Why AI Needs Knowledge Management to Succeed

Mark Smith: 0:19
Welcome to the MVP show. My intention is that you listen to the stories of these MVP guests and are inspired to become an MVP and bring value to the world through your skills. If you have not checked it out already, I do a YouTube series called how to Become an MVP. The link is in the show notes. With that, let's get on with the show. Today's guest joins me from Stockholm, Sweden. She's a founder of Knowledge Management Expert at NoFlow Value and was first awarded her Microsoft MVP in 2024. That's after working for Microsoft for 15 odd years. She's known for driving significant business value through Microsoft 365. You can find links to her bio and socials in the show notes for this episode. Welcome to the show, Rebecca.

Rebecka Isaksson: 1:13
Hi Mark. Thank you, I'm very excited to be here. I'm usually on the other side of the microphone, so tell me about your podcast. Yeah, it's called Knowledge Fika. I'm not sure if you're familiar with the concept of Swedish fika.

Mark Smith: 1:31
No, tell me.

Rebecka Isaksson: 1:33
It's similar to what Americans would refer to as water cooler conversations. Right Back in the day it used to be in most offices you had a 10 o'clock in the morning coffee break and you had a three o'clock in the afternoon coffee break when people would just stop working, go to the break room, you know, sit down and have casual conversations, sometimes work related most of the time not, but it was also, you know, I know my colleagues. When I moved to America, when I moved to Redmond, they were kind of laughing at this and they said so that's just you slacking off work, right, getting paid to have coffee.

Mark Smith: 2:15
And I'm like.

Rebecka Isaksson: 2:16
Well, you know what it's actually more than that, because those casual conversations you connect with people on a personal level and that has some magic effects on working relationships. You make new connections, you meet new people, you realize that you have common challenges or common solutions in different departments and you know it becomes very creative and innovative and also it's very good for morale and well-being. So me and an American woman who lives in Sweden since many years got connected on LinkedIn because she said she reached out to me and she said, oh my God, I'm so excited there's another knowledge management person in this country. Wow, we should connect. So we had an informal coffee and then, you know, the rest is history. We started doing this podcast and you know, similar to this one like fireside chat, casual conversation, talking to interesting profiles from Microsoft, from Microsoft partners and, you know, mostly centered around knowledge management, ai or the combination of those.

Mark Smith: 3:33
So interesting, particularly the. I want to understand the fact that you've had such a long tenure in SharePoint, and I've seen some stuff around AI, document automation that Microsoft's talked about in recent years. Then there's, you know, organizations that have a tremendous amount of data sitting in their SharePoint environments, and now along comes AI and, to make sense of it all, what are you seeing is kind of, where are these technologies merging? Now? What are you kind of seeing day to day where AI and then knowledge management is starting to cross over?

Rebecka Isaksson: 4:20
Well, I think that AI needs knowledge management, not necessarily all the dry academic theoretical frameworks and you know processes and things like that, but the foundational principles of knowledge management, because that's all about how we govern, how we manage lifecycle.

Rebecka Isaksson: 4:45
A lot of people unfortunately say the words knowledge management but what they're actually talking about is content management or actually even document management.

Rebecka Isaksson: 4:56
Right, but knowledge management is more than that, because we have this whole body of what's called tacit knowledge where, first of all, I think AI and the capabilities we have now in M365, with AI reasoning over so much more of the knowledge in our organization than just what's formalized in a document, it can really help some of the problems that knowledge managers have battled from, you know, the very beginning about 30 years ago, when knowledge management started evolving as an art and a science.

Rebecka Isaksson: 5:40
But I also think that the reverse applies AI can help knowledge managers and knowledge management solve problems. But AI, to be super efficient and for organizations to reap the benefits and get a faster ROI, knowledge management principles are needed, but it needs to be expanded and we need to shift our focus, I think, in this space from enterprise content management to enterprise knowledge management. We need governance. We need governance, we need lifecycle, we need RBACs or role-based access controls on the less tangible assets as well Our chats, our videos, our audio, all of that, email calendars, even right. We need to think governance and lifecycle in much broader terms.

Mark Smith: 6:35
But you mentioned RBAC and you know, know, I've done presentations where I talk about sharepoint's lack of rbac and um and therefore being a risk vector inside an organization, particularly in an ai driven world. How do you talk about uh, you know, rbac from a sharepoint point of view?

