Let’s get this out there: naming conventions are not glamorous. But in every serious IIoT project I’ve lived through—across oil & gas, pharma, automotive, and beyond—they’re the difference between a system that scales and one that becomes an unmanageable mess. I’ve seen both sides, and if you want to avoid the pain, you need to treat naming like a first-class citizen from day one.
Why Naming Conventions Matter (More Than You Think)
When you start out on a small site, it’s easy to get away with shortcuts. You connect a few PLCs, maybe a historian, and you just name things whatever feels right in the moment—“Line1_Temp,” “MixerSpeed,” or even “Tag123.” It works… until you try to roll out to a second site, or you need to onboard 10,000 tags instead of 100. Suddenly, nobody can find anything, and you’re spending more time mapping and cleaning up than actually delivering value.
I’ve lived through this “tag soup” scenario. One project, years ago, we onboarded tags from four different plants, each with its own conventions. The result? Dashboards that made no sense, analytics that broke, and a team that spent months just figuring out what data meant. That’s when we started taking naming conventions seriously.
What Makes a Good IIoT Naming Convention?
A solid naming convention is more than just a pattern. It’s a contract between everyone who builds, maintains, or uses the system. Here’s what I’ve seen work in real life:
- Hierarchical Structure: Use a hierarchy that mirrors your plant or enterprise structure. Start with company, site, area, line, equipment, and then the tag or parameter. For example:
Company/Site/Area/Line/Equipment/Tag. This isn’t just tidy—it’s essential for both humans and machines to navigate the data. - Consistency Across Sites: If you’re running a multi-site deployment, you need the same rules everywhere. I’ve seen sites try to “localize” naming, and it always leads to chaos when you try to aggregate or compare data.
- Alignment with Standards: Some of the best projects I’ve been part of use the ISA-95 equipment model as a backbone. This standard gives you a common language for site, area, line, and equipment, which makes integration and reporting so much easier.
- Central Governance: Don’t let every team invent their own rules. We set up a central repository—sometimes just a table in SQL—that defines the naming standard. All new tags are validated against this, and deviations are flagged automatically. This isn’t just theory; we actually built scripts in the edge solution to do this validation and logging.
Real-World Example: Unified Namespace (UNS)
The Unified Namespace (UNS) is a big deal in modern IIoT. Think of it as the ultimate naming convention: not just for tags, but for the entire data structure across your business. A UNS is a semantic hierarchy that organizes all your devices, data points, and even events, making them discoverable and usable by any system that needs them.
I’ve implemented a UNS using MQTT as the backbone, with a hierarchical topic structure that followed our naming standard. For example:
/Company/Site/Area/Line/Equipment/Parameter
This allowed us to plug in new data sources or consumers—analytics, dashboards, even AI/ML models—without reworking the whole integration. It also meant that when we needed to onboard a new site, most of the work was already done. The hardest part? Getting everyone to agree on the hierarchy and stick to it. But once we did, everything else became easier: automated tag onboarding, bulk data ingestion, and even troubleshooting were all simplified.
Enforcing Naming Conventions: Automation Is Your Friend
It’s one thing to define a standard; it’s another to enforce it. In practice, we automated as much as possible:
- Central Repository: We stored the naming rules in a central, version-controlled table.
- Automated Validation: Every time a new tag was ingested (whether from a PLC, historian, or lab device), we checked it against the standard. If it didn’t match, it was flagged in logs for review.
- Bulk Updates: When the standard changed (and it will), we could update metadata centrally and push changes out across all sites and systems.
- Edge and Cloud Contextualization: Sometimes, batch context or process steps needed to be associated with tags. We handled this both at the edge (close to the machine) and in the cloud, always using the same naming backbone.
The Human Side: Why This Is Harder Than It Looks
Honestly, the technical part is the easy bit. The hard part is getting buy-in from everyone—automation engineers, IT, data scientists, and plant operators. People have their own ways of doing things, and nobody likes being told to change. But if you want a system that scales, you need to invest the time up front to define, document, and govern your naming standards.
One lesson I learned the hard way: don’t try to do this in isolation. Bring the right people together early, get their input, and make sure the standard is practical for everyone. Otherwise, you’ll end up with workarounds and exceptions that undermine the whole effort.
Honest Opinion: Naming Is Boring—Until You Don’t Have It
Here’s my unpopular opinion: nobody cares about naming conventions… until the system breaks or the project needs to scale. Then suddenly it’s urgent. If you want to avoid firefighting later, make naming a core part of your IIoT strategy from day one. It’s not flashy, but it’s the foundation for everything else—data quality, analytics, compliance, and even cybersecurity.
So, next time someone tells you naming conventions are just paperwork, ask them how they plan to run a global IIoT platform with 100,000+ tags, dozens of sites, and multiple teams. If they don’t have a good answer, you know what’s coming.
Final Takeaway
If you want your IIoT platform to be scalable, maintainable, and useful—not just today, but five years from now—put naming conventions at the center. Automate enforcement, align with standards like ISA-95, use a Unified Namespace, and make it everyone’s job to get it right. It’s not exciting, but it’s the backbone of everything that matters in smart manufacturing.
And if you skip it? Well, I’ve seen that movie too. Trust me, you don’t want to be in the sequel.

Leave a Comment