Many organisations invest in technology without first assessing their operational challenges. We begin by defining the operating model, designing the architecture, and building the security operations capability. We then transition the SOC to your team for ongoing management.
A SOC is shaped by key decisions made before selecting tools: what to defend, what can be monitored, who oversees it, how threats are detected, and how to respond. Microsoft Sentinel and Defender XDR form the technology foundation, while people, processes, and desired security outcomes drive the SOC’s effectiveness.
Every organisation starts from a different position. You may be building a SOC from scratch, have security tooling but no operating model, or already have a SOC that needs to mature or transform. We start with what you have, identify what is missing, and define the capability you need.
A 30-minute call is enough to scope what you need — remotely or face to face in Kyiv.
starting position
Three situations, three different engagements. The first call establishes which one you are in — and occasionally that it is not the one you thought.
Microsoft Sentinel, Defender XDR or other security technologies may already be deployed, but the operational capability around them is incomplete. We turn the existing technology into a structured SOC capability.
For organisations with an existing security or IT team, we define the SOC operating model, processes and detection coverage around the team already in place.
For organisations establishing a new SOC, we define the operating model, roles and processes, and build the in-house capability to run it.
scope
scope
What the SOC is for, what it covers, which hours, and whether in-house, hybrid or outsourced is the right shape for your organisation. Including the headcount arithmetic, honestly done.
A prioritised backlog of what you need to detect, informed by the threats that apply to your sector and estate, with coverage mapped so the gaps are visible rather than assumed.
Workspace design, connector selection, data tiering and retention decided against both cost and evidence requirements — before ingestion starts rather than after the first invoice.
Analytics rules built and tuned for your environment, with a development lifecycle your team can continue: build, test, tune, deploy, review. Version-controlled, not clicked together in a portal.
Runbooks, severity classification, escalation paths and shift handover, written to be used rather than to satisfy an auditor — though they do that as well.
Role definitions and a skills matrix drawn from the operating model, so you hire against what the work actually requires. Where you are recruiting in parallel we write the job descriptions, sit on technical assessment, and build an onboarding path. Where you already have people, we assess the team you have against the roles you need — including where the honest answer is a training gap rather than a hiring one.
What gets measured, what gets reported to the board, and the improvement loop that keeps the capability from decaying after we leave.
engagement
Target operating model, coverage hours, scope and success criteria agreed with your leadership before any technical work starts. You receive an operating model document and a realistic view of the resourcing it implies.
Sentinel architecture, use case catalogue, data source plan with cost model, and the process framework. You receive the design pack and the detection backlog.
Workspace deployed, connectors onboarded in priority order, detections written and tuned, automation where it removes repetitive work rather than for its own sake.
Your analysts work alongside us on live triage before we step back. Runbooks are tested by the people who will use them, not handed over as documents.
A validation exercise against the detections we built, plus a maturity assessment you can repeat annually to show the capability is improving rather than drifting.
boundaries
We build the capability and enable your team. We do not monitor your estate, take alerts, or sit on your rota. If what you need is an operated service, this is not it — and we will say so on the first call.
No staff augmentation and no bodies on a shift pattern. We design the roles and help you assess candidates; hiring and employing them is yours.
No licence resale, no commission on your Microsoft spend. Part of the Define stage is telling you honestly if the capability you need does not justify what you were planning to buy.
faq
For business-hours coverage, two is the practical minimum — one person is a single point of failure the first time they take leave. Genuine round-the-clock coverage needs roughly five to six full-time analysts once you account for leave, sickness, training and handover, which is why it is far more expensive than most boards expect. We will do this arithmetic with you during Define, before you commit to a coverage model. It is also the reason we do not offer round-the-clock monitoring ourselves: we would rather tell you what it costs than sell you something thinner and call it 24/7.
It depends on how much of your risk is specific to your environment, whether you can recruit and retain analysts in your market, and what your regulatory position requires you to demonstrate. There are good reasons for both, and a hybrid — in-house detection engineering with outsourced overnight triage — is often the right answer for a mid-market organisation. We will give you our view with the reasoning, not a recommendation that happens to match what we sell.
Yes, and it is a common sequence — but the operating model has to come first, because it is what determines which roles you are hiring for. Organisations that recruit before defining the model routinely hire the wrong shape of person: three generalists when they needed a detection engineer and two triage analysts, or a senior lead for a scope that did not warrant one. We pace the build against your hiring so the capability and the team arrive together.
Often more so. A team without an operating model is working hard without a shared definition of what good looks like — which is demoralising for them and invisible to your board. The work here is less about building from nothing and more about giving structure to effort that is already being spent. We involve the existing team in writing the runbooks rather than handing them a document. Process imposed from outside tends to be quietly ignored; process the team helped write tends to survive.
Not yet. Sentinel is where a SOC is implemented. Without an operating model, agreed severity definitions, owned triage and a detection backlog, what you have is a log store with alerting attached. The good news is that the expensive part is already bought.
Substantially. NIS2 requires incident handling and monitoring capability; ISO/IEC 27001 requires monitoring activities and incident management processes. A SOC build produces the process documentation and technical evidence both frameworks ask for — though the wider governance requirements sit within our compliance work.
Contracts are signed with Censara Security Ltd, registered in England and Wales (company no. 17150931). Invoices are issued in EUR, USD or GBP, agreed per engagement, and paid by international bank transfer. For clients in the EU, Ukraine and other countries outside the UK, invoices are issued without UK VAT: local VAT is accounted for by the client, and EU business clients apply the reverse charge and provide their VAT number. UK clients are charged UK VAT at the standard rate. Payment terms are agreed per engagement.
Whether you are starting from nothing or suspecting that what you built is not working, a scoping call will tell you which of the two you are actually dealing with.