Rebecka Isaksson: 6:54
yeah, I guess I have to be careful now because I don't want to lose my status as SharePoint MVP, right. But no, I'm just kidding. Security permissions in SharePoint is not easy for the most experienced of us, right, and there are people who are on the deeper, technical level a lot more. You know a lot better and more experienced than I am on that, because I've always kept more at the business value, solving business problems, the functional, that's my you know niche. So it is convoluted and it is complex and it is complex. But if you don't have any kind of you know, roles and responsibilities matrix in your organization, you can't really have the governance, you won't be able to control it. The sharing links is probably the biggest course of that, because a lot of organizations do not think through who can create a site, who can share a site, who can share what document library, et cetera. So they grow organically and you have, all of a sudden, you have this forest of sharing links right, left and center, orphan sites, with what we usually call site sprawl and things like that, which is also caused to some extent by the fact that organizations don't understand what happens in SharePoint on the back end every time you create a new team. Right, so it's. But not to say that that's, you know, the fault of SharePoint or SharePoint design flaw. I think it's in human nature.


Rebecka Isaksson: 8:51
We're very good at squirreling things away, but we're not very good at tidying up and throwing out. It's you know? I just moved into a new house a year ago and for six months before we moved I said to my husband you know the 20 boxes you have in the attic that has stuff since you were 17 years old. Please go through and clean them out. What happens in the end when you move houses? Well, the night before the movers come, you realize you haven't even gotten to those boxes yet, so you move them from one attic to the next. Organizations do the same with content. Cleaning up, organizing, it's not fun do the same with content.

Mark Smith: 9:31
Cleaning up, organizing it's not fun, and, after all, it is called SharePoint. It was intended for sharing information inside the organization. I'm keen, though, to get an explanation from you, because I've not gotten any experience in this area. You talked about the difference between content management and knowledge management. How do you explain it?

Rebecka Isaksson: 9:46
The content management side is the explicit knowledge, the tangible assets, where we've codified or, you know, most cases.

Rebecka Isaksson: 9:55
We usually refer to it as the knowledge that people carry around in their heads, which is actually experience, more so than traditional knowledge that we've absorbed through education or through reading books, et cetera.

Rebecka Isaksson: 10:28
It's the know-how that we get from repeatedly doing something, learning from our mistakes, et cetera, and that's very hard to capture.

Rebecka Isaksson: 10:41
Well, this is where, sometimes, I have a little bit of a debate, shall we say, a polite disagreement, with some of the very leading, most prominent experts in the field of knowledge management, who argue that you can't codify tacit knowledge because once you write these things down, they're no longer knowledge. They are now static assets that are subject to interpretation by the reader. They become information, and I agree with that. What I mean when I say codifying tacit knowledge is not to write it down, to transcribe it, but what I mean is that AI has the capability to draw from information, whether it's in an engaged community conversation, whether it's a team chat or a team channel right chat or a team channel right, so we can pull the nuggets out of those conversations best answers in Engage, for instance and it will incorporate that, as it generates new content for you. So that's a way of codifying the tac set without killing the knowledge that turning it into information so, so interesting.

Mark Smith: 12:14
I I used to work for a company that started with I and ended with m, and you know they worked on some big you know um, you're talking about how now now right, yeah some big petroleum-type customers right that extract oil, petroleum and things like that.

Mark Smith: 12:34
And one of the things that they found with AI coming up about is that sometimes these systems had been built 50 years ago but they'd been built by engineers who have a massive attention to detail and to writing notes and things, and so they're able to extract all these schematics and things of, let's say, a massive oil rig, feed that in when this oil rig was now having challenges and those people obviously no longer work for the company, they're retired, they're gone, et cetera.

Mark Smith: 13:03
But with AI I was able to bring down that jot on the blueprint to oh note. This was able to bring that all to life and it was kind of it was tacit knowledge in that it was you had to know it because you were there, type thing. But because of those notes now, 50 years later, had a whole new lease of life and created a whole new level of value for the organization to be able to move forward and, and you know, fix these things for the, the current engineers that never had that connection, but that was a an organization obviously that was built around engineering and and, by nature, codifying into drawings and things like that information. That's not most companies.

Rebecka Isaksson: 13:52
No, it's not right. And especially not if you look at, you know, organisations that predominantly consist of knowledge workers. Right, engineering, these types of things. Because there's formulas, there's structure, there's you know, procedures, it's things are very well defined. When you're in a knowledge worker organization, let's say a professional services, a consulting organization, right, things are not that cookie cutter, so cookie cutter. So there's where you have a huge value from tacit knowledge. One of the I mean one of the predominant vehicles and actually one of the core pillars in knowledge management academia is communities of practice. So communities in Engage, basically, right, engage was and it's a pet peeve of mine when people say Engage aka Yammer, because Engage and Yammer are night and day, right platform for communities of practice, which we're now seeing, which would all the improvements that are made, all the new features around subject matter, experts, sme approved answers, et cetera, and how CoPilot will actually put more weight on an answer in Engage that has been given the check mark of SME approved or best answer, right. So that's been the only way to really scale tacit knowledge and exchange and what it is. It's conversations like these, right?

