Key takeaways
- Ownership must be explicit: IT owns shared infrastructure standards, operations owns the process and its risk, and a named OT security lead connects them.
- One change process, two speeds: OT changes follow management of change with safety review and maintenance windows.
- OT patching is risk-based and scheduled. Compensating controls cover what you cannot patch.
- A shared SOC works only if analysts have OT context and a call tree that reaches the plant before anyone isolates anything.
- Budget OT security as named lines (people, monitoring, remote access, segmentation, recovery), not as a leftover from the IT budget.
IT/OT convergence already happened in most plants. The historian talks to the ERP, the MES runs on a virtualized cluster that IT manages, and the OEM connects in over the corporate VPN. What usually has not happened is the governance that should come with it. This article covers who owns what, how changes and patches work across both worlds, how to run a shared SOC, and what to budget.
Why convergence breaks without governance
IT and OT optimize for different things. IT prioritizes confidentiality and keeping systems current. OT prioritizes safety, availability and deterministic behavior. Both are right for their systems. The trouble starts in the gaps:
- IT pushes a Windows update to an HMI and the vendor's software stops working.
- Operations installs a remote access tool to get an OEM connected at 2 a.m. and nobody tells IT.
- The SOC sees ransomware indicators on a SCADA server and has no idea who to call.
- An AI pilot gets cloud access to historian data because nobody owned the decision.
These are ownership failures. Fix them on paper first.
Who owns what: a working RACI
A RACI is the cheapest security control you will ever implement. Here is a starting point for a mid-market manufacturer with central IT and one or more plants.
| Area | IT / CISO | OT security lead | Plant manager | Controls / OT engineering | OEM / integrator |
|---|---|---|---|---|---|
| Security policy (incl. OT annex) | A | R | C | C | I |
| Network infrastructure standards | A/R | C | I | C | I |
| OT network segmentation and firewall rules | C | A | I | R | C |
| OT asset inventory | C | A | I | R | C |
| Changes to control systems (MOC) | C | C | A | R | R |
| OT patching decisions | C | R | A | R | C |
| Remote access approvals | C | A | C | R | I |
| SOC monitoring and triage | A/R | C | I | C | I |
| OT incident response and isolation decisions | C | R | A | R | C |
| AI systems touching OT | C | R | A | R | C |
R = responsible, A = accountable, C = consulted, I = informed. Adapt it, but do not leave a row blank. For AI-specific roles across the business, see roles and RACI for AI governance.
Watch out: Accountability for production risk stays with operations. If IT is accountable for "OT security" but cannot decide when a line stops, every hard decision will stall.
Change management: one process, two speeds
Do not run two disconnected change processes. Run one, with OT-specific gates.
- IT changes that touch OT (firewall rules at the DMZ, domain changes, AV signature policies, backup agents) get an OT impact check before approval. The OT security lead or a controls engineer signs off.
- OT changes (PLC logic, HMI screens, SCADA patches, new devices, AI model updates) follow management of change. That includes a safety review where relevant, test or FAT/SAT evidence, a scheduled maintenance window and a rollback plan.
- Emergency changes have a defined fast path, but are recorded and reviewed within days.
Add one rule that saves a lot of pain: no new connection into or out of an OT zone without a change record. That alone stops most shadow remote access and shadow AI.
Patching: why OT is different
"You can't patch a PLC on a Tuesday" is true, and it is not an excuse to never patch. The differences:
| Factor | IT | OT |
|---|---|---|
| Typical cadence | Monthly or faster | Planned windows, shutdowns, or not at all for some assets |
| Who approves | IT change board | Operations plus vendor qualification |
| Testing | Pilot group | Vendor compatibility list, test bench if you have one |
| Reboot impact | User inconvenience | Possible line stop, scrap, restart risk |
| Support model | Vendor patches OS freely | OEM may void support if OS is patched without approval |
| Fallback | Reimage | Known-good backups of PLC programs, HMI projects and configurations |
A practical OT patch policy triages by exposure and consequence:
- DMZ and anything reachable from IT or the internet: patch quickly, like IT.
- Windows servers and HMIs at Level 2 and 3: patch in planned windows, after the vendor confirms compatibility.
- Controllers and embedded devices: patch during shutdowns, or sooner if a vulnerability is known to be actively exploited and the device is reachable.
- Cannot patch (end-of-life or vendor-locked): document compensating controls. The 2026 secure connectivity principles from the UK NCSC, ASD's ACSC and partners advise treating obsolete products as untrusted and putting a replacement timeline on compensating controls.
Your OT asset inventory needs firmware and OS versions for this to work.
A shared SOC that does not break the plant
Most mid-market manufacturers use an internal IT security team or an MSSP for monitoring. Extending that to OT works if you set it up properly.
- Telemetry: passive OT network monitoring at key switches, plus DMZ firewall and jump host logs. Send it out through the DMZ.
- Context: give the SOC your asset inventory, zone map and a list of "normal but weird" OT behavior (old protocols, broadcast traffic, vendor polling).
- Rules of engagement: analysts never isolate an OT asset or block OT traffic without a call to the plant. Write a call tree with 24/7 contacts for each site.
- Escalation: OT alerts go to the OT security lead and the shift supervisor, not only the IT on-call.
- Exercises: run at least one joint tabletop a year with the SOC, IT and plant staff. See OT incident response.
Plant-floor reality: An MSSP that auto-contains endpoints will happily quarantine an HMI running a batch. Exclude OT assets from automated response until you have tested it.
The OT security lead role
Every manufacturer with more than one line needs a named person who wakes up thinking about OT security. In mid-market firms this is rarely a full-time hire at first.
Typical responsibilities:
- Own the OT asset inventory, zone and conduit model and network drawings.
