Rethinking Failure in L&D

Introduction

Nolan: Hello everyone, and welcome to the Learning and Development Podcast, sponsored by Infopro Learning. As always, I’m your host, Nolan Hout.

I’m very excited for today’s episode because we’re going to take a step back from talking about the nuts and bolts of learning and focus a little more on the culture within L&D itself—specifically, how we need to feel comfortable about failing early and often, and how that can be a superpower within L&D teams.

Leading us in this conversation is Patrick Peterson, Manager of Human Capital and Workforce Development at Deloitte. Patrick has been in this space for over 15 years and brings a great mix of hands-on experience along with strategic and leadership expertise in L&D.

Like me, he’s doing quite a bit, with a strong focus on AI at Deloitte—but that’s a story for another podcast.

Speaking of podcasts, let’s go ahead and kick things off and meet our guest.

Hello, Patrick, welcome to the podcast.

Patrick: Thanks, Nolan, happy to be here. How are you doing today?

Nolan: Doing wonderful. And Patrick, thanks very much. I’m excited to talk to you today about a myriad of issues. Before we even got started, I realized we’re probably going to need another podcast. I understand that you’re leading a lot of the AI conversations at Deloitte, and I’m here for it. Gosh, I’m sure we could have a whole series on that.

But today, we’re going to talk about taking risks and why that’s important in L&D.

Before we dive in, we always like to start by learning a little more about the guests we have on the show. You’ve obviously done quite a bit in your career and have been at Deloitte for quite some time, but everybody starts somewhere.

What was your origin story—how did you get into learning and development, and how did that journey lead you to where you are today at Deloitte?

Patrick Peterson’s Career Journey

Patrick: It’s an interesting story. I’ve been at Deloitte for almost 14 years now. Before working at Deloitte, I worked for the federal government in an inspector general office, and the role was to detect fraud, waste, and abuse of U.S. funds in Afghanistan. As part of that effort, I had a small team focused on going through financial transactions and detecting contract irregularities and things like that.

I was empowered relatively early in my career to lead this team. Very soon after my first day, I realized that I knew nothing about leadership and very little about using automated tools to detect fraud, waste, and abuse in Afghanistan. I needed to learn how to do that very quickly.

I essentially created a curriculum for myself, which I then refined with my team—we created our own little curriculum. In doing so, I discovered my passion for using learning to drive skills growth and creating a safe space where people can learn, consider their strengths, and grow and develop as individuals.

I feel very strongly that learning in an organization is where the culture lives. For organizations to meet this moment of unprecedented transformation, we need to preserve that culture and provide a space for workers to develop themselves, to grow, and to change.

That’s why I love learning. That was my origin story, and that’s what I’ve been fortunate to do—both for the federal government and now at Deloitte.

Finding Purpose & Becoming an Enabler

Nolan: Do you feel that’s a common thread that’s pulled through — this idea of, “I didn’t feel enabled to be successful, so I wanted to be the enabler, to create something that helped others get there”? Can you trace that back to your college days, or was that completely in the back of your mind? If you look back, would you say, “Yeah, maybe I was that kind of kid”?

Patrick: Absolutely. When I was in college—certainly during that time—one of the future visions I had for myself was to be a professor. The only reason I didn’t pursue a PhD was because I didn’t want to commit to seven years of training after graduate school. I’m also a very forgetful person, and the idea of doing seven years of research was not appealing to me.

But yes, I’ve always been someone who likes to enable and support others. I’ve always been interested in sharing knowledge, sharing other people’s stories, and making sure those spaces are open for people to grow and learn from each other. That’s been a consistent theme throughout my career.

Failing Fast & Embracing Risks in L&D

Nolan: Yeah, sticking with that—that’s how we got connected. I had read an article you wrote about failing fast and risk-taking in L&D, and I just thought, what a great topic to talk about today. A lot of what we see with transformation today is really centered around the technology—what it can do. Versus, what are the underlying things we need our teams to actually do? So, if you don’t mind, could you give a brief summary? When you’re talking about failing fast and taking risks—what’s really at the heart of that? Why is that important, and why is it valuable?

