Matt Ring

All right. Welcome, everybody. Good afternoon. The clock has started, so we will go ahead and get started here. Welcome. Thanks for coming to our talk. It's day three. We've gotten a ton of information, and you're also joining us right after lunch hour, so we're going to do our best to keep you engaged and keep that food coma away a bit.

At John Deere, our higher purpose is to help life leap forward. We do this by helping provide products and services for our customers that make them more productive, more profitable, and able to do the jobs that they need to do in an environmentally sustainable way.

For the next 25 minutes, Amy and I would like to share our approach, our challenges, and learnings from transforming 500 global teams over the last three years: hyper-focus on agility and DevOps, shifting from projects to products, and evolving our organizational culture, and how all of this has enabled our company to run faster and help revolutionize the agricultural and construction industries as a smart industrial.

So who are these people up on stage?

Amy Willard

Amy Willard. Nice to meet you. I have the privilege of leading our strategy and transformation group at John Deere.

Matt Ring

And I'm Matt Ring. I'm a senior product and engineering coach in our Agile Operating Model Foundry. I think of that kind of like a coaching center of enablement.

So that's who we are. This little company called John Deere, people are familiar with it. Our company was founded by a blacksmith named John Deere in 1837. We are a 185-year-old company.

With this DevOps conference, we talk a lot about not just wanting to hear from the unicorn companies of the world, but from the horses as well, so those that didn't start out as digital natives. Fun fact: John Deere, the very first product that we built was used by actual horses.

But just to give you a little better idea of what it is that we do at John Deere, and what's the reason that we get up each and every day to come and do our best work, we wanted to share this quick little video with you.

Video

We started breaking ground by breaking ground. We've always looked to the land for what we need and always made the tools to work with it. We won't stop making what we make better for today and for tomorrow, because at John Deere, what we bring to the table touches lives around the world.

We run for the humble, the hungry, the growing world. It's why you'll see our logo proudly stamped on machines and hats, on arms and in homes, and the loyalty to it passed down through generations. We run with the dignity that makes us Deere. We strengthen relationships. We solve problems, innovating on behalf of humanity and planet.

In service of the people who trust us and the Earth that sustains us, you'll find us where hard iron needs hard data. We don't make gimmicks, fads, or empty promises. We make machines that do more and use less. Machines that make life better. Because we don't run just to move faster than the rest. We run so life can leap forward.

Matt Ring

Thanks. That's the end of the presentation. Thanks for coming.

No, but hopefully that gives you a little better idea of what drives us, why we come to work every day, and the mission that we're after.

John Deere, we have our roots in the Midwest United States, but we are a global company. We have over 100 locations globally across every continent except for Antarctica. We have over 75,000 employees in our company, and our 2021 net sales and revenue was right about $44 billion.

Some of you may be wondering, why is John Deere a technology company? We just build machinery and farming equipment. Stuff for this innovative technology company? We just work in factories and there's tools in our hands. Is everybody coding in a cornfield or something?

Well, when inspiration strikes, you just do it. Actually, this is a picture from my office. That's Walter. He sits three rows of corn down from me.

But no, really, there's more technology in our machinery today than ever before, and it continues to grow. Just like consumer vehicles today. Has anyone shopped for a personal vehicle in the last three to five years? Some show of hands. The crazy amount of technology that's in that.

Just like that, in our equipment we have cameras and sensors to make sure that we're not going to run into something or someone. We have GPS, especially in large fleet operations or job sites, to know where our equipment is around location-finding. And even drive-assist technologies, so like stay-in-your-lane in your cornfield, and adaptive cruise control so that when we have equipment that's going side by side they're able to keep up. And that's just the technology in our equipment.

Just as a side note of that, in 2021 we had over 440,000 smart connected machines, and that continues to go up. But like I said, our technology spans the entire digital experience that we provide, not just in our equipment. So from the web experience to our mobile devices to the embedded technology that's in our equipment, including our e-commerce platform and our parts distribution network. Technology is very much a part of what drives our innovation and what we provide.

In order to do that, we needed to modernize kind of where we were at with technology. So I'll turn it over to Amy to talk about what our state was three years ago and what we've been moving towards.

Amy Willard

Yeah. Thanks, Matt, for the great intro to Deere.

Like any great transformation story, we're going to start with the why, right? This is our view from three years ago. Put very simply, we just weren't able to move as quickly as we needed to, to value. We couldn't meet the needs of our customers as quickly as we needed to, and the market was continuing to evolve and change.

