Like many in the world of industrial technology, I set out on my Industrial IoT journey full of optimism, convinced I was about to revolutionise maintenance, unlock new efficiencies, and win the admiration of my peers.
What followed was, like all the major projects I’ve run… a learning experience.
Lesson 1: Factories Are Already Full of Data. Who Knew?
In my first attempt, I spent months debating sensors, networks, and platforms—only to realise that my factory was already drowning in data. It turns out, SCADA, PLCs, and historians have been around a while. My brilliant IoT deployment was essentially a very expensive way to collect information that we already had.
🔹 Lesson learned: Before adding more sensors, check what data already exists and spend your time and resources harvesting that, before adding expensive new data. It may not be perfect, but it’s probably good enough data.
Lesson 2: Wireless Is Amazing… Until It’s Not
Like a kid with a new toy, I got overexcited about wireless technology. I imagined a cable-free paradise where machines talked seamlessly. Instead, I got signal dropouts, interference, and devices that refused to talk to each other like feuding relatives at a Christmas dinner.
🔹 Lesson learned: Wires are fantastic. They may be old-school, but they work. Wireless is also great but the choice is a tactical one based on the physical environment, not a strategic one.
Lesson 3: F.A.I.L. Means ‘First Attempt In Learning’
I tried, I failed, I tried again. Sometimes I failed better, sometimes I failed worse. I became an expert in acronyms, not because I needed them, but because every new attempt brought fresh complications. LPWAN, MQTT, OPC UA… I had enough to start a new language.
🔹 Lesson learned: Fail quickly, fail cheaply, and fail safely. If you're not failing, you're probably not innovating. By starting small, you’ll soon work out what adds vale in your environment.
Lesson 4: Cybersecurity Matters (and IT Will Catch You Eventually)
One of my more exciting lessons came when I realised that “industrial cyber security” wasn’t just a buzzword. Default passwords, open ports, and mysterious remote access sessions… It was like leaving my front door open with a neon sign saying, "Hack Me." My favourite is still the factory with the etched signs screwed to the panel next to the HMI screens with the admin passwords so they wouldn’t get lost. Fortunately, we never had a breach because I had good support and oversight from the experts.
🔹 Lesson learned: Work with IT and Security. Don’t just work with them, learn from them and become their friends. They will find out what you’re doing eventually, so you may as well invite them in early and get them excited by the art of the possible.
Lesson 5: Keep It Simple, Then Scale
In hindsight, my biggest mistake was trying to solve everything at once myself. I wanted real-time predictive maintenance, seamless integrations, full-stack visibility, and a solution so sophisticated that NASA would be jealous. What I needed was a simple, useful first step, but I didn’t need to build it al myself.
🔹 Lesson learned: Start small, prove value quickly, and expand from there. Embrace the solutions built by others with their mistakes. A lot of the technology is becoming commoditised. Focus on your core business of manufacturing and partner with people whose core business is industrial data.
Lesson 6: Have a Master Plan
Back in as previous life when I was regenerating and expanding breweries, I learnt the critical need to work to a masterplan. We can all shoehorn in a project without a vision for the future but, when you do that, you’ll live with the pain and potentially need to scrap it when you get to the next project. Industrial data is the same. There’s nothing wrong with having a few connected sensors for a specific use case, but you better make sure the payback is lightening quick because you’ll probably be removing them in the near future.
🔹 Lesson learned: Think big and have a vision but then act small in line with Lesson 5. This is the stuff you can’t outsource, but you can get advice from people who’ve been there before.
Where I Ended Up
After plenty of trials, errors, and tea-fuelled troubleshooting, I finally figured it out:
✅ Start with the business problem, not the technology.
✅ Leverage existing data before reinventing the wheel.
✅ Security is a priority, not an afterthought.
✅ Scaling too soon = scaling failure
✅ Develop, and work to, a masterplan so you understand when and why you compromise.
Would I do it all again? Absolutely. But next time, I'd start with a clear masterplan and embrace partner solutions, before starting one, rapid ROI, problem at a time.
What’s Your IoT Lesson?
If you've had your own adventures (or misadventures) in Industrial IoT, drop them in the comments! Let's make the learning curve a little less steep for the next generation of brave IoT pioneers.
#IIoT #IndustrialIoT #LessonsLearned #DigitalTransformation
No comments yet. Be the first to comment!