Patrick: Some of it is in the data. You mentioned technical skills—digital transformation is not a novel idea. Technology is changing everything at an unprecedented speed. According to Deloitte’s research, the half-life of a technical skill is between 16 and 20 months. That means if you learn how to do something today, in 16 months that skill could be out of date. For some career fields, it’s even shorter.

In that environment, for L&D to be the hub of culture, the place where people go, the enabler of the workforce—we have to produce faster. And the only way to do that is to move quickly. In software and digital product development, there’s a concept: fail fast, learn quickly, and dream big.

L&D doesn’t always operate with that mindset. But the benefit of failing quickly is that you can learn, grow, and improve. A side benefit is you also engage your workforce more frequently. If you have a new product or learning solution that rolls out every couple of weeks, that kind of rapid engagement means you’re connecting with learners in ways a traditional schoolhouse or L&D organization may not. And that helps—even if the solution itself isn’t perfectly polished.

Making Training Work Across a Diverse Workforce

Nolan: Yeah, so I heard a couple of elements in there. Maybe we can start at the end—with not being perfect. I feel like corporate training is an almost impossible job to get perfect because you’re developing things that, even when broken into smaller nuggets, still need to cater to a large demographic.

It’s not like you’re marketing razors to college kids, like the Dollar Shave Club kind of thing. You don’t get a narrow demographic where you only have to create something that appeals to a 20-year-old male in college, or whatever it is. It’s a much broader spectrum.

When you’ve got such a wide audience—and many other challenges as well—whatever you push out often hits a large portion of your workforce immediately. Who are your clients? How do you respond?

I guess the question is: how do you merge the mindset of “let’s just get something out, even if it’s not perfect,” with the need to gather results and feedback? How do you know when enough is enough?

Patrick: I think there are a few things you can do.

One is, I actually think you can do more segmenting of the workforce—and it’s not just by job. Put the accountants in one group, the program managers in another, and so on. That’s part of it. I think you can take a cut that looks at tenure, geography, performance ratings, and things like that. L&D can generally do a better job of segmenting programming for their specific workforces.

Beyond that, it’s really about adopting the mindset of: what does a minimum viable product look like? Let’s say you’re creating a web-based training—maybe on how to use the corporate credit card in compliance with policies. Every instructional designer has seen this product, right? An MVP of a solution like that could just be a series of index cards on which you’ve drawn out what the solution is. You can put that in front of one or two people and say, “Hey, I’m thinking of doing it this way. What do you think?” That gives you some good understanding.

As you accumulate that input, you start to build lessons. For instance, if you put something like that in front of someone and there are four or five screens of content, and you hear from a couple of people, “Hey, that’s a little too much content,” then take that lesson. The next time you create something, maybe don’t do that. You build that kind of knowledge base to improve your product as you go.

That’s where you really do the testing—after you have something that you can pilot and deliver. I would encourage people to do that earlier rather than at the end. Once you get to a 50% solution, that’s when you start getting feedback: this works, this doesn’t—that kind of thing.

Building Effective Training Through Targeted Pilots

Nolan: And you must face that. We’re in the development space as well. When we work with clients, I know this is a struggle we have—and I think everybody struggles with it. You want to show something that looks nice enough so people can follow along and not feel embarrassed. It’s really about finding that balance.

But I liked what you said about a couple of things, and I think a lot of those are connected—like finding a segment you feel you can test. Even if it’s supposed to be a global thing, or maybe just something for the customer service team, just say, “Hey, can I get five random people’s email IDs?” It’s not going to be a big deal. We’ve got 10,000 call center agents—what’s five going to do if I take them through a program?

Let me just show them. Talk to them on the phone: “Hey, I’m thinking about a training program, maybe about 20 minutes long. It’s going to cover this, this, and this topic. Is that what you’re looking for?”

I think we often get caught up in the idea that we have to really formalize this test batch—as if it’s a two-and-a-half-minute episode with screeners and edits and a bunch of production cuts. But a lot of it can be one-to-one—just a conversation. Even talking to somebody who’s not going to take the training program, but someone in your peer network: “What does this look like? How does this feel?”

