How Coupa Drives Internal AI Transformation
Coupa Chief Digital Officer Todd Joseph joins Colleen Call at Coupa Inspire to discuss how the company is using AI internally to change the way employees work, learn and collaborate. They examine why AI literacy, workflow redesign, community sharing and champion networks are critical to moving beyond tool adoption and creating a sustainable operating model. Joseph explains why internal AI transformation should be treated as a people and ways-of-working initiative, not simply a technology or cost-reduction project.
Transcript
Hello, we're back, Techstrong TV. I'm Colleen Call. We're at Coupa Inspire, and our next guest is Todd Joseph, Chief Digital Officer of Coupa.
Hi, Todd. Hi. Thank you for joining us.
Thank you. It's great to be here. So Todd, we're going to talk more about Coupa's own AI internal transformation.
I know that's a focus of what you do, your role, if you want to just start off with that. Yeah. Oh, that's perfect.
It's pretty interesting right now, just being a digital officer in this time and place. It's such a changing, fast-paced world, and we really sit, if you think about it, at this intersection of technology, operations, and the culture of the company. Mm-hmm.
And it's really not about picking tools or what's the next fad, but it's really about changing the ways of working within an organization. So it's really just about how do we create this new operating model to unlock a bunch of untapped capacity in the organization? And so my organization and my role is to tap into that, create an operating model that sticks.
Mm-hmm. And I think about it in three main areas. Area number one is, what does this mean for me as the individual productivity learner within our organization?
AI literacy is really critical, and how do we make sure that our teams are understanding AI, how to best leverage it, have the tools and capabilities that they need to be productive in their day-to-day work? Mm. And we feel like individuals on their own, that's the minimum bar that we have to set as an organization of where we want to go.
So that's the individual, how work gets done. Then if you take groups of individuals in these teams and their organizations- Yeah ... that's the workflow.
Mm. The workflow heartbeat of the stairstep of how do we redefine how work gets done and transform our company, that's bucket number two. Yeah.
And then the best part is bucket number three, is now we are defining the new ways of working. What are the skills, what are the capabilities, and what is the new workforce that we need to be bringing in or skilling up to, to help keep this sustainable? So my organization helps support and drive all of those areas.
So, that's a major sell. You sell it to your customers- Yes ... and now you have to sell it internally.
Was there any resistance? It sounds like it went smoothly, but was there any resistance? Because having that mindset, adopting AI into your everyday workforce- Yeah ...
your work plans. Let me hear more about that. Yeah.
It's changes, it's just everyone takes change differently. Yes. And the- Change management is interesting.
It is, and there's a one size fits no one on all of those pieces. And one of the most exciting areas is, I think we all learn by doing. Yes.
And so when we rolled out capabilities to our organization, we took that around with training, education, enablement, these Slack channels of community where people can say what they're learning and how it's helping them. Yes. One of the most exciting pieces is once the first month of rolling out our AI platform that allows our employees to work more productive.
So this is, think about the tools that everyone uses on a daily basis, whether there's their own productivity tools, their CRM, their calendar, access to other data. Now just chatting back and forth to have that structured and synthesized in a way for them. Providing that to them, in the first month, we had 60% of our employee base logging in on a daily basis to learn, grow.
To learn. Yes, which is awesome. Well, let's unpack that, or just walk me back, because you said from the first day till, I guess, the launch.
Yeah. What was the timeline to get people acclimated and ready to be an AI native company and just... Yeah, there's never this end state.
The way that we rolled it out was there's always these get ready to learn. Yes. Get ready to pre-work and understand some of the concepts around best practices around prompting.
What access connectors do you have to your daily use? What are some of the peers doing out there? So it was just- Mm ...
kind of just getting everyone to drive awareness. Yeah. But once you roll out the tool, and people are experiencing by doing, I think that became another unlock.
And this other unlock is, here's how I'm learning, here's what I'm seeing. What are you seeing? Mm.
" So what happens is that we continue to raise the water level- Mm-hmm ... of understanding and possibility- Yeah ... with this type of sharing and growing.
So this has taken over three to six to nine months, but it just continues to get richer with how creative everyone is. Well, it seems like you're taking them to a higher level, and what does that mean versus what they were doing before? Have you heard the feedback from them?
Yes. Mm-hmm. So number one, it's like we're removing the execution overhead.
Yeah. This overall manual drag. So when someone gets up, and they're logging into five different systems to pull something together- Mm ...
and that's repeatedly done, you take out the creativity, you take out the judgment, where they're adding the most value for higher level work orders. Mm-hmm. So what we're seeing is, for example, on our customer support team, when we have now an agentic chat solution interacting with our customers-That's giving their time back to make sure that, number one, they're following up, they're having these rich, deeper conversations around their usage and adoption of our product.
