LNS Research

The average tenure of a COO is now just a few years. At the front line, turnover has always been a challenge. But leadership turnover at the top is a different kind of problem, and it's one that doesn't get nearly enough attention in conversations about organizational culture. A company's culture takes time to build. It builds through consistency, through leaders who show up the same way long enough for people to really trust what they're seeing. This seems especially true at manufacturers with multiple locations, varied legacy systems, and years of "we've always done it this way." The problem is that most organizations are cycling through senior leadership every two or three years. Typically, each new leader comes in with a new strategy, a new way of doing things, and a new set of priorities. As such, the culture underneath never gets the chance to set. You end up, as one of our research partners put it recently, forever in building mode. So, the question becomes: how do you change the leader without changing the culture? Some organizations have tried to solve it by defining specific value and behavior sets, and elevating people who genuinely align to them. Johnson & Johnson, one of our World's Most Productive Companies, is probably the most cited example of this done well. Their credo, written in 1943, is still guiding leadership behavior today. But it requires enormous discipline to sustain, especially when the pressure to bring in someone who will shake things up is high. Our research keeps pointing to the consistency gap of leadership changes as one of the most underappreciated drivers of cultural dysfunction in manufacturing organizations. The front line feels it every time. #LeadershipContinuity #OrganizationalCulture #ManufacturingLeadership

Most operations leaders, if you asked them, would probably estimate their data teams spend most of their time on analysis. That feels right, given what their functional name implies. But our latest research says otherwise: 70 to 80% of the data team's time goes to cleaning and preparing data just to make it usable. Few industrial organizations have named this as a problem yet. They may have an AI initiative or a data initiative. Sometimes they have both, but usually run by different teams. And this data foundation problem sits somewhere underneath all of it, somewhat unnoticed and definitely unresolved. What we're finding is that the Leaders have gotten ahead of this. Among the top 18% of companies actually scaling Industrial AI with measurable results, 62% have actively implemented Industrial DataOps practices. Among Followers, that number drops to just 35%.  Digging deeper still, at the bottom of the maturity curve, we find that Leaders have data governance on the roadmap almost without exception. Conversely, a significant portion of Followers still don't. The industry has been down a version of this road before. "Data is the new oil" sent manufacturers into a decade of cloud migrations, data lake build-outs, and aggressive collection initiatives. What came out the other side was petabytes of ungoverned raw data, technology solutions searching for problems, and pilots that never made it into production.  Nobody pulled off what Toyota Motor Corporation did with TPS; nobody pulled off what Motorola Solutions did with Six Sigma. The data just kinda sat there. LNS Research Analyst Vivek Murugesan's latest research makes the case that we're at a different moment now: capital moving in, product roadmaps maturing, and a vendor landscape that has moved considerably in just the past few months. If your AI and data strategies are still two separate conversations, you'll want to check out Vivek's full blog, linked in the first comment below. #LNSResearch #IndustrialTransformation #IndustrialAI #IndustrialDataOps

One of the more persistent myths in industrial transformation is that if you get one plant right, the rest will follow suit. Master the model... export the blueprint... scale across the network. It all sounds so logical. In reality, things seldom work that way. What we constantly hear from operators is that every plant is different in ways that matter. The equipment is different. The legacy processes are different. The leadership is different. Moreover, and perhaps most importantly, the culture is different. A framework that worked in one facility doesn't transfer cleanly to the next, no matter how good the design. This isn't an argument against having a framework; quite the contrary, in fact. What this really is is an argument against mistaking the framework for the work. The companies that scale successfully tend to be the ones that treat each plant as its own transformation, informed by what came before but not dictated by it. They take lessons learned and adapt them, rather than rinse and repeat. There's a useful parallel in how lean manufacturing spread through industry. The companies that adopted TPS wholesale and called it done generally didn't get the results that Toyota Motor Corporation did. The ones that took the principles, internalized them, and built their own version of them did, however. This same dynamic is now playing out with industrial transformation and AI. There's also a universal truth when it comes to replication: you can't make it your own if you're too busy copying the original. #IndustrialTransformation #SmartManufacturing #ChangeManagement