A lot of that, I think, is really relevant.

Understanding the ‘Why’ Before You Test

Nolan: One of the things I want to touch on—you mentioned testing at the beginning. How important is it to know what you’re testing for? Because I always feel like when I see something go drastically wrong early on, it’s because we didn’t fully grasp the why behind the why. How important is that before you even get off to the races and start showing people?

Patrick: Yeah, I think it starts from the beginning. That’s the beginning of the instructional design process. There has really been a movement in modern instructional design around focusing on skills and building a training program that chains skills together.

So, I’m going to teach you how to do this. Then I’ll teach you how to do that. You go from one thing to the next. When you do that, you naturally get smaller chunks of content, and that also helps you figure out what you’re going to test.

Let’s go back to our purchase card example. If one of the skills you need to train people on is how to fill out an expense report quickly, that’s a skill. How to use the expense report system—that’s something you can test. You can ask: does this give you the ability to use this process? Yes or no? On a scale of one to five, where do you feel you are with this capability? That’s the kind of information that shows you what’s working and what’s not.

If I could just come back to one of the things you mentioned—everyone wants to present a polished product. Sometimes the way you present things can help reduce the stress. Rather than using a completed PowerPoint deck with all the fonts and colors perfectly formatted, try using a hand-drawn image or a wireframe model that’s intentionally low fidelity and high level. That often gets you better feedback.

And you can get it out quickly because you don’t have to spend time on font choices, colors, and that sort of thing. You can focus on what really matters.

Nolan: Yeah, like the difference between opening up a PowerPoint slide in slideshow mode and saying, “I’m going to take you through this,” versus keeping it in edit mode, build mode, and just being like, “Hey, I’m going to take you through this.” Psychologically, I’ve already framed myself for, “Okay, we’re going a little untraditional here. They’re going to be changing some words.” So yeah, you’re right—there’s so much to how you present it and being honest with whatever it is that you’re presenting.

Breaking the Fear Barrier with Fast Learning Failures

Nolan: And I think a lot of that comes from what you mentioned earlier. We talked about some tips and things, but there’s also a psychological element to being able to fail fast. Talk a little bit about that.

Patrick: It requires a sense of vulnerability. You have to be willing to do something without the certainty of knowing it’s the right choice. A lot of what we’ve discussed is about identifying what those right choices are—connecting the data, working in small chunks, picking the right people to test with. But some of it is simply having both the leadership permission and the personal comfort with being wrong.

That’s hard for everyone. I remember reading a paper a while ago that charted what professional workers are most nervous about. The top concern was being wrong about something. It’s an incredibly high number. So it’s not just limited to the L&D community—no one wants to be wrong, especially when presenting work in front of others. Some of it is about pushing through and being okay with that.

It’s also about setting the right environment—one that’s casual and open. Saying, “Hey, we’re just going to try this out. This is experimental. We’re playing around.” There’s a concept often discussed in learning development around improv. I don’t have a background in improv, but I’ve used it a lot in my training programs. The idea isn’t just about saying “yes, and…” It’s about going with the flow, really listening, and accepting what someone else is offering.

When you create that kind of environment with your test users or the communities you’re supporting, that’s when you get the best feedback. And you can even joke about it later. For instance, there was a training program I developed at Deloitte—a large milestone event with hundreds of people in a ballroom. It had high production value. We created a workflow where people would come on stage, perform certain activities, and then we’d run a training session. There were a lot of steps. It failed—massively. We had done all the research and preparation, but it was a disaster. I was in the back, furiously typing on my computer, trying to salvage it in front of 300 or 400 people. It was a mess.

But what happened after that event wasn’t panic. It wasn’t, “This was such a disaster, we need to change everything.” It was more like, “Well, we tried that, and it didn’t work. What can we learn from it?” We found a way to laugh about it.

That’s where L&D leaders, deans, and business owners can really make an impact—by giving their teams the permission, focus, and freedom to fail.

The Power of Open Conversations

Nolan: Yeah, and that’s what I was going to touch on. I think a lot of it comes down to how you model behavior for your team. I was very fortunate. When I got hired, the gentleman who brought me on is now our CEO. So I learned a lot from him—for better or worse.

