Industrial IoT has a dirty secret: most "successful pilots" never make it to production.
I've lost count of the board meetings where the pipeline looks fantastic — 15 pilots running, big-name logos, great engagement from the customer's innovation team, glowing feedback in the quarterly business review. Then you ask the hard question: how many have converted to paid production deployments?
Silence.
The problem is structural, not commercial
Pilots are funded by innovation budgets — discretionary spend controlled by a small team with a mandate to explore new technology. The innovation team has authority to spend £20k on a proof of concept. They do not have authority to roll out a system that touches operational technology, requires IT infrastructure changes, needs cybersecurity sign-off, and has to integrate with a 15-year-old SCADA system that nobody fully understands anymore.
Production is a completely different sale
Production deployments need operational budgets. They need IT sign-off. They need cybersecurity approval. They need integration with legacy systems that weren't designed to talk to anything built after 2010. They need a business case that survives a CFO's scrutiny — not "this is innovative" but "this will save us £2.3 million per year in unplanned downtime, and here's the data to prove it."
That's a completely different sale, to completely different stakeholders, with completely different timelines. And most IIoT startups aren't set up to make it.
Design for production from day one
The successful startups are the ones who design for production from day one. They think about OTA updates at scale, not just for 10 devices in a pilot but for 10,000 nodes across 50 sites. They think about device management, monitoring, and alerting. They think about what happens when their gateway sits behind a corporate firewall that blocks MQTT. They have an answer for "what's the cybersecurity posture of your device?" that isn't a blank stare.
The question founders need to ask
If you're a founder celebrating pilot wins, ask yourself honestly: is your business model built on pilot fees, or on production revenue? Because those are very different companies, and the transition from one to the other is where most IIoT startups quietly die.
The pilot graveyard is full of good technology that never crossed the chasm.