- Approve OT firewall changes and remote access requests.
- Sit on the change board and AI governance committee for anything touching OT.
- Maintain the OT incident response plan and run exercises.
- Set security requirements for OEMs, integrators and AI vendors.
- Report OT risk to the CISO and plant leadership quarterly.
Where it sits: reporting to operations or engineering with a dotted line to the CISO works well, because it keeps the role close to the process. Reporting only to IT tends to lose plant trust.
Who fills it: often a senior controls engineer with a formal allocation of time (for example 30 to 50 percent), training in IEC 62443 and support from an outside OT security firm for specialist work.
Budget lines
OT security spending disappears when it is buried in IT budgets. Name the lines:
| Budget line | What it covers | Notes |
|---|---|---|
| People | OT security lead time, training, certification | Largest and most important line |
| Visibility | Passive OT monitoring, asset inventory tooling | Often bundled; can start with one site |
| Segmentation | Industrial firewalls, DMZ hardware, switch upgrades | Schedule around shutdowns |
| Remote access | Brokered access platform, MFA, session recording | Replaces scattered vendor tools |
| Recovery | Offline backups of PLC/HMI projects, spare hardware, golden images | Cheap and critical |
| Services | Assessments, penetration testing, incident retainer with OT capability | Use for specialist peaks |
| Lifecycle | Replacement of end-of-life Windows and controllers | Tie to capital planning |
If you are subject to NIS2 as an essential or important entity, management bodies must approve and oversee cybersecurity risk-management measures under Article 20 of Directive (EU) 2022/2555. As of November 2026, national transposition is still uneven across member states, so check your national law. See NIS2 for manufacturers and the ISO 27001 vs IEC 62443 comparison to align IT and OT frameworks.
What good looks like
- A signed RACI covering IT, OT, operations and vendors.
- One change process with OT gates and a rule that every new OT connection has a change record.
- An OT patch policy that triages by exposure and documents compensating controls.
- A SOC with OT telemetry, OT context and a tested call tree.
- A named OT security lead with time and budget.
Effort and cost
Writing the RACI, change gates and patch policy takes a few workshops over a month. The ongoing cost is mainly people: the OT security lead allocation and the time engineers spend on change reviews. Tooling spend can be phased by site.
Common mistakes
- Making IT accountable for OT risk without the authority to schedule downtime.
- Running OT changes outside the change system "because it is faster".
- Treating "cannot patch" as "nothing to do".
- Onboarding a SOC without telling it what normal OT traffic looks like.
- Giving the OT security role to someone with no time allocated.
Monday-morning checklist
- Draft the RACI above for your organization and circulate it to IT, operations and engineering.
- Name an OT security lead, even part-time, and agree their time allocation in writing.
- Add an OT impact check to your change process for anything touching the DMZ or OT zones.
- Exclude OT assets from automated SOC containment until tested.
- Build a one-page OT security budget with the lines above for next year's planning cycle.
This article is general guidance, not legal advice. Confirm regulatory obligations with counsel.
Sources and further reading
- NIST SP 800-82 Rev. 3, Guide to Operational Technology (OT) Security
- ISA/IEC 62443 Series of Standards (ISA)
- Directive (EU) 2022/2555 (NIS2), EUR-Lex
- Secure connectivity principles for operational technology (ASD's ACSC, NCSC-UK and partners, Jan 2026)
This guide is general information for manufacturers, not legal advice. Confirm regulatory obligations with qualified counsel.
FAQ
Frequently asked questions
Should IT or OT own cybersecurity for the plant floor?
Neither alone. A workable split is that IT or the CISO sets security policy and runs shared tooling, operations owns production risk and decides when changes happen, and a named OT security lead owns the OT security program day to day. The plant manager is accountable for the risk on the floor. Write this down in a RACI so nobody assumes the other side has it covered.
How often should OT systems be patched?
There is no single interval. Triage OT patches by exposure and consequence. Patch internet-facing and DMZ systems quickly, patch Windows-based HMIs and servers in planned windows after vendor approval, and patch controllers mainly during shutdowns or when a vulnerability is actively exploited. Where you cannot patch, document compensating controls such as segmentation, application allowlisting and monitoring.
What does an OT security lead do in a mid-market manufacturer?
The OT security lead owns the OT asset inventory, zones and conduits, remote access process, OT incident response plan and vendor security requirements. They sit on the change board, work with the SOC on OT alerts and report OT risk to the CISO and plant leadership. In smaller firms this is often a controls engineer with part of their time formally allocated, plus outside support.
Can a managed SOC monitor OT networks?
Yes, if it receives OT-relevant telemetry and knows what normal looks like. Feed it from a passive OT monitoring tool or firewall logs at the DMZ, give it your asset inventory and zone map, and agree a playbook that says analysts call the plant before taking containment actions in OT. A SOC that auto-isolates an HMI mid-batch can cause the outage it was meant to prevent.
Related guides
AI Governance
AI governance roles and RACI for mid-market manufacturers
Mid-market manufacturers need a small AI governance group drawn from roles they already have, a clear RACI across the AI lifecycle, and a firm split between what plant…
7 min read
OT Security
OT asset inventory: the foundation for OT security and AI governance
An OT asset inventory is a maintained list of every device, system and connection in your control environment, with the attributes you need to make security decisions…
6 min read
OT Security
OT incident response with AI in the mix: playbooks and tabletops
OT incident response starts with safety and keeping the process in a known state, not with wiping machines. Plan for manual operation, include AI failures alongside…
8 min read
Frameworks
ISO 27001 vs IEC 62443: which one a manufacturer needs, and how to run both
ISO/IEC 27001 is a certifiable management system standard for information security across the whole company. IEC 62443 is a series of standards for securing industrial…
7 min read