One of the things I didn’t understand at the time was that I thought every person with two years of experience would yell at the CEO if they thought he was wrong. He wasn’t the CEO at the time, but we’d have heated discussions. We had someone join the team who wasn’t from our company, and they asked, “Is that how all your meetings are? Is this person just saying you’re wrong to your face?” And he responded, “Well, why wouldn’t he? Don’t we just want what’s best?”

So I’ve grown up in this organization where you have a spirited debate about what’s right and what’s wrong—and whoever has the best idea wins. I think that creates a lot of psychological safety. Of course, you have to understand when that kind of debate is appropriate and when it’s not. But I think modeling that behavior sets a tone.

I’ve also seen other leaders join our organization who didn’t share that approach. They struggled a bit, trying to lead in a more traditional, top-down way—pushing information instead of inviting discussion. It’s fascinating to see the difference in the quality of ideas that emerge from each approach.

Patrick: Yeah, I feel really strongly that it’s important for teams across disciplines—whether it’s learning, development, or any other function—to be intentional about setting the norms and practices they’re going to follow as a team.

One of the things we do at Deloitte is focus on empowered wellbeing. Part of that conversation is about how we want to work together.

Deloitte, like many other consulting and professional services firms, is spread all around the world. We have people with very different styles, backgrounds, and needs. Often, there’s a conscious conversation when you start working with someone new: What do I want my core hours to be? What’s the best way to reach me? How do I like to be challenged? How do I make decisions? Am I more of a data person or an ideas person?

Having that shared understanding as a team is often how you create psychological safety. And that’s something I think anyone—regardless of whether they’re in L&D—can really adopt.

Skills at the Speed of Technology

Nolan: Yeah, so I want to switch gears a bit and get back to this idea of failing fast and why it’s so important today. I was just talking to someone about skills and the big shift toward a skills-based organization, where skills are truly at the center. You touched on that recently too, Patrick.

I was telling them that we just didn’t used to have the ability to reach a consensus on skills quickly. AI has really helped us fill in the gaps. And to your point, I always say, it’s not perfect. But honestly, if we tried to go back to the old way—working with every department to identify the key skills for a marketer, a salesperson, or any role—it would take at least six months. You’d spend all that time reaching an agreement, acting on it, then start over with another group. It was all about job analysis or role specifics, rather than focusing on skills.

So how do you feel about that? I feel like the push to move quickly is now being set by the technology itself—if we let it. How do you think technology is changing not just what we can do, but also what’s expected of us in such short timeframes?

Patrick: People tell you what skills they need. They’re sharing that in some way, shape, or form. As an L&D professional, you’re writing it down and putting it into an Excel spreadsheet or whatever platform you’re using. In my view, that’s not high value. If all you’re doing is gathering information and spitting it out, that’s a tough business to be in.

L&D needs to reposition itself from an order-taker model to a strategic talent advisor. What that means is having a baseline understanding of the state of development of your workforce. You can gather that from all kinds of data. Microsoft, for example, through its Viva product, has the ability to see how people are engaging with each other in email, Teams, and other applications. That gives you an understanding of how connected people are to each other.

Nolan: That’s through their graph database, right?

Patrick: Not individual gaps — not gaps in terms of ‘oh, this person doesn’t have the skills necessary to deliver’ — but rather, ‘here’s where I need to start targeting learning so that our workforce can really meet those current and evolving needs.

Nolan: And it’s all connected because you mentioned the half-life was, what, 18 months? Yeah. So, who usually puts this out? Not the World Health Organization—I forget which organization, but some other group usually publishes this number. They recently said it used to be around 10 years, then it dropped to five, and now it’s down to two to three years. That data came out about two years ago.

Patrick: That is indeed the case. I’m 18 or 22 or something.

Nolan: Absolutely. It tracks that we now have 18 months to recycle these skills. That means if you were to look at that and say, “Well, the old way took six months just to identify the skill,” by the time you’ve actually identified it and done something to develop or change it, the skill has already moved on. It’s like—“Thanks, but we don’t need that anymore.” Of course, we still need it, but in a different form.

