Courtney Kissler

Thank you.

All right. Well, thanks for being here. We are very excited to give an update on what's been going on at Nordstrom for the last year, and share what we've learned and what we're focused on for this year.

So I just wanted to give a lay of the land to start with. At Nordstrom, we think about the customer as at the center of the universe. Everything that we build and deliver is framed up in this context.

We have our full-price business, which is inclusive of all of our stores. We have stores all across the United States, and then we opened in Canada and Puerto Rico this year as well. And then we have our online properties, nordstrom.com and our mobile apps in the full-price space.

And we acquired Trunk Club, which is basically, if you haven't tried it, try it. We have Trunk Club for men, and we just launched Trunk Club for women. But essentially, you get products shipped to you based on your style and preferences, and then you can keep what you want, send back what you don't want. Great service.

And then we have our off-price business as well. So from a brick-and-mortar perspective, that's our Rack stores. In the digital space, we have nordstromrack.com and HauteLook, which is basically a flash-sale business.

And then everything that goes on in all of those areas of our business, technology is a key enabler. We're really focused on strategic flexibility so that we don't paint ourselves into a corner.

Our digital experience, I say that because our in-store experience is very high touch. We're trying to figure out how to bridge that gap into the digital space and make the experience more meaningful. And then in our stores, it's really about convenience. We have the high-touch experience, but it's not as easy to get in and out if you want to come in and do something quick.

And then everything's about speed. Gene mentioned the cycle time goal that we have. We'll go deeper on that. And then also improving our reliability.

So first, I'm going to tell a little bit about where we're at with modernization, what our path has been, and where some of our teams are at on that journey, because we have varying degrees of maturity, as you can imagine. He'll talk about that. And then Jason is going to talk about his journey at Nordstrom, and we'll talk about our need for speed.

And then I'm going to share a case study. It's a mainframe application that we have that our cosmetics team uses, and it's a really good example of how we had a lot of assumptions about what we should do about it, and then where we ended up at the end of that case study.

So, our journey. I talk about it in three stages. We've got people crawling, walking, and running. And earlier, we were joking around, Mark Peterson and Gary Gruver and I, that there's probably even a stage before crawl. So we've got people who haven't even started crawling yet.

But really, for us, over the last year, we've really been focusing on hiring in engineering thought leadership. Jason's an example of that. Focusing on shipping product, which probably sounds like a no-duh, but for us, that was a big shift in how we deliver solutions.

And really figuring out what is our strategy around cloud and microservices, so defining that. And then lots of investment in Lean mindsets, practices, tools, techniques. We have some programs that we're using at Nordstrom. We have a senior continuous improvement coaching program. I'm actually one of the people who's participating in that, where you get one-on-one coaching from a Lean practitioner and try to change mindset and behavior.

We've got a lot of adoption around DevOps. Mileage varies on that as well. And we have some teams that actually have started implementing microservices and putting product in the cloud.

We're really trying to get to on-demand releases and starting to establish baseline metrics and figuring out how we can be more data-driven in our decision-making. And we're also spreading talent across the organization. I keep pointing at Jason. He recently moved out of our web team and into another team to actually help them accelerate some of the modernization that's going on in that team.

So I've got my little animation.

So not all teams are at the same pace. For example, we've got some of our web organization practicing things in the walk and the run. We also have some of our web organization still crawling.

Our store team, we've got some people who are practicing these mindsets, and they've gone pretty far with DevOps adoption in those teams. But we're not as far along on our cloud strategy and implementation.

And then our superstar team is our customer mobile team. I talked about them last year. They're one of the only teams that's in the running phase. And what's great about them is they know they're never done, but they want to continue to be running. But they have the best structure and framework for continuous learning.

All right. So now Jason's going to talk about his journey.

Jason Josephy

All right. Yeah. All right. Thanks.

My name's Jason Josephy. I've been with Nordstrom now for a year. Courtney invited me to come up here to talk about my personal journey here, my personal transformation, and how this applies to all the things that we're trying to do, both with Lean, DevOps, et cetera.

So when I came into Nordstrom, I landed on the web team. Our team builds all of the web experience, both the full desktop and mobile site.

It was a team that you could consider in a classic state. It had a lot of developers doing a lot of development. We had a lot of testers doing a lot of testing. We have a release team that manages builds and deployments and environments, which then goes to the ops team, which then goes to prod support. And you've probably all been a part of something like that at some point.

