Why IT Departments Struggle to Drive Innovation

Every year, organizations convince themselves they're on the verge of a data-driven renaissance, only to find themselves facing familiar challenges when December rolls around. Let’s make this year different! Year after year, companies hire specialists, license analytics platforms, and launch transformation initiatives, yet remain entangled in cumbersome spreadsheets, conflicting definitions, and isolated information. Even companies with cutting-edge tech stacks continue to wrestle with fragmented databases and incompatible data models—the legacies of countless tactical compromises. The key to finally tackle these issues is to realize that at its core, their root cause isn't technological, but human and organizational in nature. Messy and siloed data stems from misaligned incentives, entrenched cultural patterns, and expedient solutions that calcified into permanent architecture. When performance metrics are focused solely on operational targets and no rewards for data quality or sharing, information remains locked in departmental strongholds, each with their own language, priorities, and interests. Doing it differently starts with strategic planning, where business leaders tend to passionately debate product launches and expansion plans, only to later ask the data teams to provide the supporting data pipelines. Instead of being decision co-pilots, data teams become post-hoc service providers—a telltale sign of data's relegation to a support function. This year, give them their rightful place as a strategic driver. The path forward requires elevating data to the same strategic level as people, capital, and core products. Data must finally become the connective tissue binding everything together, not a mere byproduct of operations. This means rewarding data sharing, dismantling organizational gridlock, and redesigning culture around data as a strategic asset—all while systematically addressing the technical debt that holds innovation hostage. The good news? The path to meaningful change doesn't need another major technology investment to start with decisive steps: tie executive compensation to data quality metrics, establish empowered cross-functional data councils with real decision-making authority, and create data ownership roles that transcend departmental boundaries. For early-stage companies, this means embedding data professionals in product teams; for enterprises, it requires establishing federated data governance that effectively balances central control with departmental autonomy. The question isn't whether you'll invest in new tools—it's whether you'll finally dare to reshape the human systems and organizational architectures that determine your data destiny.

If you’re trying to implement change and it keeps stalling - your company isn’t broken. It’s doing exactly what it was designed to do: protect the status quo. Most organisations aren't designed to change. They're designed to survive. AI? Agents? Large scale transformation? These aren't plug-and-play upgrades. They're threats to the current state. And systems - by nature - resist threats: - They reassign ownership - They slow approvals - They "agree in principle" while dragging in practice It's not dysfunction. It's defence. So if you’re leading change, stop expecting applause. Start expecting antibodies. You can blame culture. Or leadership. Or “change fatigue.” I blame physics. Specifically: Le Chatelier’s Principle. In chemistry Le Chatelier’s Principle describes how systems respond to disturbance. When you apply pressure to a stable system, it reacts to restore balance. Add heat? It cools. Add tension? It redirects. Sound familiar? - AI pilot's that die in committee - Shiny new tools that got quietly shelved - Transformation project that fizzled after the kickoff deck So... what can we do: - Bringing in AI? Don’t just focus on the model - focus on the humans it unsettles. - Scaling analytics? Don’t just show dashboards - remove the friction in decision-making. - Driving innovation? Don’t expect open arms - design with counter-reactions in mind. Change doesn’t always stick because it’s smart. It sticks because it finds a new equilibrium. Disrupt. Stabilise. Repeat. The system isn’t broken. It’s doing exactly what it was built to do.

Technology isn’t a cost center, it’s your competitive edge. If you can’t shift this perspective, you won’t be able to innovate 📈 Over the past decade, I've guided numerous companies, from pharmaceutical giants to financial leaders, on their journey to becoming product-led. One common mistake I've seen is treating technology as a mere expense, not a strategic advantage. This view undermines transformation efforts. Here's how it unfolds and what to do about it. In many organizations, technology is seen as a cost center. When executives talk strategy, it's often about cost reduction. They can articulate market differentiators well but stumble when asked, "How does your tech vision enhance your competitiveness?" Silence. The competitive edge dulls as rivals leveraging tech strategically catch up. This approach is like playing corporate whack-a-mole: solving cost issues while missing opportunities. What if the process you streamlined wasn't needed at all? Or if you could innovate beyond traditional methods? Many transformations start with Agile to address slow development cycles. But speed alone doesn't equate to success. Agile without product thinking can lead to an output-focused mindset: success measured by backlog clearance rather than solving real business problems. Transformations stall when teams build features quickly without building the right ones. It's crucial to view software products as strategic enablers, not just tools to "run the business." Without this shift, product strategies remain uninspiring. Even if your software isn't sold, it can be a major strategic differentiator. Consider Capital One's journey: disrupting the banking sector by using data analytics for credit risk models and improving customer experiences by eliminating unnecessary processes. What about internal tools? For pharmaceutical companies, bringing drugs to market is essential. Instead of merely speeding up processes, your tech could identify study participants and predict outcomes better than competitors. It's about asking the right questions. "How do I make this cheaper?" leads to outdated solutions. "How do we re-imagine this process for an exceptional experience?" drives innovation. If you're on this journey, start by changing the conversation. Ask "why?" and "what if?" Shift from cost-cutting to value-creation, from outputs to outcomes, from project management to product thinking. Real transformation isn't about new processes or team reorgs. Those are secondary. The core shift is viewing technology as a strategic asset driving business value. That’s where the real transformation begins.