Like many of you, I assume, we had a lot of localized agile efforts, not a lot of effectiveness when you scale that. We had a lot of project teams, very few product teams. They were incentivized to deliver on time, on budget, but maybe not think holistically beyond that project. We had very little customer feedback and customer interactions for engineering teams or even our product owners at the time. We had lots and lots of handoffs between teams, many, many, many steps to get value out the door slowing us down all the time. We had lots of business analysts, lots of portfolio managers managing what we did, and we had lots of managers whose sole job was really to prioritize and give work direction.

So just to make sure I'm talking to the right folks, do any of these sound familiar? Any of you have this in your company? All right. Yeah, not unique to John Deere in any way, but we're excited to tell you what we did about it.

The first thing we wanted to share is how did we want to change the way we work. I'm just going to highlight these real quickly. We really wanted to create a product model that drove our investment and our funding choices and our prioritization. We wanted a strong product management function within our business units, not within IT. That's a critical thing for our transformation, and one of the things that has really allowed us to accelerate forward with alignment with the business partners that we work with every day.

We wanted persistent cross-functional teams that were really organized around those products. So the top three said product three times. That's intentional. That's a very, very important part of our strategy.

Along the bottom, we really wanted to shift from know-it-alls to learn-it-alls, right, a growth mindset. We don't have to know everything before we start. We can get started and learn as we go. We wanted to ensure that all of our employees had access to coaching and training to make this transition, so we wanted to bring them along regardless of what role they landed in. We wanted to insource a lot of engineering talent. We were highly outsourced before. We shifted to a much higher insourced footprint, and we really wanted to significantly increase our cloud and our automation technologies.

That's sort of the where-we-wanted-to-go view, and we knew that that amount of transformation wouldn't happen accidentally. I had to really think holistically about it. To do that, in 2019, we introduced this very simple visual. Only three things, three things we need to transform: what we work on, how we work, and our foundation.

What that really means is: what is the product mindset, the product structure, thinking about value, thinking about how we innovate, thinking about our customers in ways we never have before. How we work: what frameworks do we use? What mindsets do we have? How do we manage our investment? How do we talk about total cost of product ownership? What's our team topologies? What do we want our roles to be moving forward? And our foundation, to us, that means our people and our tech. How do we invest in both of those to make sure we are equipped to succeed?

Generally speaking, a very simple visual that more or less said that we probably needed to change almost everything. Over the last three years, we have transformed almost every element about our 500 teams. The roles employees are in, the work they do, all of that has morphed. We can't tell you everything, so I'm going to share two beliefs that we have, Matt's going to share two approaches we took, we have five lessons learned, and then we'll be done.

The first belief we have: we believed in mindsets. You can actually see some iterations here of a journey that we made. I think Matt actually drew the one over there on the right, while he held the marker, as we tried to figure out how do we take a holistic approach to transforming a product team and all of the pieces of that.

This is the visual we landed on, and what it really is meant to show you is that in all of the philosophies we took, whether that was training, coaching, putting people into roles, we really wanted to think about these four triangles. The product management approach: know who your customers are, the problems they're trying to solve. Are we building the right thing? From an engineering perspective, are we building the thing right? Are we thinking about resiliency? Are we thinking about flexibility in the future? Are we thinking about quality? Are we thinking about the right tech stack for integration in the future?

Thinking about the agility: are we building it with the smallest increment possible to get the most amount of value? Are we nimble in how we operate, and can we change direction if we need to very quickly? And user experience: are we building something that our users can use and will enjoy using?

Our approach was really to tackle all of the product teams, all of the products, with all four of these in mind, and lean into roles along the way. We have, obviously, these aligned to certain leadership roles on the product teams, but make sure that we were holistically focusing on that so that we could accelerate and that team could accelerate further once they left the experience they had with us in our Foundry.

The second belief we have is we believe very strongly in coaching, a very strong focus on coaching. As we mentioned earlier, we have a Foundry. It's our version of a dojo, which is really an immersive experience where teams can come in, and Matt will talk about this in a second, and they can learn about all four of those disciplines holistically as a team working on the real work. It's not training. They come in, they bring their real work, they bring their real problems. They have coaches that are agility coaches, product coaches, engineering coaches, UX coaches that work with them and try to transform the way they work and help them do that with their actual jobs.