So the first thing we wanted to try to start doing was bring these teams together. We were operating with a Scrum model, and so my background in Scrum was long. I've been doing Agile, primarily only Agile, for about probably 10, 12 years. I got my Scrum Master certification in the early 2000s. I went on to do Agile, lead Agile teams. Everything that I knew was Agile. I knew that to be a great success platform. I had been successful with it. It was the way that you do things, period.

So as these things evolved and I started getting into the team and trying to bring this stuff together, it wasn't working, and I didn't really know how to do that. I had been in the environment for a long time and I had made a lot of assumptions, and so someone new coming to the team, that's challenging.

There's also sometimes a lack of trust involved there. The credibility of me, who I am. I showed up, I'm telling you what to do with all these things. It's very difficult.

And so as these things evolved, I knew I needed to try something else, and I was getting a lot of that feedback. And so these conversations led to all this stuff about Lean, and I was like, "Yeah. Right. Sure. I've seen that. I've heard that story." Honestly, I was completely skeptic that it had any value to add to software engineering, software delivery. I didn't even think it could be applied, honestly. I always thought Lean was this way to create units of measure that you would manufacture.

And so we began sort of our journey. Oops, wrong way.

Courtney Kissler

No, you're welcome.

Jason Josephy

Sorry. I'm new to this too.

So we scheduled this value stream map session. I don't know how much you know about this, so I'll explain a little bit.

This is part of what's called the continuous improvement process. I was brand new to this. I had no idea what it was. Courtney had set it up. She said, "I want you to participate in this." I said, "Okay." I can't say no, so.

Courtney Kissler

It went really well. It did not go like that, actually.

Jason Josephy

Yeah.

Courtney Kissler

"Would you please participate?"

Jason Josephy

Take a leap of faith. Yeah.

So what it is all about is that you take the whole team together, and we were, again, in a place where we were very siloed, if you want to call it that way, and bring the team together to build your map, basically.

This is a representation of our map, a physical representation, and the team builds it together. And it highlights all your processes, so you'll actually be able to see very clearly what's going on, and you'll be able to talk about it.

And at first, I thought, "This is not the way to solve this problem. What you do here is you write software and it goes away. Why are we spending our time doing this?" And of course, that was contentious at the time.

So we started to go through this, and it actually almost immediately developed this thing that I had not experienced at all. The people that were challenged with these problems came together for the first time, it seemed, and all of a sudden there was this commitment, collaborative commitment to understanding these problems, really understanding what was really going on.

I learned a ton that I didn't actually expect to even learn out of the process, and it was quite amazing. It really built this foundation of collaboration where we could share this experience.

And so then the idea of the value stream is to start creating experiments, try to reduce your cycle time, try to improve your process.

So it was shortly after this happened I got invited to go to a summit in Florida, the KataCon it was called by some people. So I said, "Okay, let's check this out," and I went down there. And this was when I first got introduced to the whole process, like what improvement kata really was and continuous improvement.

And so the first part of the conference was about this learning aspect, and it was fascinating to me because the whole idea was you create this system where you have learners on your team, and you do your work, and then you learn from it.

And when I walked in the room, the first thing they said to me was, "It's not about outcomes." And I thought, "That cannot be true. Everything we do is deliver. That's all we do. We deliver stuff."

And so they walked us through this framework, and it was fascinating to think about just the pure scientific method and how simple this aspect of learning was that we almost just don't even behave that way anymore. And so we learned this process of learning.

The next day was about this coaching aspect, and the coaching aspect was really fascinating to me because it's based on the coach and learner relationship, and the coach's job is really not to manage people, but to actually direct us towards our target, ask questions that help us learn, help all of us learn, and continue to learn, create that space of learning, and help the team learn.

And the great thing is that coaches and learners can switch. Everyone can be a learner. Everyone can be a coach.

And so this was really a powerful thing because out of this coaching, the coaching kata, what I learned is that we don't want to provide solutions to things. And that was what I always was about. Like, "Here's our solution. This is the way we're going to do it. Let's go do it, and we'll deliver it." Outcomes, solutions.

And then you lose the learning, and that was something that I had been doing a lot of.

So I left the conference with kind of a couple things. While I was there, I saw this great quote, and it said, "At Toyota, we do not make great cars. We make great people who make great cars." And that was kind of very meaningful for me because it was something I hadn't really considered.

I just had this assumption that everyone was just going to do the right thing for us to deliver, and I never optimized on the culture of greatness applied to that. So it was like, if you make great things, you can make great things happen. If you get great people, your tools get great, your process gets great. Everything you do becomes great. You make great things. And that was a very powerful moment there for me.

So what we tried to do is we left the conference, came back, and we started trying to coach. And so this was originally the five coaching questions of the kata. We have adapted it.

