Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

How Do OT and IT Teams Work Together?

OT and IT teams have different objectives, change processes, and vocabularies, and most industrial security failures trace back to that gap rather than to any missing technology.

Illustration for the cybersecurity term OT and IT Collaboration
Home / Blog / Cybersecurity Terms / How Do OT and IT Teams Work Together?
17 November 2026

How Do OT and IT Teams Work Together?

How Do OT and IT Teams Work Together?

Ask an IT security manager what an asset is and the answer involves a server, a laptop, or an identity. Ask a plant engineer and the answer is a compressor. Both are right, both are using their industry’s standard vocabulary, and the gap between those two answers is where most industrial security programmes get stuck.

The disagreement is not really about words. It is about what each group is accountable for. IT is measured on confidentiality, data protection, and service delivery. Engineering is measured on availability, safety, product quality, and regulatory compliance for the process. When those objectives conflict, as they do over patching, scanning, and remote access, neither team is being unreasonable.

Where the Friction Actually Shows Up

The Operating Model That Works

  1. Put the system owner in engineering. The named owner of the control system should be an engineer, advised by security, not a security manager issuing instructions into a plant they have never walked. Accountability has to sit where the consequence lands.
  2. One asset register. Not an IT inventory and a maintenance management system that disagree. Reconciling them is tedious and is usually the single most valuable early piece of work.
  3. A joint forum with teeth. A standing group including engineering, IT, security, health and safety, and procurement, that approves architecture changes and accepts risk in writing. Monthly, with a chair who can decide.
  4. Pre-authorised emergency actions. Decide in advance, in daylight, who may disconnect what and under which conditions, and write it into the shift manager’s instructions. Decisions made during an incident are worse decisions.
  5. Risk-based patching. A vendor-validated patch stream, batched into planned outages, with compensating controls documented for the interval: allow-listing, segmentation, and monitoring. Engineering accepts this far more readily than a demand for a thirty-day cycle.
  6. Security in engineering specifications. Procurement is the cheapest point of leverage. Requirements drawn from IEC 62443-4-1 and 4-2, a software bill of materials, a stated support lifetime, and remote access through the client’s own gateway cost nothing at tender and are impossible to retrofit.
  7. Monitoring that reaches the SOC. Passive OT anomaly detection feeding the existing security operations centre, with analysts who have been given enough process context to tell a commissioning activity from an intrusion, and an engineering contact on the escalation path.

Make the Exchange Go Both Ways

Programmes that treat this as educating engineers about phishing fail. The traffic has to run in both directions. IT and security staff who will work on a plant should complete the site safety induction, understand permit-to-work, and walk the process with an engineer before they propose anything. Engineers benefit from understanding credential reuse, how ransomware actually arrives, and why the laptop that moves between the office and the control room is a genuine problem.

Secondments work better than training. A security analyst who spends a fortnight in a control room learns more about alarm floods and operator workload than any course will teach, and comes back able to write detection rules that engineers will not immediately disable.

Two Failure Modes to Watch

The first is IT pushing standard tooling into the plant without consultation: an agent deployed by policy to an operator station, a scan that takes down a line, a password policy that locks out a console during an upset. One of these sets the relationship back a year.

The second is engineering solving an operational problem quietly: a 4G router installed so a vendor can support a machine, a USB drive used to move recipes, a laptop that lives on both networks. Neither is malicious; both are rational responses to a process that did not offer a legitimate alternative. The answer is to provide the legitimate alternative, which usually means a monitored remote access gateway and a sanctioned file transfer path, as described under OT security.

Exercises are the fastest way to find both. A scenario run with the control room, the SOC, and the incident management team in the same room surfaces the authority gaps, the communication assumptions, and the undocumented connections in an afternoon.

At Privia Security, we run joint OT and IT scenarios with operational teams through our Cyber Exercise Services.

FAQ – Frequently Asked Questions

Should OT Security Report to the CISO or to Engineering?

The most durable arrangement gives the system owner role to engineering, with the security function setting standards, providing monitoring, and holding the assurance view. A purely IT-owned model produces requirements that plants quietly ignore; a purely engineering-owned model loses the security expertise. What matters most is that one named person is accountable and that the reporting line is written down.

How Should Patching Be Handled in OT?

Risk-based and vendor-validated, batched into planned outages rather than applied on an enterprise cycle. Between windows, compensating controls carry the risk: application allow-listing, network segmentation, restricted removable media, and monitoring. Document which vulnerabilities remain unpatched and why, because that record is what a regulator and an insurer will ask for.

Why Does Vulnerability Scanning Cause Problems in Plants?

Many field devices have minimal network stacks and respond badly to traffic patterns that modern servers ignore, with documented cases of controllers halting under a routine scan. In a plant that means lost production or a safety event. Passive monitoring on a tap or span port gives most of the same inventory and vulnerability information without touching the devices.

What Is the First Thing to Fix?

A shared asset register, followed by a documented decision about who may disconnect the plant from the corporate network and under what circumstances. Neither requires budget approval, both surface the disagreements that would otherwise appear during an incident, and almost everything else in the programme depends on knowing what you have and who decides.

Related Terms

You May Be Interested In These

What Is TruffleHog?

Cybersecurity Terms

17 November 2026

What Is TruffleHog?

Read More
What Is a Backdoor?

Cybersecurity Terms

17 November 2026

What Is a Backdoor?

Read More
What Is an Infostealer?

Cybersecurity Terms

17 November 2026

What Is an Infostealer?

Read More