Why iPaaS Platforms Fail at OT Integration in Manufacturing
Why the Integration Platforms You’ve Heard of Won’t Solve Your OT Problem
You’ve probably been through at least one of these conversations. You bring in a vendor demo for Boomi, or MuleSoft, or Workato. The platform looks capable. The connector library is enormous. You ask about connecting your shop floor devices to your ERP.
The answer, delivered politely and with a lot of qualifying language, is that device connectivity is out of scope. They’ll point you toward an integration consulting partner who can build something custom. Or suggest you look at a middleware layer first. The conversation moves on.
What they don’t tell you is that this isn’t a capability gap. It’s a business model decision. And once you understand why they made it, you’ll stop expecting them to change.
The iPaaS market hit $12.8 billion in 2025 and is growing at over 33% annually.¹ None of those companies have meaningfully addressed OT device integration.
The economics of “build once, sell thousands”
The entire iPaaS business model is built on a single economic insight: solve an integration problem once, package it as a pre-built connector, and sell it to as many customers as possible with minimal incremental cost.
A Salesforce-to-NetSuite connector takes a team a few months to build. Once it’s live, it sells to 10,000 customers. The same connector. The same code. The same support documentation. The margin on that is extraordinary.
OT device integration doesn’t work that way. Industrial devices communicate through a landscape of protocols that vary by manufacturer, product generation, firmware version, and customer configuration. OPC UA, Modbus, MQTT, dozens of proprietary SDKs. A scanner integration built for one customer’s Epicor environment shares almost nothing with a scanner integration built for another customer’s SAP environment. You can’t build it once. Every deployment requires bespoke work.
For a platform company with investors expecting scalable revenue growth, that’s a structural “no.” Not because the problem isn’t real, but because the economics of solving it don’t fit their model.
What the factory floor actually requires
There’s also a genuine technical reason most platform vendors can’t serve this space, even if they wanted to.
OT systems operate under constraints that don’t exist in cloud-to-cloud integration. Real-time control systems need sub-second response times. Latency that’s completely acceptable in an API call is catastrophic in a machine control loop. You can’t run an industrial process through a cloud integration layer with unpredictable network lag.
The equipment itself adds another layer of complexity. A typical medium-sized manufacturing plant has over 200 individual pieces of equipment, often purchased across decades from different suppliers, each running its own automation platform.² Some of that equipment predates modern connectivity entirely. Legacy CNC machines, older PLCs, process control systems running software from the late 1990s. Connecting them requires industrial engineering expertise that no SaaS integration platform has built.
And then there’s the physical environment. Factory floors introduce electromagnetic interference, temperature extremes, and network reliability conditions that a cloud-native platform isn’t designed to handle. The same architecture that works perfectly for a Shopify-to-HubSpot integration doesn’t translate to a foundry floor.
70%
of Industry 4.0 pilots fail to scale beyond initial proof of concept²
>200
individual equipment types in a typical medium-sized manufacturing plant²
The cultural problem nobody puts in the pitch deck
Even when the technology is capable, most OT/IT integration projects stall for a reason that has nothing to do with code.
IT teams and OT teams have been operating as separate worlds for decades. Different leadership structures. Different priorities. Different definitions of what “working” means.
IT thinks about standardization, security governance, documented change management, and scheduled maintenance windows. OT thinks about uptime, real-time operational stability, and not disrupting a production line that’s been running reliably for ten years.
When IT drives an integration project, it tends to produce something technically correct and operationally unusable. When OT drives it, the solution works on the floor but creates security exposure that IT won’t sign off on. Most projects get stuck somewhere in the middle, with both teams frustrated and the integration backlog growing.
More than half of manufacturers report cultural misalignment between IT and OT teams as a primary barrier to integration progress.³ That’s not a technology problem. No amount of better tooling solves it. What solves it is having someone in the room who understands both sides of the conversation and can translate between them before the technical work starts.
The last mile no one owns
McKinsey gave this a name: the last-mile IT/OT problem. It’s the step where data chains from shop floor device to enterprise system break down, and where 70% of Industry 4.0 pilots stall after a successful small-scale proof of concept.²
The vendors who handle the first part of that chain do it well. Device manufacturers like ProGlove, Zebra, and Cognex have built gateways and SDKs that handle protocol translation for their specific hardware. Cloud integration platforms handle the enterprise side with precision. The gap is in the middle: getting device data from the OT layer into the integration layer where it can flow to ERP, analytics, and business systems.
Nobody has built a business specifically for that gap. The device manufacturers don’t want to be in the integration business. The iPaaS platforms don’t want to be in the industrial protocol business. The result is a gap that both sides acknowledge and neither side owns.
What this means if you’ve been told it’s “out of scope”
You weren’t being failed. You were receiving a rational answer from a vendor whose business model points away from your problem. The issue isn’t that OT/IT integration can’t be solved. It’s that the incentive structure of the platforms you’ve evaluated doesn’t reward solving it.
What the right approach actually looks like
The manufacturers who’ve made real progress here didn’t find a better version of the same platform. They changed what they were looking for.
Instead of a platform with a pre-built connector library, they found a partner with deep manufacturing operations experience. Someone who understood OT protocols because they’d worked with OT systems, not because they’d written a connector for them. Someone who could walk a production floor, understand the workflow, and build an integration that operators would actually use.
Instead of a self-service tool that assumed standardized APIs, they engaged a model that was willing to do the bespoke work the problem requires, but build it on a managed platform rather than custom code, so the result was maintainable after delivery.
And instead of separate IT and OT conversations, they found someone who could run both conversations simultaneously, map the workflow before writing a line of code, and deliver something that IT could support and OT could trust.
That’s a different category than what the major platforms are offering. It’s also the only approach that actually works at the last mile.
The honest caveat
None of this means the major platforms are bad tools. For SaaS-to-SaaS integration, they’re often the right answer. If you’re connecting Salesforce to HubSpot, or syncing your ERP to a logistics platform, the pre-built connector model makes perfect sense. Use it.
The problem is applying that model to a problem it wasn’t designed for. OT/IT integration is a different category of work. It requires a different kind of partner, a different engagement model, and a realistic understanding that the solution is going to be more complex than a pre-built connector and less complex than a full custom build.
The gap is real. The tools to close it exist. What’s missing, for most manufacturers, is the partner who knows the difference.
Want to go deeper?
Download the OT/IT Integration Playbook
This article is part of a three-part series. The full playbook covers the market gap in detail alongside the cost of inaction and a practical framework for evaluating integration readiness.
See what the gap is costing your operation
The OT/IT Integration Cost Calculator takes about 90 seconds. Enter your floor headcount, shift structure, downtime hours, and revenue per production hour to get a cost estimate based on your numbers.
Sources
1 Global Growth Insights, “iPaaS Market Size, Growth & Forecast Insights 2026–2035,” February 2026. Market valued at USD 12.81 billion in 2025, projected CAGR of 33.92%. https://www.globalgrowthinsights.com/market-reports/ipaas-market-100579
2 McKinsey & Company, “It’s the Last IT/OT Mile That Matters in Avoiding Industry 4.0’s Pilot Purgatory,” 2018. https://www.mckinsey.com/capabilities/operations/our-insights/operations-blog/its-the-last-it-ot-mile-that-matters-in-avoiding-industry-40s-pilot-purgatory
3 Smart Industry / Dassault Systèmes, “IT-OT Convergence as a Driver for Manufacturing Innovation,” September 2025. https://www.smartindustry.com/benefits-of-transformation/it-ot-convergence/blog/55318152/it-ot-convergence-as-a-driver-for-manufacturing-innovation