Uwan Mukbong:
You're touching on something interesting, which lands to the cultural adaptation and a cultural change where you're beginning to talk about cross-functional teams needing to come together to start to think differently, to start to think about how we can get teams to innovate.
So what do you do on that aspect as it comes to that cultural change and adaptation there, where you're talking to your leader, but then how do you take that down below the permafrost layer that may block things as you move forward there?
KeyAnna Schmiedl:
Yeah, so listen. I do not pretend to believe that I know for sure exactly where our people are at at any given time. And so I insisted to our leaders that we just don't know. And so we sent out a survey. And we sent it out to everybody across the company and we asked questions as broad as, "What do you think about Gen AI?" To, "What's your level of understanding of it? As you think about your work, how do you think it might help or hinder you?"
And we got all of these responses back and we just looked at that information. And from that we took those themes and we arranged sub councils to the AI Council to then be able to tackle things like, okay, we should be thinking about ethics and governance. We should also be thinking about, well, how do we operationalize this? There are some opportunities for learning and development. But as we kind of went across all of these sub councils, what we really pulled out of that information and those data were that people had a little bit of a misunderstanding, or at least they were conflating the idea of Gen AI and the idea of just straight-up automation.
And so we were hearing things from folks that we were like, "Oh, that's just an automation solution. We could do that now and we probably should have done that a few years ago."
So it allowed us to also have a conversation about what is our level of understanding here? And that's where we started the benchmark for creating a series of, there are four courses that we have, where one is just generally how to understand AI at Workhuman. And then each course builds on itself so that, depending on where you are and your level of expertise or understanding, there's a course that makes sense for you to understand a little bit more and build on that next step.
But it really came from first hearing from our people. Because without that, we maybe could have landed in the same spot, but I guarantee you we would've had a lot more missteps along the way. And we also would've missed the opportunity to identify a person who responded to the survey and said, "Oh, I teach a master's level course in Generative AI Theory."
And I was like, "Come help us build the learning, maybe facilitate a course or two."
But getting people to showcase to us what their level of expertise was and how in some cases that exceeded our own, I think also allowed us as leaders to realize we really needed to partner with our people to generate the right solution.
Uwan Mukbong:
Just to clarify, this was someone in Workhuman who was-
KeyAnna Schmiedl:
That's right.
Uwan Mukbong:
... Teaching a class externally in the university on Gen AI.
KeyAnna Schmiedl:
That's right.
Uwan Mukbong:
Interesting. Amanda, what's your perspective?
Amanda McClerren:
Bayer is a very large company. We have three different business divisions. We've got over 150 years of history, and that can create an opportunity to get a little bit stodgy perhaps. So we're actually exploring a new system right now called "Dynamic Shared Ownership" at our company. And the idea is that we're going to have many fewer layers of leadership, we're going to have larger spans of control, and we're really going to push decision-making down into the organization. So kind of extreme peer-to-peer and team-to-team accountability. And it turns out when you do that and you give teams, empower teams, the right access to data and information, they really can make decisions more quickly.
But the other thing that we learned is that that radical transparency of information and insight is super important. So the role of digital becomes even more critical. We have to help the business have that transparency to understand how the business is performing and how the decisions that they're taking are impacting the P&L.
And so that's been a lot of the work that we're focused on right now, just really early days, but I think there's a lot of credence to the system and now it's about getting everything operating together.
Uwan Mukbong:
Absolutely. Very good. Ira?
Ira Rubenstein:
Yeah. So look, digital change is hard. It's not easy for companies that are historic like Bayer, or even a PBS, we've been around 50 years, a traditional broadcast entity and moving into this digital media streaming landscape, which has required us to completely shift. The way I've approached it is trying to help people stay focused because there's so much noise and so much distractions, and doing that by sharing a vision of, I call, my digital core beliefs, which really comes from actually my daughter who has anxiety.
And I learned that when you have anxiety, you have these core beliefs that drive it. And I realized I had digital anxiety at work. And that digital anxiety was when we weren't being consumer focused, we weren't building to scale, we weren't embracing change, we weren't embracing data. And so sharing that with the team definitely helps. But because there's so much work to be done in building our apps and doing our marketing and everything else just goes into the day-to-day, we're fortunate that we have what's called an innovation team.
And I've had that at other organizations I've worked at before. And that allows the team to deep dive, rapid prototype on whatever the shiny new object is. And that keeps the core team focused on the change you're driving without getting distracted. And this innovation team has worked very closely with AWS and we're thrilled with the work. We work closely with Bedrock. They did a rapid prototype over a year ago now, maybe two, on a recommendation engine that powers all the PBS platforms now. When you go in, it knows what you want to watch.
They're working right now with the AWS Gen AI innovation team on a project I'm super excited about. But that's the rapid prototype. And if it becomes a product, great. Then the rest of the team can embrace it. And that drives a change, but it keeps people focused on the day-to-day of what we need to do to move the organization forward.