This is an uncomfortable truth many GCC leaders recognise but rarely say out loud. You cannot ask Indian GCCs to innovate, disrupt, or think strategically while the system they operate in still rewards deference, compliance, and fast agreement. But this is not just a Global HQ problem. Yes, innovation struggles when India is treated primarily as an execution engine. When key decisions sit elsewhere. When challenging assumptions is mistaken for resistance. When “alignment” quietly means “don’t disagree.” However, the responsibility does not stop there. Culture is not only imported. It is also built locally. And ultimately, Indian GCC leaders own the culture their teams experience every day. If teams hesitate to say no, to challenge, or to push back, it is often because local leadership has (sometimes unintentionally) taught them that safety lies in compliance. Here’s the harder part of the conversation: If GCC leaders want true innovation, they must actively unlearn and rewire certain patterns: ·      Create clear boundaries between respect and deference ·      Signal, repeatedly, that disagreement is not career-limiting ·      Protect people who speak up, especially when it’s uncomfortable ·      Push decision-making down instead of escalating everything up ·      Hire leaders who are capable of independent judgment, not just operational excellence The same behaviours that once made GCCs incredibly successful at scale can now quietly limit their next phase of growth. Innovation does not come from permission. It comes from psychological safety, trust, and leadership courage. And that courage is required on both sides: ·      from Global HQs, to genuinely invite challenge ·      and from India GCC leaders, to model it, reward it, and hire for it Until both happen together, the innovation mandate will remain aspirational.

5 invisible forces that block innovation (and hide in even the best teams). By the time you realize innovation is stuck, it’s probably already been stalled for a while. Leaders often assume innovation gets blocked by lack of ideas or talent. In reality, it’s much more subtle. 5 quiet blockers of innovation: 1. Success Becomes a Straitjacket ↳ When what’s always worked keeps working, there’s no urgency to try something new ↳ Teams get optimized for consistency, not creativity 2. The Pressure to Perform Kills Risk ↳ Innovation needs room to fail ↳ When every miss feels costly, bold ideas stay buried ↳ High expectations + low psychological safety = quiet compliance 3. Over-Optimization Leaves No Slack ↳ If every hour is scheduled, it crowds out creativity. ↳ Innovation lives in the white space. ↳ No slack = no spark. 4. Groupthink in Disguise ↳ Alignment is good, until it becomes “don’t rock the boat” ↳ Breakthroughs require dissent, debate, and diverse perspectives. 5. Too Much Focus on the Now ↳ When everything is urgent, nothing is strategic ↳ Short-term wins quietly crowd out long-term bets ↳ If no one owns the future, it never gets built Don’t assume innovation will just “happen.” → Make space for exploration → Reward smart risks → Invite diverse thinking → Tolerate failure along the way Performance delivers results. Innovation expands what results are possible. You need both... just on different time horizons. ♻️ Repost to help another leader 🔔 Follow LK Pryzant for daily ideas on leadership, strategy, and career growth

Almost every manager has said it: “Don’t just bring me problems, bring me solutions.” It sounds reasonable and pushes against empty complaining. But here’s the unintended consequence: You silence the very people who are best at seeing problems. Because the skill of identifying nuanced, real-world friction is not the same as the skill of designing solutions. And expecting both in the same person, nonetheless, at the same time, can shut the door before the conversation even starts. So what happens? Problems go unspoken. Friction gets normalized. Workarounds become culture. I see this all the time in healthcare. People notice inefficiencies constantly: “Why do we say ‘if it’s not documented, it didn’t happen’?” “Why do we use this tool this way?” “Why does this process feel so… off?” But the system is complex. Solutions are hard. And when the bar to speak up is “have a fully baked answer,” most people just stop raising their hand. 🚫🖐️ You don’t get fewer problems. You just get fewer people willing to point them out. And that’s far more dangerous. Maybe the better approach is: “Bring me problems. Let’s build solutions together.” Because innovation doesn’t start with answers. It starts with awareness.