Rebecka Isaksson: 15:44
When I worked at Microsoft Consulting Services in Redmond for Global HQ, I was part of the office of the CTO and we ran the knowledge management including content and collaboration platform, built on SharePoint, of course, and the communities of practice, and we had around 105 different communities of practice. We had over a thousand community leads and SMEs and, on average, we would have between 43 and 45,000 unique members engaged a month, right Month over month, in these communities. So they were an incredible way. For, you know, a consultant is working on client side. They might not even have access to their own laptop, they can't access any of the knowledge bases or resources in the Microsoft tenant, so but what they could do was they could always throw out a question to the relevant community and we measured all of this. We had a lot of data. We measured response times, number of responses, all of these things, and we saw that most of the time when somebody asked a community for help with a customer problem, they would get collateral. They would get detailed recommendations and instructions of how to tackle a certain problem within minutes. So you know that drove up customer satisfaction, but it also helped the consultant who was battling a certain problem that they hadn't encountered before. They learned. So now they could be a better. You know they could provide more value for the next engagement and so forth.

Rebecka Isaksson: 17:45
But it's very hard to motivate to certain. You indirect and there's a ripple effect over time. Because even if 80% of the community members never participate and share or ask questions, they're reading these conversations, they're seeing the collateral that's being shared. It's hard to quantify, but now we have Copilot, for instance, that can actually help pull out these nuggets and incorporate that in content that is then, you know, created and widely distributed. So you can now all of a sudden scale this. It's still hard to measure, but you can really start scaling tacit knowledge sharing and tacit knowledge exchange.

Mark Smith: 18:28
You can really start scaling, tacit knowledge sharing and tacit knowledge exchange. I like that. I've learned a lot. Tell me, how did you become an MVP, like? You've obviously been in Microsoft for a long time? Were you aware of the MVP program when you were in Microsoft, or were you more aware of it when you came out of Microsoft? And I know that you went in and out of Microsoft a few times? But how did becoming an MVP at this time come about for you?

Rebecka Isaksson: 18:53
well, yeah, I left in October 18 because I wanted to go back to Sweden for, for personal reasons, family, friends, etc. Um. So I left the company for a short while, but very quickly. Within six months I was asked to come back to work for product marketing in M365 on this new and exciting AI product Project Cortex, which became Syntex and Beaver Topics. So I was aware of the MVP program at that point. I didn't know about it for my first 12 years but when I was in product marketing because, as we were preparing the product for general availability or products as it became, we worked very closely with the MVP community and we had MVPs who worked for organizations that were part of several private previews that you know work very closely with marketing and with engineering to refine and fine tune and perfect the product before they were released. So I was aware of it. I never thought of it myself that, you know, I should try to become an MVP, but because I'm a communities person at heart, I enjoy helping people, I enjoy sharing my expertise and experience, not necessarily expecting a return for it, and that's why I publish so much on LinkedIn and share so much on LinkedIn, for instance, because I like to contribute to the bigger community.

Rebecka Isaksson: 20:38
So I started speaking at these different events like we have in Europe and North America collab days, for instance, which used to be SharePoint Saturday. So I started, you know, speaking at some of these and I met some of the MVP OGs, as they call them, some of the folks that's been in this. Eric Ships is one of them the SharePoint cowboy. You might have heard of him or your listeners were into SharePoint, probably know who he is and Sue Hanley and Mark Anderson and all these guys. And that's how the idea came about, because some of them said, you know, we should probably nominate you because we actually need more people that has the business value, business problem focus, and not only the technical experts, developers, admins, right. So I just and I thought, well, okay, that would be nice, but you know it was still so I. That's how it came about.

Rebecka Isaksson: 21:40
I just started attending more of these Collab Days events and speaking at all these things and then someone in the UK nominated me after the Collab Days, bletchley Park last year and yeah, and it was approved. So I was very of your listeners. I feel like 15 years as an employee. I've never had as much impact and close interactions and conversations with Microsoft engineering than I do as an MVP, and I'm very humbled by that fact, but I also thrive on it. It really motivates me because my personality is just I like to have an impact. I'm an impact junkie, you know. Impact and influence for improvement, making things better, is kind of what makes me jump out of bed in the morning.

Mark Smith: 22:49
Hey, thanks for listening. I'm your host business application MVP Mark Smith, otherwise known as the NZ365 guy. If you like the show and want to be a supporter, check out buymeacoffeecom forward slash NZ365 guy. Thanks again and see you next time. Thank you.