Techbanq

Buyer's checklist · industrial security

Know what is connected before anyone assesses your plant

Every OT security engagement begins by discovering what the operator could have known: the assets actually on the network, the paths through the supposedly separate zones, the vendor connections nobody logged, the workstation that could not be restored. Establishing those facts yourselves makes every proposal sharper and every bidder honest.

Process piping and steel structures at an industrial plant

Serving Saudi Arabia from our Riyadh office and our engineering hubs in Lahore and Karachi

Cybersecurity credentials

  • National Cybersecurity Authority, Saudi Arabia

    NCA registration

    Registered cybersecurity service provider

  • ISO

    ISO/IEC 27001

    ISO/IEC 27001 certified

Certifications & registrations

The five passes

The OT self-assessment to run before any outside engagement

All five passes are observational — records, configurations, walking the plant. None injects traffic into a process network or touches a controller, because a self-assessment must never create the incident it exists to prevent.

STAGE 01

Asset honesty — list what is actually connected

Walk the plant with the network diagram and record what the diagram missed: the controllers, HMIs, historians, printers and forgotten laptops actually cabled or on wireless. The gap between the diagram and the floor is the first finding, and it is yours to make. A connected-asset list reconciled against the diagram, gaps marked.

STAGE 02

Segmentation — check the reality behind the zones

Read the firewall rules and switch configurations between the corporate network and the process network, and list every rule that permits traffic across. A boundary with "temporary" any-to-any rules from a project three years ago is a drawing, not a boundary. The actual rule set at each boundary, with every crossing justified or flagged.

STAGE 03

Remote access — count every path in, vendors included

List every way the plant can be reached from outside: staff VPN, vendor support tunnels, cellular modems inside panels, remote desktop on a historian. Ask each equipment vendor what standing access they hold; the answers routinely surprise the operator more than any scan would. A remote-access register with an owner, a purpose and an expiry for each path.

STAGE 04

Recovery — prove an engineering workstation could come back

The engineering workstations hold the project files, controller logic and licensed tools the plant would need after any incident. Establish whether backups exist, where they live, and whether anyone has ever restored one onto clean hardware. An untested backup of the only copy of your control logic is a single point of failure with a green icon. A restore actually performed, or the honest finding that it failed.

STAGE 05

Safety boundary — write down what no one may touch

Name the safety instrumented systems and any equipment whose interruption could endanger people or the process, and record the rule in writing: these are out of scope for any active security work, by anyone, ever. This document precedes every future engagement and goes into every RFP. A written safety-boundary document every future vendor signs against.

The first finding

The OT asset list is usually wrong, and learning that is the point

Industrial networks accrete. A vendor commissions a machine and leaves a modem in the panel; a project connects a historian to the corporate network for one report and the cable stays; an engineer bridges two segments for an afternoon that becomes a decade. None of this was decided by anyone whose job was the network, which is why the official diagram describes an intention, not a plant. Operators who run the walkthrough honestly almost always find the floor differs from the drawing — and each difference is either an asset nobody is protecting or a path nobody is watching.

This is also why the walkthrough must stay observational. IT-style discovery — active scanning, probing, credential testing — can disturb industrial equipment that was never built to receive unexpected traffic, and a self-assessment that trips a production line has answered the wrong question expensively. Records, configurations and walking with a clipboard find most of what matters at zero operational risk; what needs more than that is precisely what a carefully scoped outside engagement is for.

What surfaces

Four gaps the plant walkthrough surfaces most often

Not predictions about your plant — patterns from how industrial estates grow. Finding one is normal; finding none usually means the walkthrough was gentle.

  1. 01

    The vendor tunnel nobody documented

    Standing support access installed at commissioning, still live years later, known to the vendor and to no one on your side. It is found by asking vendors directly, which is why the register pass writes to them.

  2. 02

    The flat network behind a segmented diagram

    Zones exist on paper while the rule base permits broad crossings added for projects and never removed. The diagram passes audits; the rule base is the truth.

  3. 03

    The workstation with no way back

    The only machine holding the controller project files, backed up never or backed up somewhere no one has restored from. The plant's recovery time after an incident is the time to rebuild this machine.

  4. 04

    The dual-homed machine bridging the boundary

    A workstation or historian with one interface in the corporate network and one in the process network, quietly bypassing every control on the official route between them.

Our role

What we assess on the plant floor and the lines we do not cross

  1. 01

    Assessment that starts from your facts

    When the five passes exist, our assessment support starts at interpretation and risk — which gaps carry real exposure, in what order to close them — instead of billing weeks to rediscover your own estate.

  2. 02

    Production-safe method, in writing

    Discovery on process networks is passive, active techniques are confined to what your operations team approves in writing, and your safety-boundary document governs everything we scope.

  3. 03

    Engineering for the gaps that need it

    Segmentation redesign, remote-access consolidation and workstation recovery are engineering work we deliver with your plant team — scheduled around production, not over it.

Safety systems remain governed by your safety engineering processes and their own standards; our work stops at the boundary your document draws, and we would decline a scope that crossed it.

Common questions

What buyers ask

How do we inventory OT assets without touching the process network?+

Combine sources that already exist. The physical walkdown — following cables, opening panels, photographing nameplates — finds the assets no tool sees, including the cellular modem and the forgotten laptop. Switch configurations and their address tables list what has actually communicated, read from the network equipment without sending anything to the industrial devices. Maintenance records, purchase orders and vendor commissioning documents fill in models and firmware. Where deeper visibility is justified later, passive monitoring — observing a mirrored copy of traffic without transmitting — exists precisely for industrial networks, and choosing it is a scoping decision for an engagement, not a prerequisite for the self-assessment. What the self-assessment should never do is run an IT-style active scanner against controllers; equipment that predates the assumption of hostile traffic can fail in ways a scan report does not warn about.

Do vendor support connections belong in our OT remote-access register?+

They usually are the register — in many industrial environments, vendor paths outnumber staff paths and are governed more loosely. Equipment vendors install standing access at commissioning for support and diagnostics, and those tunnels survive staff turnover on both sides until nobody remaining can say what they reach. The register should record, per path: which vendor, into which equipment, by what mechanism, authenticated how, active when, and reviewed by whom. Then apply the discipline that each path exists by a current decision — access that comes up for a support session and closes after it, individual accounts rather than shared logins, and an expiry that forces the renewal conversation. A vendor who says the machine needs permanent unattended access is stating a commercial preference, not an engineering fact, and it can be negotiated like one.

Why do engineering workstations need their own backup answer?+

Because they concentrate a unique kind of value the corporate backup regime rarely covers. An engineering workstation holds the project files and logic for the plant's controllers — often the only current copy, since what runs inside a controller and what was last saved to a file can drift apart across years of small changes. It holds licensed engineering tools, some tied to hardware and painful to reinstate, and it frequently runs an operating system old enough that corporate backup agents were never installed. After an incident — ransomware, hardware failure, an error — this machine is how the plant reprograms, diagnoses and recovers; if it cannot be rebuilt, recovery time is measured in vendor callouts and licence negotiations rather than hours. The self-assessment question is deliberately concrete: has anyone actually restored one onto clean hardware, uploaded a project and connected to a controller? Until that has happened once, the backup is a belief.

Next step

Discuss your OT picture with us

A conversation about what your five passes surfaced — which gaps your own team can close, and which need outside engineering support.