For the Kubernetes management platform, visit portainer.io · For AI solutions, visit portainer.ai
One architecture for 13 offshore rigs: read how Seadrill built it with Portainer →
Blog

Securing OT When You Don’t Have the Specialists

Many OT security assessments recommend bringing in dedicated OT security expertise. For most plants, that hire is hard to fund, slow to fill, or both.

Meanwhile, your existing team still has to do the work.

The gap is about skills more than seats

The 2026 SANS | GIAC Cybersecurity Workforce Research Report found 60% of organizations said skills gaps were a bigger problem than staffing shortages, up from 52% the year before. 27% reported breaches directly tied to workforce capability gaps, and 95% said regulatory directives are now affecting their hiring.

OT security is a narrower specialty that needs people who understand control systems and security, and plants compete with utilities, consultancies, and vendors for the same limited talent pool. For most manufacturers, waiting for the right hire isn’t a plan.

One routine change, done the usual way

Take a change that happens all the time. The MQTT broker your line data flows through is moving to a new server. Twelve edge gateways across three sites need the new address.

Today, that often means an engineer logs into each gateway remotely, usually with the same admin account everyone uses, and edits the configuration. One gateway at the third site is offline that afternoon, so it gets skipped with a mental note to circle back. A contractor at the second site changed a timeout on two of the gateways last month, and nobody knew. The work is finished by the end of the day, and the only record is an email saying it’s done.

Six weeks later, when the engineer who made the change is on vacation, data from one line stops arriving. Whoever picks it up has to work out from scratch what’s different about that gateway.

Nobody on that team was careless. The process asked for expert attention at every step and gave nothing back when someone missed one. That’s the real exposure for a generalist team when the process leans on memory.

The same routine change with guardrails

Using that same MQTT server example, when the gateway configuration lives in a Git repository, an engineer changes the broker address in one file, and a colleague reviews it before it merges.

If Git is new to your team, our Git for Non-Developers primer covers what you need.

Portainer watches that repository. When the change is committed, it rolls the update out to the edge group that holds all twelve gateways. The gateway that was offline gets it automatically when it reconnects, so nobody has to remember to go back. Per-device status shows all twelve on the new version, so nobody has to log into each one to check.

The contractor’s earlier timeout change would never have happened outside that process because his access is scoped to his site’s devices, and changes go through the repository, not straight to a gateway. Roles only help if they’re reviewed, which we cover in Why Role-Based Access Control Isn’t Enough for OT Security. And if the new broker misbehaves in the middle of the night, the technician on shift rolls back to the last working version from one screen, the same way at every site.

The gateways never needed an inbound firewall rule for any of this. Each one runs a small agent that connects outbound to Portainer, so no remote-access exceptions need to be opened at each site, justified, and then forgotten about.

The work is the same, and it took about as long. What changed is that the secure path was the easy one, and going around it would have taken more effort than following it.

Evidence you didn’t have to write

Six weeks later, the question “what changed on that gateway?” has a documented answer: the commit, who approved it, when it deployed, and which version each device is running. NIS2 and IEC 62443 increasingly expect that record, and it exists because the work was done that way, not because someone sat down to document it afterward. Our post on OT patch management covers why workarounds alone can’t produce that kind of evidence.

Keep the specialists for the hard calls

None of this replaces security expertise. Network segmentation design, risk assessment, and incident response still need coverage, whether that’s a consultant, an integrator, or a partner. What guardrails change is where that scarce expertise goes. When routine changes take care of themselves, you can spend a specialist’s day on the decisions that need one, not on checking twelve gateways by hand.

If you’re not sure which of your routine changes still depend on someone’s memory, the OT Security Readiness Scorecard is a quick way to find out.

More blog All resources Get a demo