Why knowledge management is broken

And how AI can fix it...

Many of my articles until now have been focused on productivity at the individual level. This month, however, I’m switching gears to look at it from an organization and company standpoint.

Organizations suffer from many of the same productivity challenges as individuals do... except they’re compounded. The more people that get involved, the more opportunities there are for productivity breakdowns. This is the same for knowledge management.

What is knowledge management?

Established in the early 90s as a discipline in its own right, knowledge management is defined as the “process of creating, sharing, using and managing the knowledge and information of an organization” and achieving organizational objectives by “making the best use of knowledge.”

While this concept sounds simple in theory, it is less so in practice. In fact, it could be that no organization -- from small businesses to enterprises -- has truly managed to crack the code on knowledge management. Knowledge management is simply broken.

Why is knowledge management broken?

When we look at the stats, we see that 19% of everyone’s workday is spent looking for “stuff”. That could be searching your emails, hunting for a file, or sifting through items on your desk. Moreover, 44% of the time, we can't find the information we’re looking for. That’s a lot of time wasted and a lot of frustration.

When we extrapolate this over a team, department, or entire business, the problem gets even more complex. Just as everyone has different working styles, they also have different knowledge management approaches. Some people are hyper-organized, others are barely holding it together, and everyone else in between has a different way of creating, sharing, using, and managing information. 

Throw the cloud in there, along with all the productivity platforms, and it’s easy to see how knowledge management can quickly get out of sync.  This can result in individual frustrations, organizational inefficiencies, business risks, and stress.

We’ve tried to fix this problem...many...many times

Centralizing and managing knowledge has been ‘solved’ many times over. When the category of Knowledge Management first emerged in the 90s, the discipline spurred many new technologies to ‘solve’ the challenges of managing knowledge within the enterprise. This included tools like enterprise asset management systems, document management systems, collaboration tools, enterprise portals, and workflow systems. 

As we moved into the 2000s and cloud storage became more widely accepted, the internet boom created more data, mounting an urgent need for tools that made it easier to find information.

Enterprise search emerged as another ‘solution’ to this problem. Enterprise search started in the 1970s primarily in academia but slowly made its way into the consciousness of business in the 90s. Many startups built businesses around enterprise search until the industry consolidated in the 2010s and Dassault, IBM and Oracle scooped up their flavor of enterprise search. 

If you go to any larger organization, they will likely have one of these enterprise search solutions but if you ask any employee how much they love using it...crickets. 

Why is the knowledge management challenge at a tipping point?

Despite the fact that we’ve tried to fix the problem with various solutions, I would argue that knowledge management hasn’t gotten better with technology, it’s only gotten worse. 

Plus, with more emphasis on collaboration, remote and hybrid teams, a tool for everything, we’ve grown the complexity of knowledge sharing exponentially. 

Remote work over the last year has highlighted this problem for every organization both large and small. . No longer can we lean over our desk and ask a colleague for a document. No longer can we walk to the company’s bank of filing cabinets and pull out the file we need. 

Now, we have to ask a colleague over Slack, try to use a half-baked global search tool, or explore with trepidation the cloud-based world of Google Shared Drives, Teams, or DropBox. Many times, we actually end up creating more content (like Slack messages, emails, etc) just to find where that one piece of missing content is. 

Yes, I think we’ve hit peak content chaos! 

With content siloed, some large organizations have tried to revive their old enterprise content management tools, but if you ask folks at these organizations how it’s going for them, you’ll probably hear “not very well”. 

More platforms, apps, and tools aren’t solving the problem. Instead, we’re in a situation where we’re creating lists in Notion to track Google Docs, and Google Docs to track workspaces in Notion.

How can AI fix these problems?

All of this to say: it’s time for a better approach. As someone who spends their days immersed in the world of AI, I see opportunities for this everywhere. Here are a few ways I think AI can do a better job than content management systems (or humans, for that matter) can at managing knowledge:

1. Embrace “find” versus “search” - I’ve written on this topic previously, but to summarize, AI has the capacity to ingest and understand huge amounts of data, paving the way for true “find” capabilities, versus merely “search” ones.

2. Break up with folder structures - Our brains have been conditioned to think and dream in folder hierarchies. That means, traditionally: A top-level folder, followed by sub-folders, containing even more sub-folders. Every document has its place, and it can only live in one folder at one time. It requires the cognitive ability to understand a document’s content and then put it in the right place. However, inevitably, not everyone sees eye to eye on the “right” folder structure, and each person has their own preferences or requirements for working. This is where AI is a game-changer. Even when information is spread across multiple platforms, AI enables organizations to pull together what’s needed, when it’s needed, and scrap traditional folder hierarchies.

3. Say hello to automation - Organizing content has typically been a very manual process. You must first read a document to understand it, then hunt to see if an appropriate folder exists to store it in, and finally, move the document to that location (or create a new folder first if none exists). This doesn't sound like a lot of effort, but it all adds up to time and brainpower wasted. AI changes this because it can scan the document’s content for you, generate appropriate meta tags to classify and store the document, and then quickly find it for you when you need it again. 

It’s exciting for me to see Charli moving in this direction of helping teams. Our AI will soon have the capabilities to mobilize smart knowledge management at the organizational level, making it a game-changer for organizations big and small. Charli will let individuals work as they prefer while keeping content organized, quick to find, and easy to share at the organizational level. 

👉 Stay tuned for more announcements on Charli for teams in the coming weeks.