Uwan Mukbong:
Now, that's fantastic. Now, I like the comments that you stress that point about really that leadership engagement and organizational curiosity.
So if we think about the cultural adaptation, we've solved the cultural adaptation, leaders get it, everybody's on board, you've got cross-functional teams beginning to work together. What have you seen? Give me an example of something that's come out of it based on what you've done so far.
Amanda, any perspectives there?
Amanda McClerren:
Sure. I think it's really important to have a good story around this. And so I'll tell you my story, my origin story, my first exposure to data science, which fundamentally changed how I think about problems ever since then.
In my story, we had many, many years of data in our seeds pipeline on the genetic information of each of our seed lineages. So think about the grandparents, the parents, the children, the progeny. And we also had a lot of information about how those seeds performed when they were planted in the field, data that we collect outside once a season.
And so our data scientists were able to develop an algorithm that could predict the performance of that next generation of seed based purely off the genetic information that we could collect in the lab. And that was an incredibly powerful innovation because it allowed us to shave a full year off of the product development life cycle.
And it also allowed us to sort of fundamentally change the scale of our pipeline because now we could perform that really important yield testing in the lab all year around versus once a year in the field. And so it's had an incredible impact on our ability to create better innovation faster for our farmer customers.
Uwan Mukbong:
Fantastic. Good example there. Ira?
Ira Rubenstein:
So trying to think of cross-collaboration among teams. So I guess the best example I have is when we were trying to figure out live linear streaming in the broadcast model. And that involved a lot of the lawyers, programming, et cetera to make sure we had the rights, but it also involved leveraging AWS and the elemental servers.
And within three months, we were able to bring all 300 stations up to live linear streaming. But it involved, not that we had to get the lawyers and then the technical, but they had to understand what we were trying to build and then making sure we had the rights and to configure all that.
Uwan Mukbong:
Interesting. KeyAnna?
KeyAnna Schmiedl:
Sure. So I'm going to give you three examples. One of them has a story. So I previously mentioned my team being customer zero. We're customer and we're prospect zero. We make HR tech that tries to make a better working experience for all humans. And so I have got to partner directly, and also via my team, with our product teams, with our engineering teams as they're brainstorming ideas for the next new launch, for the next new innovation to say, "Okay, I love it in theory. Let's try putting it into practice."
Because anytime you have a new technology, the most unpredictable thing is the human element. And so let's just get a few examples of the humans humaning around the technology and then figuring out what unanticipated something is going to happen. And in some cases, that's a discovery that's even better. Or you learn actually this whole section is unnecessary because people jump from point A to point Z.
And so those are really fun conversations and that's also where my team gets to get out of the day-to-day thinking about people work as more traditional and get into the experimentation, the exploration, and the play that makes it more fun.
The second example I had previously mentioned. We started this AI Council and we had these different lanes. So from a cross-functional standpoint, you have governance and ethics and sure, yes, you have legal involved, but we also have a Workhuman IQ team. And they were heavily involved hand-in-hand saying, "Here's how I think we should start crafting and thinking about this."
But for me it was until we understand how our folks are already using tools that exist out there, I don't want to box them in into what they can and can't do. I want to first understand what they're trying to do, to your point. And then say, "How do we ensure that they can keep doing those things?" And that's how we started crafting that policy.
Similar for PR. As we go across all of the different lanes. But also what we did was we identified the importance of tacking those culture ambassadors, those AI first ambassadors, identifying at least one person across every single department and saying, "You're going to get that exposure first. You're going to get access first, and you're going to be the one who then helps your teams along," so that they have somebody right there who's in the day-to-day with them that can help as soon as they have a question.
And then the last one is we started AI hackathons about two years ago. And so this last year, one of the ideas that were submitted came from our customer service team. So these are our call center operators. And they had this idea around how they thought AI could really help them deliver a better customer experience.
And so they submitted their idea. For all of the ideas, we paired up engineers, product developers, Workhuman IQ specialists, and they would vet the ideas, go talk to the teams, and then help them build a prototype. And with this team in particular, they built a prototype. And what it did was that, as soon as a phone number came in that the center recognized, it would then go into the history of every time that person is called, what was the most recent call that that person made, and then what's the summary of the issues that this person has? So it eliminated the call center operator from needing to say, "Can I put you on a brief hold while I review your customer notes?"
How many times have we heard that and gone, "I'm going to lose it."
And so now they don't have to do that anymore. And so the unintended consequences there was that that's a very high turnover role for us because it's a low paying job, it's a 24/7 center. And the turnover is high because you don't like having to deal with an angry person on the other end of the call. They're typically not calling to tell you you're doing a good job.
And so, because they implemented this, we're actually retaining more talent, they're having a better experience, but our customers are also having a better experience. And this technology has won awards, as well.
Uwan Mukbong:
Very good. And I like the culture ambassador piece there that you talked about.