As we all know, I think coaching can be hard to measure the impact. It's difficult to know, did we succeed? Our coaches actually mobbed on that, and they came up with this list here of 10 immersion principles. Come into your immersive experience with us in the Foundry. These are the things that we look at from a behavior perspective as a team: is value flowing quickly, freaky fast, through the system? Are there feedback loops? Are learning happening in your team? Is the value flowing in small increments? Are you customer-centric? Are we big and visible with our information? Are we sharing across teams? Is it predictable? Data-driven? Are we experimenting?

We believe very strongly that if a team demonstrates these behaviors, they are far more likely to leave that immersive experience and continue the momentum and have all the other outcomes we want: more value more quickly, more customer-centric, ability to pivot, time to market, all of those wonderful things that we want. This is what we look for in our teams, and most of the teams take these with them and continue to talk about them as they leave. Are we doing these things? What are the indicators of these behaviors that we want to encourage?

With that, Matt gets to live this world. I'm going to hand it back to Matt to talk to you about the actual immersion experience.

Matt Ring

Thanks, Amy.

I want to talk just a little bit more about the immersion and our Foundry experience. I'm not going to go into too much depth in terms of the dojo experience necessarily, because there's other talks that have talked about that, but I want to talk about some of the things that we did that were a little bit more unique.

We talked about this wave immersion program. This was our vehicle to amplify and accelerate these new ways of working across our company, across our product and IT organizations. You can think of it like a dojo where we're bringing teams in for dedicated coaching. But what was a little bit different about ours is what I often see in dojos is that they will bring in a team or maybe a disparate set of teams, and they will do a lot of team-level coaching. What we did with the waves is that we took a thin vertical slice of an organization, so teams and then the leadership of that, and we coached across that vertical slice, all the way from the senior leadership of whatever that organization was down into the individual teams.

Organizationally, we were looking at this scope of approximately 500 teams, about 4,000 technology professionals. We have our little wheel here on the left that shows what our progress has been over the last three years here. If it makes you think that we basically at the start created a Gantt chart and used the word wave slash sprint slash wave 10 times, you're not totally wrong.

But what we did was, when we started with the first wave, we built in feedback loops both within and in between those wave immersions, so we were getting fast feedback from those that we were coaching and how we were working to adjust. Wave three looked way different from wave one, wave four from three, et cetera. We continued to inspect and adapt and sense and respond to what was working and what wasn't working.

The wave program itself on the right there: it was basically 20 weeks overall of an immersion experience. Once we identified the slice of the organization that was ready to go through this wave immersion, the first 10 weeks of it was this prep, we call it. This is where we were helping them with their own reorganization from where they were at to these persistent product teams and the products and services they wanted to have. There was definitely some training involved in terms of the mindsets and the behaviors that we look for, as well as a lot of more product prep. Okay, you got a product. Awesome. What's your purpose? What do you do? Where are you going? What direction do you want to go? There's a lot of that prep there to get them ready.

Once we got into that, then the 10 weeks of immersion. This was very similar to kind of a dojo experience. The teams are bringing in their real work. Coaches are working side by side with them to introduce some of these new ways of working. So the real world, real work.

Then the last part was basically the sustain and evolve. Part of this was we have our center of coaches that were helping with that, but we also had scrum masters, coaches, and then these leadership that are embedded in the organizations that stay and live along with those organizations. They are responsible for also continuing these efforts, these new ways of working, once they exit out of immersion.

That's basically our wave. Like the wheel shows here, we're just about finished with 11 waves, and that will be right around 500 teams by the end of the series.

Awesome. Last thing we wanted to focus on was we wanted to measure for outcomes. I want to avoid agility theater. Prior to 2019, in our Agile Operating Model, we were already focused on measuring some things around quality and security: availability, MTTR, some other measures that were probably introduced with ITIL and ITSM and all these fun things, but also security. Before we had embarked on this, we already had some measures that were telling us whether we were delivering and running safer and better.

With AOM, we introduced some delivery metrics and outcome-based metrics, but also people metrics. Time to market: that's really the overall lead time from idea to release. Functions delivered, but also deployment frequency: not just getting more stuff out the door, but are we doing it in smaller batches? As well as the people measure, we wanted to focus on NPS. We measured team NPS and org NPS. We wanted to make sure that our employees were engaged and we weren't burning them out in digital feature factories. And also the engineering ratio, to measure how we were doing in terms of that strategic insourcing of our folks.

In 2021, we recognized, hey, we're focused on how we're working, but are we actually delivering the right thing? That's when we started to invest a bit more on the value side of it. We leveraged things like OKRs as well as product KPIs, so key performance indicators, one of the things that we believe would be valuable for our products and services. User satisfaction was kind of that lagging indicator that we used. By doing the things that we think will improve, is our customer satisfaction actually improving?

