RHI Magnesita is the kind of company most startup founders never think about. They make refractory products — heat-resistant materials used in steel production, cement manufacturing, and industrial processes. It’s not sexy. It’s not disruptive. It’s a EUR 3+ billion company making things that make other things possible.
Working as Innovation Manager there taught me that enterprise innovation operates under completely different rules than startup innovation. The frameworks, the pace, the constraints, and the success criteria are all different. And most of the startup-world advice about innovation is not just useless in enterprise settings — it’s actively harmful.
This isn’t a critique of RHI or any specific enterprise. It’s a set of observations about what innovation actually looks like inside a large, complex, highly regulated industrial organization. If you’re building for enterprise customers, or if you’re inside an enterprise trying to innovate, this is what I wish someone had told me before I walked in.
Enterprise Innovation Is Not Startup Innovation With More Money
The first lesson was the most fundamental: enterprise innovation and startup innovation are different disciplines, not different scales of the same discipline.
Startup innovation is about finding product-market fit with minimal resources and maximum speed. You build fast, fail fast, iterate fast, and let the market tell you what works. Ship it ugly is the right approach. Speed is the competitive advantage.
Enterprise innovation is about creating new value within an existing system of constraints — regulatory requirements, existing customer relationships, brand reputation, operational infrastructure, organizational politics, and shareholder expectations. Speed is not the primary advantage. Alignment is.
An enterprise can’t “ship ugly.” Their ugly ships to real customers who depend on product consistency, who have contractual quality standards, and who will hold the company liable for failures. The stakes of a bad product in a regulated industrial environment aren’t a bad review — they’re a factory shutdown or a safety incident.
This means enterprise innovation requires a different cadence: more validation upfront, more stakeholder alignment, more testing, and more documentation. Not because enterprises are slow and bureaucratic (though some are) — because the cost of getting it wrong is orders of magnitude higher.
At RHI, a new product concept went through a validation process that would make a startup founder’s head explode: lab testing, pilot testing with one customer, expanded testing with three customers, regulatory compliance review, manufacturing feasibility study, commercial viability analysis, and then — maybe — market launch. This process took twelve to eighteen months minimum.
Was this too slow? In some cases, probably. But the alternative — shipping a flawed refractory product that fails in a steel furnace — wasn’t an option. The validation process existed for reasons, and respecting those reasons was the first step to working effectively within the system.
What Actually Drives Enterprise Innovation
After spending several years inside enterprise innovation processes as Innovation Manager, I identified what actually produces results — and it’s not what the books say.
Internal champions matter more than external consultants. The projects that succeeded at RHI always had at least one senior internal champion who understood both the innovation opportunity and the organizational politics. External consultants (including me) could bring frameworks and facilitation, but the person who moved the project through the organization had to be someone with internal credibility and relationships.
My role was most effective as a tool for the internal champion — providing methodology, facilitating workshops, creating documentation that the champion could use to build support. When I tried to drive projects directly, without a strong internal champion, progress stalled because I lacked the organizational context to understand and address internal resistance.
Cross-functional teams accelerate everything. The biggest bottleneck in enterprise innovation wasn’t the speed of work — it was the speed of handoffs between departments. R&D develops something, hands it to manufacturing, manufacturing identifies problems, sends it back to R&D. This loop could consume months.
The solution that worked: cross-functional innovation teams with representatives from R&D, manufacturing, sales, and quality from day one. When all perspectives are in the room during development, handoff delays shrink dramatically because potential problems are identified and addressed in real-time rather than sequentially.
Small bets, not big bets. The most successful innovation projects at RHI were small, focused experiments — not company-wide transformation programs. A new product for one customer segment. A process improvement in one manufacturing line. A service extension for one market. These small bets could be executed quickly, produced real data, and built organizational confidence for larger initiatives.
The Subtraction Audit was originally developed in a different context, but the principle applies to enterprise innovation: remove complexity before adding it. The most impactful innovations I saw at RHI were simplifications, not additions.