Courtney Kissler

Nordstromized it.

Jason Josephy

Yeah. A little bit at Nordstrom, so now we turned it into the eight questions.

Courtney Kissler

It's more fun with eight.

Jason Josephy

Right.

So you can go look up the coaching kata. What this really is about and why it's so great is, it's actually a framework that you can use in your day-to-day activity, and that's what I use it for. It's a reminder of how to try to do this because it's not natural at all to just simply have conversations like this, regardless of the protocol or the process you use, just to talk in these ways to try to lead people through the state of learning, myself included. So it was actually very meaningful.

So we went back to Nordstrom, went back to Seattle, started trying to apply some of these things.

So fast-forward back to where we are now. The web team now has made great gains in some of these applications. We've brought the teams together. We're starting to really work on CD, CI, starting to do DevOps. All teams are now contributing to the culture.

And the great thing now is we're bought in. People are excited. There's change happening. Everyone's talking about it, and everyone's on board with it. And it's a testament to how these things will take off. They will take hold. And so it's a really great thing.

So that's my story. I'm going to hand it back to Courtney.

Courtney Kissler

All right. Thank you.

All right. Now we're going to talk about the goal of reduction of cycle time by 20% across all of our customer-facing teams. I'm going to explain how we got there and what we're doing to, basically, the countermeasures that we're using to actually achieve the goal.

So in 2014, we were really focused on how do we get more productive, and there were a lot of different opinions about how we could make that happen. Luckily, our continuous improvement method won, and we were able to demonstrate that if we have teams understand what the goal is and then give them these techniques, then we actually would get more productivity, and that would just be a natural outcome.

And essentially, we landed on, let's pick a goal. We'll pick the 20% reduction in cycle time for 2015, and then we sent everybody on their journey to figure out what were the appropriate tactics and countermeasures that they could use to meet the goal.

So the current condition was we weren't really measuring cycle time. We had some teams that were. Lots of teams had different definitions. Definitely was not happening consistently across all of the customer-facing teams.

So our target was let's measure it and make it visible, and let's conduct experiments to reduce the cycle time.

So this is actually a picture of the board that we use to manage, through the coaching kata, of how we're doing against each of the team's target and what countermeasures they're leveraging to reduce it.

So I talked about cycle time visibility. That's kind of first and foremost. We've got to measure it. And then a lot of teams are leveraging value stream mapping. Jason mentioned the web team. Other teams are doing that so that they can really understand what would be the biggest opportunity and problem to solve in order to get the 20% reduction.

Continuous delivery is probably not surprising to see that on there.

Microservices is another mechanism because we have, I guess the best way to describe it is we kind of have a hybrid going on right now on the web team where we've got run-ahead teams that are really building product in a new way, and then we have our existing way that we release features into the web.

And then we're leveraging cascading Hoshin. We do that with our selling and customer-facing business initiatives. We actually use an A3x, and we stack-rank all of them, and then we cascade that to the teams.

We do the same thing for these initiatives as well. So we make sure that we're working on the right things and that we have the capacity to actually make progress against those.

So again, can't fix what you can't measure, so make cycle time visible.

So this is an example of one area that I support, basically how we're keeping track of cycle time by month. Basically, take a monthly snapshot and then also output.

So there's a lot of conversation about how you define output, but essentially this is anything that the team is working on that shows up in the customer experience. And this enables us to have transparency.

We all get together once a month, and we ask why when we see patterns that don't make sense. Take one example. In August, cycle time went up, and output went up, which is typically you see what we saw in June, where cycle time goes down, output goes up.

So this gives us that opportunity, though, to really dig in and understand what's going on in those teams and help them if they need help getting unblocked or if something's in their way.

Now I'm going to tell the story of our cosmetics business office. This is a mainframe application. It is used by our floor managers in the stores, so all of our beauty and cosmetic store managers and department managers.

When we started this, so I actually, at one point in my career at Nordstrom, I supported this technology team. And every year we'd go into planning and we'd say, "We got to get that thing off the mainframe. We got to do it this year. We're going to do it this year. We're going to move it off the mainframe. We're going to get the support." And then we would never move it off the mainframe.

And then the next year would come, and we'd have the same conversation. So this year it happened again, and we said, "Well, why don't we actually leverage a value stream mapping exercise, and maybe the outcome is that it needs to come off the mainframe. It's possible. But maybe there's a bigger problem to solve."

And this will give us that opportunity. As Jason said, it really brings together all of the people who are part of the value stream. They collaborate. It's fact-based. They're able to look at the whole end-to-end process and then go after what makes the most sense.