Mm-hmm. The customers are more satisfied because they can actually now see faster response time, more comprehensive approach, and it's just a better all-around experience for both our customers and our employees, specifically just leveraging the tools that we have. So you are getting a lot of feedback from real customers that the- Real customers saying- ...
it's more positive? Absolutely. Absolutely.
So not only on the customer support, we have our account executives that are interacting with customers and prospects on a daily basis, and now we're collating information to them, where maybe 30 to 40% of their time historically has been in manual work efforts to pull information together. We're supporting that with a new agent called Sales Buddy. It's their- Oh ...
personal chat assistant that can help them- Good name. Yeah ... yeah, help them just meet the customer's needs in just a consistent, maybe a more higher quality artifact, where the account executive can spend more time in the account understanding the needs and matching that to what- Mm ...
I guess, our product documentation, our product quality, and what we're seeing in other customers to help better position us. Well, it sounds like it's going smoothly, but is there anything that didn't work? Yeah, I think what you have to do is, there's not a silver bullet with AI.
No. You have to continue to leverage it because it learns. Mm.
It's like our knowledge was spread across the emails that we send, the chats that we go into, the documents, and being able to piece that together where it is personal for you- Mm ... because the brain continues to learn. Our AI Coupa brain is every decision that every employee is making or an artifact they're creating.
We want to collate that and let our organization learn more and let everyone tap into it. And that takes time. And the results, even though they're good, they can get better.
So I think the more we leverage it, the more we learn. The more we learn, the more we tune. Yes.
And the more we tune, I think the larger value prop that we get out of it, which is kind of a learning process. It might get frustrating at times because- Yeah ... we all want the speed of information to be immediate.
But the more we use it, we have been really excited about the results that we're seeing. I like the term AI Coupa brains. Yeah.
Do you use that often with- Oh, we've been passing it around because we feel like the conversations that you and I have, our interaction, and others, that's just part of a collective community. And the value is to make sure all of our employees and our customers have access to it. Okay.
And it's not us as individuals trying to capture and retain it. We need to let the system store and present that to us. Okay.
And then wrapping up, because, again, this sounds like it went really smoothly. For any other CDO, is that what we're calling, CDO? It's funny that, because I always thought it would be a CAIO, the Chief AI Officer, but I'm glad they went with- With every org there's a new structure.
With every new structure, there's a new title. Yes, exactly. Yeah.
But for any CDO at a large enterprise like Coupa, what's the one thing you tell them to get right if they're going through the same, what you've gone through with your employees, your staff, your team? Yeah. I think first and foremost is a people change- Yeah ...
not a technology change. Oh, that's definitely- I think we lead with our employees, and we look at it as a ways of working transformation. It's just not a cost play at all for us.
Mm-hmm. This is about how do we enable our employees to be more effective and efficient at the job that they do. And we want to give them the tools.
Mm-hmm. But we also want to show them how to leverage those capabilities. And so there's a really big part, I think, of our job is to be a practitioner, to show by doing, and then also how do you continue to amplify what is really working in the day-to-day of everyone's job.
So just, I would say it's a people transformation- People transformation ... not technology. Yeah.
And number two, it's a champion network that is integrated into your company in each of the different functions that can really live and breathe the solutions and the feedback within each of the different roles. Because at the end of the day, it is you're transforming roles- Mm ... to new ways of working, which are use cases that drive value.
Yeah. And that just is really best done at the grassroots within those individuals' teams. Yeah, grassroots AI innovation.
So I know I said I had a last question, but if you were to give Coupa a grade on its internal, this whole process, what would it be? Honest question. Oh.
Honest answer. I would say it is early innings for every single company. Of course.
And I'm probably biased, but here's what I would say. I would definitely give us an A, and here's- Awesome ... the rationale why.
Okay. Number one, I think it starts with the top down. And I do feel like the investment, the support, starts with Lee, our CEO, and she champions this every day in her message.
Not only do we want to practice it in our product, but we want to bring that to leading that change internally with the own ways of working. So I think number one, it starts at the top. Number two, then I say every employee has taken on the challenge to take the initiative and be proactive to learn, and the results show.
Over 60% are using it daily. Everyone is learning. They're sharing what they have built and how it's transforming their own work.
And so I think if you start at the top down with a great vision and message, the team rallies around that, and they're leaning forward on how they approach it. Mm. It just helps everything in the middle go a lot smoother.
That's fantastic. And Todd Joseph, thank you for sitting down with us. I learned a lot.
AI is something that I've been a big fan of, and I'm learning because I know that people are skeptical, and I blame the movies. Yeah, exactly. AI.
Well, we all are learning. Something new comes out all the time, so. Yeah, but I think there's a lot of good in it, so appreciate you coming, and- Appreciate your time ...
hope you have a wonderful day one of Coupa Inspire. And we will be back. More to come.
Stay tuned.