If there’s one thing I’ve learned from two decades in the trenches of manufacturing IT and IIoT, it’s that a good strategy isn’t about chasing the latest technology or ticking off a checklist. It’s about making real improvements—connecting machines, data, and people in a way that actually makes life easier on the plant floor and delivers value to the business. So, let me break down what I’ve seen work (and not work) when designing a new Industrial IoT (IIoT) strategy, using real stories and lessons from the field.
Start With the Business Need, Not the Tech
It’s tempting to start with the shiny stuff—edge devices, AI, cloud analytics. But every successful IIoT program I’ve been part of started with a clear business problem. For example, at one large manufacturing site, we kicked off our IIoT journey not because we wanted “IoT,” but because we needed to reduce unplanned downtime. The tech only came after we mapped out the pain points with operations, quality, and supply chain.
If you skip this, you risk ending up with a patchwork of disconnected solutions—a historian here, a dashboard there, a “pilot” that never goes anywhere. I’ve seen this happen more times than I can count, and it always leads to headaches and wasted money down the road.
Design for Integration: Legacy & Future
No plant starts from a clean slate. You’ll have old PLCs, proprietary SCADA, maybe a historian from the early 2000s, mixed with newer OPC UA and MQTT-enabled devices. The trick is to design an architecture that brings all these worlds together.
Standardize Early: Naming, Hierarchies, and Governance
Here’s an unpopular opinion: Standardization is more important than the technology itself. I’ve seen projects fail because every site named their assets differently or used their own data models. We solved this by enforcing a unified plant hierarchy and strict naming conventions across all sites. It wasn’t glamorous, but it made everything else—analytics, reporting, even AI—possible at scale.
The Unified Namespace (UNS) approach is key here. Think of it as a real-time, structured “directory” (often modeled after ISA-95) that organizes all your data, no matter where it comes from. It breaks down silos and lets both OT and IT speak the same language.
Plan for Security and Compliance From Day One
This isn’t just about firewalls and passwords. In regulated industries you need end-to-end encryption, role-based access, audit trails, and compliance with standards like GxP and FDA 21 CFR Part 11. We follow a strict zone model, segmenting networks and controlling data flow between OT and IT.
Embrace Hybrid—Edge and Cloud Together
A pure cloud or pure edge approach rarely works in real plants. We’ve had the best results with hybrid models: edge nodes for real-time processing, buffering, and local failover (so you keep running even if the WAN drops), and cloud for scalable storage, analytics, and machine learning.
Think About Scalability: Not Just in Tech, But in People and Process
It’s easy to get a pilot running at one site. Scaling to a dozen or more is where things get tricky. You need consistent governance, clear roles, and a support model that doesn’t rely on a few “heroes” who know the magic incantations. In one global rollout, we found that training and standardized templates were as important as the platform itself.
Avoid the “One Problem = One Solution” Trap
This is a big one. Over the years, I’ve seen organizations bolt on new tools every time a new requirement pops up—track and trace, temperature monitoring, digital thread, etc. The result? A spaghetti mess of point solutions that don’t talk to each other. The best strategies are built around a unified reference architecture, even if it means saying “no” to some requests or pushing back on vendors who promise quick fixes.
Get OT and IT on the Same Page
This is more about people than technology. The lines between OT (operations) and IT (information technology) are blurring, especially as sensors and data start flowing everywhere. But turf wars and lack of trust can kill an IIoT project before it starts. We had to run joint workshops, define clear ownership for device management, and build cross-functional teams to make progress. It’s not easy, but it’s essential.
Don’t Underestimate Change Management
IIoT isn’t just a technical upgrade—it changes how people work, how decisions get made, and how problems are solved. In my experience, the projects that succeed are the ones that invest in training, communication, and hands-on support. If you just “throw it over the wall” to the plant and hope for the best, you’ll end up with expensive shelfware.
Prepare for Challenges: And Learn From Failure
Let’s be real: most IIoT projects hit bumps. Common pitfalls include unclear business goals, lack of integration with legacy systems, security gaps, and pilots that never scale past one site. I’ve seen projects stall because of firewall issues, vendor lock-in, or simply running out of steam when the “heroes” move on. The trick is to treat these as learning opportunities—iterate, document, and keep moving forward.
Honest Opinion
Here’s my unpopular take: The hardest part of IIoT isn’t the technology—it’s getting people to agree on a common approach, stick with it, and resist the urge to reinvent the wheel every time. If you can get the organization aligned, the tech will follow. But if you don’t, no amount of cloud, AI, or digital twins will save you.
Real-World Tools
For those who like specifics, here are some examples of the tools:
- Protocols:
- OPC UA for legacy and modern equipment
- MQTT with Sparkplug B for scalable or Kafka structured messaging.
- Platforms:
- Kepware, HighByte, Ignition, SiteWise IoT, Azure AIO Edge at the plant.
- AWS SiteWise, Azure AIO, Snowflake in the cloud.
- Integration:
- Unified Namespace for real-time, contextualized data access.
Final Thought
Designing a new IIoT strategy isn’t about picking the hottest new tech or copying what the “big guys” are doing. It’s about understanding your own business, building on what you have, and making deliberate, sometimes unglamorous choices that set you up for long-term success. If you keep these points in mind—and learn from your own scars along the way—you’ll be ahead of the game.
Leave a Comment