What small SCADA integrators actually ship when the tag database drifts

By Yves Habchy · August 2026

What: Small industrial automation integrators (2-15 EE) running multi-vendor client projects across Ignition, Kepware, Wonderware, FactoryTalk, and Iconics keep shipping HMIs that reference tags the PLC no longer has. The drift lands at commissioning: faceplates rendering blank, alarms firing on wrong points, trends pulling nothing while the client is on-site for go-live.

Why: Two teams name the same signal two different ways: the PLC engineer for the logic structure, the SCADA engineer for the display. Nothing keeps the two vocabularies aligned across the project, so mismatches surface at commissioning. Horacio Resendiz at Hallam-ICS put the same read plainly earlier this year: "The PLC team names a signal one way because it makes sense inside the logic structure. The SCADA team names the corresponding HMI tag a slightly different way because it makes sense for display purposes."

On top of that, the tools that exist today are single-vendor. Nothing polices what happens across the boundary between one vendor's PLC and another vendor's HMI. Drift stays manual.

Where: Small integrator shops delivering multi-vendor projects across water-wastewater, food and beverage, discrete manufacturing, and building systems.

Who: Integrator project leads or owner-operators at 2-15 EE shops delivering multi-vendor projects.

What I built

scada-tag-audit v0.1 alpha. Reads Ignition tag CSV exports + Rockwell Studio 5000 L5X exports, reconciles cross-vendor references, and produces a self-contained HTML report grouping findings by severity: orphaned HMI bindings and type mismatches at the top (commissioning breakers), naming-convention drift below, informational rows and clean matches collapsed at the bottom.

Runs on the integrator's laptop. No cloud upload. No vendor API dependency. No client data leaves the machine.

Install:

Windows: download scada-tag-audit.exe (13 MB, double-click). Windows SmartScreen warns on first run because the alpha isn't code-signed; click "More info" then "Run anyway".
Python: pip install scada-tag-audit then scada-tag-audit --ignition tags.csv --rockwell routine.L5X

Source + issues: github.com/Yveshby27/scada-tag-audit.

v0.1 scope + honest limitations (README section): Ignition CSV + Rockwell L5X only. UDT member references not resolved (UDT-heavy PlantPAx projects will surface false orphans on UDT members). Alias chains not resolved. Wonderware, Kepware, FactoryTalk, Iconics, OPC UA custom info models on v0.2+ roadmap if the wedge holds.

I could be wrong on which drift shape actually hurts most, or on whether v0.1 scope covers enough of your typical project to be useful. If commissioning-week looks different than I've described, that's what I need to know.

The ask

If you run a small integrator shop (2-15 EE) delivering multi-vendor projects, reach out. I'm not selling anything; I want to know whether the cross-vendor tag drift I've described matches what commissioning-week actually looks like inside your projects. If it doesn't, where it doesn't is exactly what I need.

Also possible: the drift shape I've named doesn't exist at the small-integrator tier, or it exists in a totally different shape than I've described. Say so and I move on.

My LinkedIn and email are below if this is interesting.

Or, if you're not ready to reach out

Drop your email as a signal that this pain is real for you, plus a message if you want to say more. Enough signals and this becomes worth building. Might reach back out if this goes somewhere worth talking about.

More detail

Verbatim quotes and sources from the research.

"Inconsistency in tag structure creates a ripple effect that touches everything: rework, miscommunication, commissioning delays, and long-term maintainability problems."

Horacio Resendiz, Hallam-ICS blog, Before You Build the HMI: Why Tag Planning Makes or Breaks a SCADA Project, July 2026.

Resendiz is naming the same pain from the integrator side that the vendor tools don't solve. The rework, the miscommunication, the commissioning delays are downstream costs an integrator absorbs on their own margin when the drift surfaces at go-live. The fix he prescribes is a living IO-list document. In practice that document either doesn't exist or lives in someone's SharePoint from three projects ago.

"Reusable objects only work reliably when the tag structure feeding them is consistent and predictable."

Same source.

Reusable HMI objects (faceplates for pumps, valves, motors) are how integrators keep project costs down. When the tag structure drifts, the reusable object breaks or shows wrong data. The two things integrators bank on (reusable objects and tag structure) are coupled in a way vendor documentation understates.

Recent integrator-voice discussion of faceplate tag errors during multi-vendor proxy-PLC integration.

PLCtalk.net thread, Proxy PLC communication with SCADA, May 2026.

Different vendor stack from the Hallam-ICS piece but the same shape of pain: the HMI references tags in a way the underlying PLC no longer supports. When two independent sources (a vendor thought-leadership blog and a peer-integrator forum) converge on the same failure mode, the shape is usually real.

LinkedIn DM or yves.habchy@gmail.com.

Other research notes →