"Why aren't we talking to each other?" I've asked this question as a frustrated engineer. So have many others I've worked with. In one case, a team spent six weeks redesigning a component another department had already optimized. Nobody knew. This isn't a communication problem. It's structural. Organizational silos don't just hinder communication; they systematically destroy innovation and experimentation. Gartner and IDC research shows data fragmentation and silos cost companies millions in inefficiencies, delayed launches, and duplicated efforts. Yet these costs never appear on financial statements. The real damage isn't wasted resources. It's the impact on innovation velocity: ➡️ Problems get fragmented When challenges span departments, each team optimizes their piece without seeing the whole. I've seen quality issues persist for months because departments hit their targets while the overall process failed. ➡️ Knowledge gets trapped Critical insights never reach teams that could use them. One manufacturing leader told me: "We solved the same problem five times in five facilities because we had no way to share lessons learned." ➡️ Decision-making slows to a crawl Every handoff between engineering, operations, supply chain, and quality adds delay and distortion. When markets shift, this friction becomes fatal. How to transform siloed organizations: First, create shared outcomes. Replace department-specific metrics with cross-functional KPIs that require coordination. Second, establish structural bridges. Rotate high-potential team members through different functions for 90-day assignments. This builds human connections that span silos. Third, implement structured experimentation across departmental boundaries. Collaborative problem-solving dissolves silos naturally. The highest-performing manufacturers aren't those with the strongest departments, but those with the most effective connections between them. --- If this is a problem in your organization, let's talk.

The hardest part of IT isn’t technology. It’s culture. Most organizations think their biggest problem is tools, platforms, or architecture. It rarely is. Technology problems are visible. Culture problems are systemic. You can redesign processes. You can hire new skills. You can modernize platforms. But if the culture still rewards: • silos over ownership • speed over discipline • activity over outcomes The organization will eventually fall back into the same patterns. That’s why so many transformations stall. Not because the technology failed. Because the organization never changed how it works. The best IT leaders eventually learn this: • Technology transformation is the easy part. • Organizational transformation is the real work. “Culture eats strategy for breakfast.” — Peter Drucker

In 11 years as CIO, I watched the same pattern repeatedly. Business and IT fighting over a new tool or platform. If the business brought it to IT, IT became defensive. Another platform bypassing our controls. Another threat to the team we'd built. If IT brought it to the business, the business didn't trust it. Another technology being pushed on them without understanding what problem it actually solved. Both sides had valid points. Neither was wrong. But innovation died in the middle. I had a conversation with a partner recently who highlighted that nothing has changed. His company works closely with large enterprises trying to move faster on digital and AI. Same dynamic, every time. Business brings it to IT: defensive. IT brings it to the business: ignored. His conclusion was interesting. The only way through is a third party in the room. A Digital Office, an innovation team, someone who sits between IT and the business and translates in both directions. They show IT that better tooling means stronger governance and a business that stops going rogue. They show the business that working within standards doesn't have to mean waiting months to get started. How are you making sure business and IT actually work together in you organization?

👉 Takeaways from the #HonestNoBS, one-table, one-discussion, Chatham House rules dinner with data and leaders discussing what’s working and what’s not at Gartner Data & Analytics London ✅ What’s working: • Data engineers are getting it done. Ingestion and pipelines? Happening! • AI tools like ChatGPT are unlocking productivity. • Governance is starting to get its seat at the table: from “we don’t need governance” to “it’s critical and integrated.” • Communities that spotlight wins are fueling adoption. • Internal hackathons, giving people freedom to solve hypothetical problems, sparks product ideas and have led to real customer solutions. • Cross-functional teams = better innovation. Diversity of thought matters. • AI is pushing us to finally rethink what should be automated (yes, even Excel). • Younger folks are diving in headfirst. New talent wants documentation and governance. “New blood gets it.” Tracking metrics like metadata completeness makes this real. • Semantics is no longer a ‘nice to have.’ We’ve shifted from “it’s the right thing to do” to “it’s how we make AI actually work.” Leadership is open to talking semantics and it must tie to business value. • "Don't ask for data. Tell us what you're trying to achieve." That mindset shift is real. • Federated models work—even without central teams. Governance by design. • Buy > Build, if it’s best of breed. • Data Marketplace establish some control while enabling self service. ❌ What’s not working: • Departments still operate in silos. Money making business units pay IT to be left alone. • Legacy thinking: “It works, don’t touch it.” = innovation death spiral. • Forcing hackathons doesn’t build culture—freedom does. • Excel is still king. Why? People trust the illusion of control. • BI tools have created semantic silos that no one wants to share. • Knowledge hoarding. Job security via secrecy. • Older employees won’t adopt. Younger ones leave too fast. Stalemate. • Reorgs are the new norm. The people who adapt fastest win. • Ownership is murky. Business should own data—but they just say, ‘That’s the data team’s job.’” Engineers end up doing product work without support. • Engineer-led delivery often misses the user. • Black box semantics are dangerous. Be skeptical when semantics are opaque. 🧠 Final Thought BI—as we know it today—won’t exist in 5 years. The future? → A world where semantics are foundational. → A culture where AI doesn’t just support humans, it understands them. → A workforce that works backwards from intent, not forwards from tooling. And the companies that work backwards from AI? They’re going to win. I'm incredibly lucky for the opportunity to chat with so many data leaders. Thanks Tim Gasper and Ryan Cush for facilitating the discussions! AND... we are hosting this dinner in San Francisco during Snowflake and Databricks conference. Ping me if you want an invite. Exclusive for data leaders 👇 Let’s compare notes. What do you agree with? Disagree? What’s missing?