There's something special about when a fully remote team gets together in person. This past week, the LNS Research team was in Cambridge, MA, for our annual all-staff meeting. Each year, it's about sharing ideas, improving how we work, and some much-appreciated fun. We do all of this every day in a virtual setting, but there's nothing quite like being in the same room together to align on strategy and plan for the year ahead. Not surprisingly, the topic of AI was front and center this year. Departmental hackathons, AI show-and-tells, and a very fun "Prompt Olympics" (including a little team competition, because we're a competitive group, after all) gave us the chance to dig into real business challenges and what AI actually makes possible. And with our mission of helping industry leaders transform their organizations at the heart of everything we do, we spent real time on what that means for our clients in this new AI landscape. Even with a slight Titanic moment on a dinner cruise — fountain, not glacier — a fantastic time was had by all! Thanks to Matthew Littlefield, Mehul Shah, Diane Johnson, Sera Holt, Niels Erik Andersen, James Wells, Allison Kuhn, Vivek Murugesan, Michael Carroll, Ryan Cahalane, Cristina Barragan, Madeline K., Jessica Rodriguez, Brady Aitken, Lee Shiro, John MacDonald, Jim Beigel, Thomas Souza, and Jason Cariglia. #TeamCulture #IndustrialTransformation #AI

Ask most manufacturers what they want from industrial transformation, and the answer sounds reasonable enough: find what works, standardize it, and then scale it across the network. Sounds clean, logical, and efficient, right? The plants, however, have other ideas. Of course, every facility carries its own history. The equipment was chosen at different times (some, quite long ago) by different leaders with different priorities, and it shows. Some companies have tried to impose top-down standardization all the way to the equipment level; we have even seen some try to go remarkably far down that rabbit hole. But most find it becomes an endless project that never quite delivers and has to start over every time there's an acquisition or a technology refresh. What we're seeing work better is an entirely different kind of standardization. Instead of trying to make the equipment itself uniform, you standardize how you interact with it. This builds the layer that lets different machines, systems, and data sources speak a common language to the business above them, without requiring everything underneath it to be identical. Sure, it's less glamorous than a uniform factory floor, but it's also much more realistic. The goal has never been about making every plant look the same. It's about being able to see, understand, and act across all of them with trust and transparency. Making every plant identical and being able to operate across all of them aren't the same problem, and a lot of transformation programs run into trouble when they're treated as if they are. #IndustrialTransformation #SmartManufacturing #OperationalExcellence

Most manufacturers deploying AI are focused on what it can generate. That's the natural starting point, and it makes sense. But the more interesting opportunity, at least in the industrial space, can often be found on the other side of that equation. That might sound like a knock on the technology. It isn't. It's actually a pretty solid design principle. When you use AI to generate something and then run a second system against it to pressure test the output, check it against your original requirements, and flag inconsistencies where they exist, something shifts. The two systems working against each other tend to get you somewhere more reliable than either one working alone. Usually, the more you run it, the better it gets (provided you don't over-engineer it). What we are seeing more and more of at LNS Research is that manufacturers who are getting the most traction out of AI aren't always the ones using it to generate answers. Rather, they're the ones using it to interrogate answers, to ask whether something holds up before it moves forward in a workflow. In an environment where the tolerance for error is low and the cost of getting it wrong is real, that's not a consolation prize. Most of the value we're seeing isn't in what AI generates. It's in what it catches. #AITransparency #ResponsibleAI #AIGovernance

Most manufacturers have spent the last decade building systems to capture machine data. Asset health, OEE, and process parameters all have dashboards. What hasn't kept pace is capturing what the people know. And that gap is about to get very expensive. What we're seeing more and more of at LNS Research is two ends burning toward the middle. At one end, the workforce that carries the deepest institutional knowledge is retiring out faster than it can be replaced. At the other end, AI-driven change is moving faster than most transformation roadmaps were written to handle. The window to capture what your most experienced people know and then build it into how the next generation works is shorter than most organizations are planning for. Safety is the most visible consequence when that knowledge walks out the door; our research shows it goes the wrong way fast when knowledge management breaks down. But the operational impact runs even deeper still. On-time delivery, operating margin, and new product introduction all follow suit as knowledge quality slips. The organizations that treat this as a workforce issue, not just a technology issue, tend to see it show up across the board. However, the more important shift is in what empowering the frontline actually means. It's not putting a tablet on the line and calling it a connected workforce. It's giving workers exactly what they need, when they need it, in a form they can actually use. Our research shows Leaders are building that kind of system; Followers are still working on the former. The runway is shorter than most people think: Not five years; closer to two. We're digging into all of this at LNS Research's The Productivity Event, June 9-11 in Huntington Beach, CA, specifically in our Industrial AI Insights session on Knowledge and Workforce Empowerment. Leading the session is LNS Research Analyst Allison Kuhn; she'll be joined by session executive speakers Barrett Hopper of MKS Inc., Cory Jensen of Continental Manufacturing Chemist, Inc., and Nick Jansen of Green Bay Packaging. See comments below to register. #IndustrialProductivity #TheProductivityEvent #KnowledgeManagement #IndustrialAI

