears do not come naturally together, and switching between them is a skill in itself.

Two things stuck.

One: a multi-year strategy only works if the first year is visibly delivering. I backloaded nothing. Every one of the three ambitions had something real standing by the end of year one: the standardization baseline moving, the SOC taking live cases, the first external service contracts in place. Strategy that only shows up in year three does not survive the boards and budget cycles that happen in year two.

Two: the service-provider ambition was the test. It was the one most likely to fail, the one with no internal template, the one the organization instinctively resisted. Getting it to actually work, with real customers, real revenue, and real SLA, taught me more about what it takes to change an organization than the other two ambitions combined. Security as a cost line is easy to run. Security as a product forces every other capability to sharpen up.

And the residue. Running a multi-year strategy inside a company still stabilizing its own model taught me how to keep a long horizon visible while staying useful on short-horizon problems. A different muscle from the programme-delivery one I built through the separation, and both of them shape how I think about this kind of work now.

Sources (public record on the group and its post-separation operation):