Documented Knowledge vs Lived Knowledge

This is the distinction that explains why companies believe they have organisational memory when they don't.

Documented knowledge is the stuff that's been written down. The wiki page. The SOP. The onboarding deck. The CRM record. The quarterly report. Companies invest heavily in documentation, and when someone asks "do we have good organisational memory?" they point to these artefacts.

But documented knowledge is a tiny fraction of what the company actually knows. It's the knowledge someone had time to write down, in a format that made sense to them, on a day when they thought it was important enough to bother. It's filtered, delayed, and incomplete by design.

Lived knowledge is everything else. The context from the call where the deal terms were shaped. The reason the team chose one architecture over another. The relationship dynamics between the buyer and their CFO. The throwaway comment that turned out to matter three weeks later. The pricing exception that was approved and why. The customer's specific concern about data migration that never made it into the handover doc.

Lived knowledge is where the real value sits. It's also the part that almost never gets captured. Not because people are lazy, but because the volume of it is overwhelming and the mechanisms to capture it don't exist. You can't wiki your way to organisational memory when most of the knowledge that matters is too contextual, too fleeting, or too nuanced to fit in a document.

Why Most Companies Don't Have It

The default state of organisational memory is zero. Knowledge has to be actively captured, stored, and made accessible for it to count. Without a deliberate mechanism, what people learn stays in their heads until it fades or walks out the door.

Most companies rely on three mechanisms that feel like organisational memory but aren't.

Individual memory. The senior rep who knows every account inside out. The engineer who remembers why the system was built a certain way. The founder who holds the full picture of every customer relationship. This isn't organisational memory. It's personal memory on loan to the company. When that person is on leave, promoted, or gone, so is the knowledge.

Documented artefacts. The wiki, the shared drive, the Confluence space, the handbook. These capture what someone chose to write down at a point in time. They're static, they go stale, and they rarely contain the contextual knowledge that matters most. The wiki tells you the process. It doesn't tell you what happened on the call that changed the process.

Institutional habits. "We do a handover doc when someone leaves." "We write call notes after every meeting." "We have a weekly sync to share updates." These are attempts to create organisational memory through discipline. They work until someone is too busy, which is most of the time. Any system that depends on consistent human behaviour in a high-volume environment will degrade.

Real organisational memory doesn't depend on any individual remembering. It doesn't depend on someone choosing to document. And it doesn't depend on a process being followed perfectly every time.

The Symptoms

Companies without organisational memory experience the same problems repeatedly, but they tend to misdiagnose the cause.

Slow onboarding gets blamed on inadequate training. But the real issue is that the new hire has no access to the context that made their predecessor effective. The relationships, the history, the account-specific knowledge. None of it transferred.

Generic follow-ups get blamed on weak sales skills. But the rep heard the details on the call. They just couldn't retain all of them across 30 other calls that week. The follow-up reflects what they half-remember, not what was said.

Repeated customer questions get blamed on poor internal communication. But the customer already told the sales team everything. The problem is that none of that context survived the handover to CS. So the customer repeats themselves, and trust erodes.

Bad forecasts get blamed on CRM data quality. Which is true. But the CRM data is bad because the people who have the knowledge don't have the time or the mechanism to put it there.

Each of these is a symptom of the same underlying condition: the company doesn't remember what it has learned.

What Real Organisational Memory Looks Like

A new rep joins the team and has full context on every account they inherit. Not just the deal stage and the last activity date, but what was discussed, what was promised, what matters to this person, and what happened three months ago that's still relevant.

A CS manager prepares for a renewal and already knows the customer's concerns, the relationship dynamics, and the specific commitments made during the sales process. The customer feels like the whole company has been paying attention, not just the one person who sold them.

A founder steps into a call with a prospect they met six months ago. They know the prospect was going through a reorg, that they were worried about implementation timelines, and that they'd mentioned their team's resistance to new tools. The conversation picks up where it left off.

A product team reviews customer feedback and finds not just NPS scores, but actual context from hundreds of conversations about what customers are struggling with, in their own words, captured as those conversations happened.

In each case, the knowledge that was created through normal work persisted beyond the person who created it. That's organisational memory.

Documentation Won't Get You There

The instinct when companies recognise a knowledge problem is to ask people to document more. Write better notes. Update the CRM more diligently. Build a wiki. Start a knowledge base. Run a documentation sprint.

This instinct is understandable and almost always wrong.

The volume of knowledge created through daily work is too high for manual capture. A company running 200 customer calls a week generates more context than any team could ever write down, even if writing things down was their only job. And for the people who are actually on those calls, documentation is in direct competition with the work itself. Every minute spent writing up notes is a minute not spent on the next call, the next deal, the next customer.

The companies that build real organisational memory don't do it by asking people to work harder at documentation. They find ways to capture knowledge as a byproduct of work, not as an additional task layered on top of it.

The reframe is this: the goal isn't better documentation. The goal is organisational memory. Documentation is one input. But if it's the only input, the memory will always be incomplete.

Where This Is Headed

Organisational memory has historically been treated as a nice-to-have. Something that mature companies get around to building once they've scaled past a certain size. A knowledge management initiative on the roadmap, somewhere behind the CRM migration and the new sales methodology.

That's changing. The cost of not having it is becoming measurable. Companies are quantifying what it costs to lose context during employee turnover, to ramp new hires without institutional knowledge, to run customer relationships on incomplete information.

At the same time, the technology to capture knowledge passively has advanced significantly. On-device language models, real-time transcription, semantic understanding, context-aware systems. The building blocks for organisational memory that doesn't depend on human discipline now exist.

The companies that figure this out early will compound an advantage. Every week of captured context makes their teams more informed, their relationships deeper, their decisions better grounded. The ones that don't will keep losing what they learn, one day at a time.