There's a mismatch at the heart of most industrial AI deployments that doesn't get examined closely enough. Manufacturing is built around precision and repeatability. However, generative AI, by its nature, is probabilistic. That's not a flaw, it's just, it is what it is. But those two things are not in the same neighborhood, and glossing over that gap is where a lot of projects get into trouble. It gets even more complicated when you factor in that manufacturing isn't just a single decision; it's a sequence of them. Think about something as routine as a batch release process: an AI flags an anomaly in raw material testing, a second system assesses whether it affects downstream quality, and a third determines whether the batch meets spec for release. Each step depends on the one before it. If each system is highly accurate but not perfect, those imperfections don't cancel each other out. They stack. By the end of a five or six-step sequence, your cumulative reliability might look very different from what it looked like at step one. None of this means AI doesn't belong in manufacturing. It absolutely does, in some form or another. But deploying it without understanding where that risk lives is how you end up with outcomes that are hard to explain, and even harder to defend. The real issue isn't can AI do this. It's where does probabilistic output create acceptable risk, and where does it create unacceptable risk. Those are two different things that require very different approaches. #IndustrialAI #ManufacturingRisk #AIInManufacturing

Industrial data problems didn't start with AI, and they won't end with it either. Most manufacturers are sitting on years of operational data that never quite reaches the place where it could actually change a decision. AI raises the stakes on closing that gap considerably. These old data problems persist, but now with new consequences. LNS Research's just-released 2026 Industrial AI Data Platforms Solution Selection Matrix is the follow-up to our Advanced Analytics edition. This time, we examine the data layer that determines whether your AI investments have anything real to stand on. In the Data Platforms SSM, Analyst Vivek Murugesan evaluates eight vendors: Braincube, C3 AI, Cognite, Palantir Technologies, Quartic, Sight Machine, SymphonyAI, and TwinThread. Sure, they come from different backgrounds, have built different architectures, and are targeting different parts of the industrial market. But what they do have in common is a serious attempt at solving the same data connectivity, quality, and contextualization challenges manufacturers have wrestled with for years... now with considerably more capable tools at their disposal. One of the key findings from our latest Industrial AI SSM research is that most mid-to-large manufacturers will need two to three types of platforms, not one. The assumption that a single vendor can solve the full industrial AI stack is still driving a lot of buying decisions, and it's still getting companies into trouble. Use the SSM to help you build and evaluate your vendor shortlist. Moreover, it's time to rethink your data instincts and resist a common pitfall. Fixing your data before launching your analytics initiative sounds logical, but it isn't practical. If you wait for your data to be fully fixed, you'll be waiting forever. Data problems have to be solved alongside analytics work, not before it. And when evaluating vendors, the Solution Selection Matrix flags something we hear often from operations leaders: time-to-value and the ability to deliver insights at scale matter considerably more than how elegant the underlying model is. To read the LNS Research Solution Selection Matrix on Industrial AI: Data Platforms, click the link in the first comment. #IndustrialTransformation #IndustrialAI #DataPlatforms #SolutionSelectionMatrix

After the year industrial companies just had, most COOs have a version of the same question: Did we actually make progress in 2025, or did we just get through it? The Industrial Productivity Index™ measures exactly that. LNS Research's annual Index (IPI) tracks hundreds of publicly traded companies across 10 major industries and dozens of peer groups, measuring how efficiently they convert inputs into outputs across the full value chain and how that performance shifts year over year. The 2026 Industrial Productivity Index results, based on nearly 700 companies this year, are being released live on May 14th. Join the LNS Research team for the hour-long LinkedIn Live starting at 11am ET. The team will walk through what the data is actually showing this year, including early signals on how AI may be moving the productivity needle, and what it means for COOs managing operations and transformation right now. We'll also take your live questions. Industries covered once again include aerospace and defense, automotive, chemicals, consumer goods, energy, food and beverage, high-tech, industrial equipment, life sciences, and materials. And just ahead, the IPI will lead us to the big reveal of the 2026 World's Most Productive Companies™ Top 100 at The Productivity Event in Huntington Beach, CA, June 9-11. These are the manufacturers achieving outstanding productivity growth despite a turbulent and ever-changing industrial climate. The 2026 Industrial Productivity Index drops May 14th. Join us live! Matthew Littlefield, Niels Erik Andersen, Vivek Murugesan, Tom Comstock, James Wells #LNSResearch #IndustrialTransformation #IndustrialProductivityIndex