Three years in, we've had some pretty good, impressive results here. Time to market overall for the 400-odd teams that we have so far measured, we've seen a 60% reduction in time to market. We have increased almost 3x in terms of the throughput, the output that we've had there. But more importantly, our deployment frequency has increased 4x. We're not just shipping more, but we are shipping in smaller batches, so we are having a better impact that way.

Also, our team NPS is up. We are a plus 68. If you're not familiar with the NPS, it's on a negative 100 to positive 100 scale, so plus 68 we're very, very proud of. While I can't get specific numbers, in the three years we have surpassed 100% return on investment for what we invested in going through this journey.

All right, so key learnings from the past three years. I'll run through these. Love that timer. It just keeps ticking down. I'm going to go through these a bit fast here.

First one: get comfortable being uncomfortable. This was especially true for our leadership, but it really applied across the board. Going from know-it-all to learn-it-all is going from that deterministic mindset to that emergent mindset.

Growing a learning ecosystem. Gene talks a lot about scene-ius in his thing. This was really it. How do we foster those communities of practice, those communities? There's a CoP, a guild, hackathons, dojos, whatever, what have you. But how do we create those forums and those communities that allow people that are passionate about something to innovate, ideate, and experiment and learn whenever that opportunity strikes?

Continuous attention to all-the-things excellence. This goes back to our holistic mindsets. Not just technical excellence, not just being great at agility or DevOps, but also product and UX, that holistic piece there.

Building trust through action. This has broad appeal, but this was especially something that our tech coaches shared with us. It wasn't enough for them to say, hey, you need to go build automated pipelines or do all the DevOps practices. It was really about showing and building that trust and coming in to help. You need help with automation? You're having some quality issues? Hey, I have some experience containerizing an app. Let me work with you. How can we help? Let me help you with what you have there. Show, don't tell, and then promote that experience you're learning.

And then using data to drive insights, avoiding the gut feel, the HIPPO, the highest-paid-person-in-the-room opinion, but focusing on data to help inform those decisions. Those are our five key takeaways.

All right. So where are we going from here, Amy? Where are we going from here?

Amy Willard

I get to lead this transformation at Deere for the last two years, and we are nearing the end of our first horizon, with the 500 teams having that foundational learning we wanted. We get to now figure out where do we go on top of that foundation.

Along the bottom here, we absolutely want to double down on the foundation we've built, so continue to invest, continue to grow, not let up on the gas at all. But now we want to not only focus on shifting from outputs to outcomes. How do we get the maximum value for the investment that we have in a digital organization?

Earlier this week I heard, how do we start to say no to the good so we can do the great? It's exactly what this means. We want a transformative tech stack. We want to use our tech stack to accelerate the outcomes that we want to accomplish as a company. Outcome-driven lifecycle events, DevEx, platform engineering, making it easy to do the right thing, leaning into DORA and elite engineering, that's all here in this space. A tech stack that enables our purpose, that's what we want to do there.

And digital mastery. We spend a lot of time focusing on T-shaping our organization and our teams. What we really like to do is continue to focus on helping individuals master their craft. Whether you're an engineer, you're a product manager, you're a scrum master, you're a coach, you're a leadership role, we want to help you master that craft in all the roles while we continue to drive more innovation into the system. We improve our overall digital delivery, and at the end of the day we want an awesome employee experience.

And we also want to have a relentless focus on removing constraints. That's where we're going next. That's the horizon, and we're really, really excited to build on top of the foundation we have.

Matt Ring

With that, we'll wrap it up with a few last-minute details. The help we're looking for: we would love to connect with you. There's so many amazing stories here this week. We have lots of things we'd love to talk about and we'd love to learn from you. You can find Matt and I there, and we'd love to connect and talk more after this.

Please give us feedback in the app. We'd love to hear that as well. And the last one, as I mentioned, we have a pretty lengthy story to tell, and we had 25 minutes here to tell it. If you're interested to know more about what we did, we do have a case study. You can check that out in the spirit of learning even when we're not here to talk to you. We'd love for you to be able to take some of our learnings from that as well, and then ask us questions if you want to click a little bit deeper into any one of those.

With that, thank you for having us today. We've loved telling our story. We'll be around if anyone has questions. I think we have exactly two seconds left, so if you want to go, feel free, but Matt and I are going to hang out if there's any questions.