There’s a quiet tax most organizations pay every single week, and almost no one accounts for it properly. It doesn’t show up as a line item on the P&L, but it drains momentum all the same. People spend hours hunting for information, re-creating work that already exists, waiting for answers that live in someone else’s head. APQC’s research puts a number on it: more than 13 hours per employee per week lost to friction like this.
The uncomfortable truth is this: most teams are not slow because they lack talent or effort. They’re slow because knowledge doesn’t move.
And that’s where the conversation needs to change. This is no longer about “knowledge management” as a back-office function or a dusty repository. It’s about whether your organization can think, learn, and act as a system instead of a collection of individuals.
Knowledge Isn’t Power If It’s Trapped
One of the biggest blind spots leaders have is assuming that expertise equals resilience. It doesn’t. Expertise locked inside people’s heads is fragile. APQC highlights a risk many organizations are underestimating: by 2030, more than half of frontline workers over 55 are expected to leave the workforce, yet only about a third of organizations document their critical knowledge.
That gap is dangerous. Not because people are leaving, but because organizations never built a way for experience to turn into shared capability.
The organizations that get this right treat knowledge transfer as a deliberate act. Sometimes that means structured interviews with experts. Sometimes it’s as simple as capturing instructions, templates, or decision logic and putting them where others can actually find and use them. The key insight is this: knowledge only becomes an asset when it survives its original owner.
TechnipFMC’s approach, described in the paper, is instructive. They didn’t just extract information from retiring experts. They created a process that honored expertise, mapped what was truly at risk, and intentionally closed gaps before people walked out the door. That’s not documentation. That’s stewardship.
Innovation Is a Team Sport, Not a Genius Act
Another myth worth challenging is the idea that innovation comes from star performers. In reality, innovation comes from recombination. APQC frames this well: collective intelligence emerges when people can easily build on what others have learned.
NASA’s “Pause and Learn” practice is a great example. Instead of waiting until the end of a project to reflect, teams regularly stop, capture lessons, and feed them into a shared system. Even failures become reusable assets when they’re made visible and understandable.
If you think about it, lessons learned are one of the highest-ROI activities an organization can do, yet they’re often treated as optional or ceremonial. That’s a leadership failure, not a process problem.
Silos Are a Design Choice
We talk about silos as if they’re natural. They’re not. They’re designed, often unintentionally, through structure, incentives, and tools.
APQC’s data shows fewer than one-third of organizations use communities of practice to drive productivity. That’s staggering when you see what’s possible. Collins Aerospace has run more than 90 active communities for over two decades, connecting thousands of employees across functions and geographies. These communities don’t just share knowledge, they accelerate problem-solving and make a massive organization feel navigable.
Here’s the uncomfortable question leaders should ask themselves: do your systems reward people for sharing what they know, or for hoarding it?
AI Won’t Save You From Messy Knowledge
This is where many organizations are currently fooling themselves.
They invest in AI expecting magic, only to discover that the outputs are inconsistent, outdated, or untrustworthy. People stop using the tools. Confidence drops. The problem isn’t AI. It’s the foundation underneath it.
APQC is blunt about this: AI amplifies whatever knowledge environment you already have. If your content is fragmented, outdated, or ungoverned, AI just helps you reach bad answers faster.
Infosys got this right by embedding AI into everyday tools and feeding it curated, reusable knowledge. The result was a 50% reduction in search time and an 80% boost in user satisfaction. Notice the sequence. Knowledge first. AI second.
This is the part many leaders miss: AI is not a shortcut around discipline. It’s a multiplier of it.
Learning Has to Move at the Speed of Work
The final shift the paper points to is learning. Traditional training can’t keep up with regulatory change, new technologies, or evolving customer expectations. When knowledge isn’t embedded in daily work, organizations stall while people scramble to interpret what changed.
Communities, reusable examples, expert mentoring, and microlearning aren’t “nice to have.” They’re how organizations keep moving when the environment won’t slow down for them.
The organizations that win here measure what matters. Not course completions, but cycle times, error rates, onboarding speed, and risk reduction. Learning only matters if it changes outcomes.
The Real Edge
The most important idea in APQC’s paper is also the simplest: the future belongs to organizations that treat knowledge as a strategic asset, not an afterthought. Collective intelligence isn’t about technology. It’s about intent. It’s about designing systems where what one team learns today makes another team better tomorrow. Where AI has something worth learning from. Where people spend less time searching and more time deciding.
If there’s a challenge to leave you with, it’s this:
Don’t ask whether your organization has smart people. Ask whether it gets smarter over time.
That difference is where performance is being won now.