The Innovation Theater Problem
I need to address something uncomfortable: much of what enterprises call “innovation” is theater. Innovation labs that produce no products. Hackathons that generate excitement but no execution. Innovation teams that exist to make the annual report look progressive but have no authority or budget to actually ship anything.
I’ve participated in some of this theater. Early in my career, I ran innovation workshops for companies that had no intention of implementing the results. The workshops were about looking innovative — creating the appearance of forward-thinking leadership without the discomfort of actual change.
Recognizing innovation theater — and declining to participate in it — was an important professional boundary I eventually established. The 5-Conversation Sprint partly exists as a theater detector: if a company isn’t willing to have five real conversations with potential customers about a new idea, they’re not seriously innovating. They’re performing.
At RHI, the distinction between real innovation and theater was clear. Real innovation had budget, authority, customer access, and executive commitment. Theater had a nice office, a creative name, and no path to implementation. Learning to tell the difference quickly saved me enormous amounts of time.
What Enterprises Can Learn From Startups (And Vice Versa)
The most productive perspective isn’t “enterprises should act more like startups” or “startups should be more like enterprises.” It’s a selective integration of the best from both worlds.
What enterprises can learn from startups:
- Customer proximity: talk to customers directly and frequently, not through layers of market research
- Speed of decision-making: some decisions don’t need committee approval
- Willingness to experiment: small experiments that might fail are cheaper than large projects that must succeed
- Iteration over perfection: version one doesn’t have to be final
What startups can learn from enterprises:
- Stakeholder management: understanding who needs to be aligned and why
- Quality systems: building reliability into the process, not just the product
- Documentation: capturing knowledge so it survives personnel changes
- Risk management: understanding what failures cost and planning accordingly
My career — moving between startup acceleration, solo consulting, and enterprise innovation — gave me exposure to both worlds. The most valuable skill I developed wasn’t expertise in either world. It was the ability to translate between them — to bring startup speed into enterprise contexts and enterprise rigor into startup contexts, adjusting the mix based on the specific situation.
AI as a business tool is reshaping both worlds equally, and the enterprises that integrate AI into their innovation processes — not as theater, but as genuine capability — will have significant advantages over those that don’t.
The Lasting Impact on My Thinking
The enterprise innovation experience fundamentally shaped my approach to all business building. Three specific impacts:
I became more rigorous about validation. Watching enterprise innovation fail when assumptions went untested made me more insistent on validating assumptions in my own businesses. The Subtraction Audit, the 5-Conversation Sprint, the Owner Dependency Score — all of these frameworks include validation steps because I’ve seen what happens when you skip them, at both startup and enterprise scale.
I developed more patience with process. Enterprise work taught me that some processes exist for good reasons, even when they feel slow. This patience translated to my own business: some things take time, and rushing them produces worse outcomes than respecting the natural cadence.
I learned to read organizational dynamics. Understanding how large organizations make decisions — who has influence, what motivates them, where resistance comes from — is a skill that transfers directly to any business context. Even a two-person partnership has organizational dynamics.
Enterprise innovation isn’t glamorous. It doesn’t make for exciting conference talks or social media posts. But the discipline, rigor, and stakeholder awareness it teaches are genuinely valuable — and they’ve made me better at building my own, much smaller, much faster businesses.
Key takeaways:
- Enterprise innovation and startup innovation are different disciplines — speed is the startup advantage; alignment is the enterprise advantage.
- Internal champions are more important than external consultants — the person who moves the project through the organization must have internal credibility.
- Cross-functional teams from day one shrink handoff delays dramatically — include R&D, manufacturing, sales, and quality in the room from the start.
- Distinguish between real innovation (budget, authority, customer access) and innovation theater (nice office, no implementation path) — decline to participate in the latter.
- The most productive approach integrates startup speed with enterprise rigor — bring customer proximity and experimentation into enterprise contexts, and stakeholder management and quality systems into startup contexts.
