Follow-on blog: why now is the knowledge management moment

The transformative potential of large language models (LLMs) like ChatGPT is undeniable. Unfortunately, few organizations have the foundation in place to maximize their value.

The biggest blocker isn’t the tech itself. It’s the quality (and accessibility) of organizational knowledge. As Liz Fosslien, Atlassian’s in-house expert on making work better, recently put it, teams today have “more information than ever, yet they’ve never been less informed.”

Why the disconnect? Because scattered, outdated, or siloed knowledge doesn’t just slow teams down — it turns promising AI initiatives into expensive disappointments. In a recent Atlassian webinar, Fosslien sat down with Julie Mohr, Guest Speaker, Principal Analyst at Forrester, to unpack why the generative AI boom is putting pressure on companies to get their knowledge management (KM) house in order.

They landed on three key takeaways that leaders should have on their radar:

1. The value you realize from generative AI hinges on the quality of your knowledge.
2. Knowledge is much more than just documents — it’s all the ways your teams communicate.
3. Collaboration and culture are still the backbone of effective knowledge management in the AI age.

Let’s dive into why knowledge management deserves a spot at the top of your strategic agenda.

Generative AI shines a spotlight on knowledge quality

The excitement around generative AI in the workplace is palpable. Companies are eager to train LLMs on their internal data to power everything from customer support bots to HR assistants. But Mohr warns that rushing into AI without solid knowledge foundations is a recipe for disappointment. “What makes the LLM run is that knowledge,” Mohr explained. “You don’t get the power of the AI without the quality of the knowledge.”

Think of AI as an amplifier. Give it solid, organized data, and you’ll unlock meaningful value. Feed it outdated, scattered, or inaccurate information (the kind most companies are already dealing with), and you’ll quickly see diminishing returns — or worse, dreaded “hallucinations” that undermine your efforts.

The hidden costs of poor knowledge management extend beyond AI. Consider how much time your teams spend just searching for information — looking for a report, login, or piece of code. It adds up more than you might think. According to Atlassian’s 2025 State of Teams report, employees and executives alike estimate they spend about 25% of their work week tracking down information. Over a year at a 10,000-person company, that equates to 5 million hours lost to fruitless searches. 

When knowledge isn’t easily accessible, teams make slower decisions, duplicate work, or miss key facts — all of which drag down productivity. And when AI tools are layered on top of that chaos, the consequences scale just as quickly.

So, before you dive into generative AI, ensure your knowledge ecosystem is clean, connected, and collaborative.

Knowledge is more than documents

When you hear “knowledge management,” your mind might go straight to wikis, SharePoint sites, or overflowing document folders. But in modern organizations, knowledge isn’t confined to documents. It’s flowing through Slack messages, Zoom calls, documents, powerpoints, and countless informal conversations that never make it into a formalized knowledge base. 

That’s a lot of valuable insight that can easily slip through the cracks. As Mohr put it, “knowledge is a continuum” — and too often, companies opt for the wrong medium. “They’ll put it in a PowerPoint when it probably should have been in a document,” she noted.

This issue is magnified when generative AI enters the picture. LLMs need the right data in the right context to perform at their best. If critical details are locked within meeting recordings or buried in someone’s chat history, these tools won’t be able to access them, constraining the value they can provide.

Leading organizations are rethinking how knowledge gets captured — and making it part of the day-to-day:

• A key takeaway from a Zoom call gets logged in Confluence.
• A Loom video is embedded alongside project updates so the full context is front and center.
• An off-hand insight in Slack becomes a quick FAQ entry or gets added to a decision log.

When knowledge capture becomes part of how work gets done, it’s both easier to find and becomes a high-quality input for generative AI tools. The result is a shared organizational brain that’s accessible, practical, and genuinely useful for employees and AI alike.

Collaboration and culture: The human core of knowledge management

You can have the best tools and processes in the world, but if the human element is lacking, your KM strategy is going to fall flat. Generative AI thrives on context, and the richest context comes from your people.

Think about it: When you’re stuck on a problem, is your first instinct to scroll through a lengthy wiki or reach out to a trusted colleague? For most people, it’s the latter. There’s a natural human tendency to turn to one another for information. “People still prefer to talk to their buddy,” Mohr explained. “Conversations are active, engaging, and memorable.”​ But unless we capture those insights somewhere accessible, they’ll vanish the second the conversation ends.

What does that look like in practice? It means treating knowledge-sharing as a cultural norm. If someone asks a great question in Slack or surfaces an insight in a meeting, don’t let it disappear. Jot it down in a place others can find later — whether that’s a shared FAQ, a Confluence page, or a running doc of lessons learned. These bits of day-to-day knowledge might seem insignificant in the moment, but they’re the connective tissue that helps generative AI tools provide responses that are relevant, accurate, and grounded in how your company actually works.

Getting there takes more than just good intentions. It means breaking old habits — like the idea that hoarding information creates job security — and replacing them with a new mindset: Knowledge sharing makes everyone better. As Fosslien noted, Atlassian’s own culture of openness (one of our core values is “Open company, no BS”) means that new hires (and new technologies like generative AI) can access a wealth of information from day one instead of finding everything locked behind permissions​.

How do you create that kind of culture? Start with the basics:

• Model the behavior. Share what you know, even if it feels small.
• Don’t penalize mistakes that come from thoughtful risk-taking or learning.
• Encourage teams to create “evergreen” project pages or playbooks in a tool like Confluence and recognize those who contribute to them.
• Incentivize knowledge-sharing by making it part of performance reviews or by celebrating top contributors to internal knowledge forums. Even small practices, like ending a project debrief with “Let’s write up our lessons learned and key decisions,” can reinforce the habit. 

And, of course, give your people the time and space for this “work around work.” If every minute of their calendar is accounted for, they won’t have the breathing room to document or share. Treat knowledge activities as a first-class part of work, not an after-hours nice-to-have.

Seize the moment to transform knowledge work

Knowledge management used to be an afterthought (“we have SharePoint for that, right?”). Not anymore. In the age of generative AI and distributed teams, KM has moved to the forefront — and the organizations that act now will reap the greatest rewards. Mohr summarized that as companies embrace AI and expand their knowledge practices, they must focus on a few essentials: data quality, diverse knowledge formats, and human-centric collaboration.

Get those right, and you create a powerful engine for learning and innovation. Neglect them, and you’ll pay the price in wasted time, frustrated teams, and unrealized potential.

Interested in learning more? Watch the whole webinar, on-demand, now.