So, as I said earlier, the assumption was mainframe, it's unreliable, it's unsupportable, and we've got to move this app off.

By the time we got through the value stream mapping exercise, what we uncovered was at the very beginning of the process, we were asking for information from those department managers they didn't have.

So they'd submit the request. Sometimes they'd leave that field blank, sometimes they'd put garbage in it, and it would go to a central office, and they would kick it back because it didn't have what they needed. And that would go on for up to a week. They'd go back and forth, back and forth, until they either found the information through some other means and submitted it accurately, or that central team would just look it up for them.

And so basically what we found out was, well, we should stop asking for that information.

The other problem that we had was, in order to submit the information, they had to go off the floor, which is not very practical. They're working. They're on the floor. They didn't have access to this application unless they actually went into a back room and used a PC to submit the data.

So what we did is we built them a, I don't know if it's even really fair to call this an iPad app, but essentially we put the form on an iPad. The iPads are on the floor. And we only asked for the information that we needed just in time, and we simplified the form.

And essentially now the central office looks up that second piece of data. So it's not that we didn't need it, we just put it with the right people.

So this is the view of what it looked like before we did that experiment. It would take up to seven days of back and forth between those two teams. We had a hypothesis. We thought we could get it down to same day, and we actually got it down to near real time. So they're submitting it as they go, and basically it's happening in real time.

This is the other fun metric. So prior to the experiment, we would see up to almost 50 discrepancies per month. And we put the iPad app in, and in June we only had three. July, we had a bit of a spike. If you don't shop with us, that's our big event is in July, our Anniversary Sale, so we tend to hire a lot of people for that month. So we saw data discrepancies go up for that reason. But then we came back down in August.

So basically, the outcomes of all of that was we had behavior change. The people who thought that the right thing to do was to move it off the mainframe had their kind of aha moments and said, "Oh my gosh, I was wrong. It's actually not the biggest problem to solve."

The leaders were really engaged. We had executive leadership support in all aspects of the value stream.

And what was really fun when I went to their report out, the business leaders were the most excited. They were like, "Well, we're going to do this one, but then we can't wait to do the next experiment. We already know what it is. We've got a big backlog. We're just going to keep doing these experiments." And they were very energized.

Trust went up because the business teams were frustrated. They couldn't understand why we weren't solving this problem. The technology teams were frustrated because it wasn't a super fun application to support. And trust just went up across the board when we did this exercise.

Team morale, same thing. Very happy when they got through this. And they learned something new. So everybody learned how to problem solve, what value stream mapping was, how to take a small experiment, because that's another thing that we're not always so great at. We want to boil the ocean. So picking something that would be small enough so they could see a win and then move on from there.

So my top five takeaways are actually the same takeaways that I shared last year.

So Jason mentioned it earlier: people are our number one asset. The customer's at the center of our universe. Everything that we do should be grounded in delivering customer value. Really trying to create a learning culture and just have that be how we get our work done.

And this one's my favorite because it's, well, maybe the last one's my favorite, but this one in particular is something that I'm not very good at, is being disciplined. So it takes a lot of discipline to practice these techniques. You have to remember. You have to try every day. It's very challenging.

And then being persistent, because in the example that Jason used when we did the web value stream map, there were a couple of times where we tried experiments and they didn't work.

Jason Josephy

Mm-hmm.

Courtney Kissler

So the team, they can get frustrated with that. Like, "Why did I do this at all? This isn't working." So having that persistency and just keep going.

And then leaders have to evolve. This is something I believe in very passionately. I think this is a different way of leading. You have to be connected to your teams and the work that they're doing. You have to teach, not tell. And it's just a very different way of going about how you get your work done.

Jason Josephy

Yep.

Courtney Kissler

And then what we need help with. I think there have been some themes across all these presentations. For us, we have some executive leadership that understand the value proposition with DevOps, but we need more of that.

And then case studies at scale. I think Ross and Heather kind of said the same thing when they kicked off on Monday.

And then we're hiring. We're always hiring.

Jason Josephy

Yeah.

Courtney Kissler

Talent. Yeah.

Jason Josephy

Very excited.

Courtney Kissler

Yes. Come join the fun.

Jason Josephy

Yeah. Come help us.

Courtney Kissler

And if you have any questions, you can find Jason and I on Twitter.

Jason Josephy

Mm-hmm.

Courtney Kissler

And with that, I actually think we're under time.

Jason Josephy

Yeah.

Courtney Kissler

How'd that happen? Oh, we can? Sure.