So, I think it’s about adapting. You hear a lot about learning “in the flow of work,” and I think that’s where we need to head. If employees and learners expect knowledge to stay right in front of them, it’s a strong call to action for us. If this is the new paradigm, how do we create a new game? How do we change the way we operate?

How do we shift from thinking in terms of “an e-learning asset” to simply delivering the knowledge they need? I know Deloitte is doing incredible work here with connected agents and other initiatives. That’s a great example of taking a calculated risk early—realizing people may no longer want to take a 30-minute e-learning course.

We just launched a leadership coach powered entirely by AI. It’s built into our model. The person who wrote the content—our IP owner—said, “I don’t know if people will go for this.” And I told him, “Honestly, I would prefer it.” I’m super forgetful, and while I can grasp broad concepts, when I’m about to enter a difficult conversation, I want to ask a coach in real time: “Hey, I’m about to address poor performance—give me some tips.” That’s real-world feedback.

And maybe that’s what makes us rethink traditional leadership training. Maybe I don’t need Nolan to sit through a two-day leadership summit. It’s not just about failing early—it’s about taking big swings when the payoff could be huge.

Prioritizing the Right Opportunities for Fast Wins

Nolan: And that brings me to one last point I wanted to touch on: these days, we have so many opportunities and options on the table. How do you choose where to make the bets? If you’re going to fail fast, where do you say, “You know what, this is low-hanging fruit—let’s go after that”?

Patrick: Yeah, I think it’s a tough question. I mean, you just kind of have to go with your gut. I usually start with something that’s not very consequential. Compliance training is a good candidate.

To your point earlier—are people going to take a 30-minute e-learning? No. People hate the 30-minute e-learning. Regardless of how nice it is, people hate the 30-minute e-learnings that I’ve created, I’m sure.

I think the reality is, for learning to be effective, it needs to be human-centered—person-centered—which means it needs to be short and focused. So, take compliance training and ask some tough questions:

What if, instead of having this WBT that we’re required to do every year, which is only mildly updated and consumes a lot of resources, we simply ask people to read through a policy, click to agree, and take a quick test to ensure they understood?

What if we created a small game out of it? What if we tried something completely different? What if we engaged the workforce and asked them, “What would you prefer for this training?”

These are the kinds of experiments I’d start with—something low-stakes that won’t impact your leadership strategy for the next 10 years. If it doesn’t work—if, for whatever reason, people don’t like it—you can always revert to the previous version next year. That’s where I would begin.

I think the idea of learning in the flow of work—we finally have the ability to deliver on that vision with technology. Not just AI, but technology more broadly.

It’s time for instructional designers to pick that banner back up and ask: What does learning look like on an hour-to-hour basis? How are we engaging the workforce so they’re not just learning, but actively using content, growing, and interacting with these systems more effectively?

AI is a Leveler & Talent is the Multiplier

Nolan: Yeah, well said and interesting. I just saw a post from a lady I follow on LinkedIn, and she said something like, “Is learning dead?” And of course, it’s not. What I wrote to her was, “It’s interesting that you would even make a post like that, because I feel now is the time.” I feel people—and CEOs—are looking for an answer to talent, because talent is the finite resource.

I personally feel that AI has been a great leveler, for better or worse. It’s leveled a lot of the playing field, just like tech did way back when. That allowed Microsoft to surpass competitors that were a fraction of its size. PayPal grew to where it is—not because it was the most secure financial institution or the oldest. It wasn’t Wells Fargo or US Bank. It simply created something by leveraging technology and was able to scale really quickly.

I think AI is going to level a lot of that competition for people who may not have had the same opportunities. I met a person at—actually, were you at HR Tech this year, Patrick?

Patrick: I wasn’t, no. I’ve been in previous years.

Nolan: At an event at HR Tech, I was talking to a founder, and he said, “I went to this startup event in Silicon Valley, and I felt so bad because these people got up and said, ‘We’ve been building this tool for the past two years, and it’s so great.’” Then he told me, “I turned to the person sitting next to me and said, ‘We developed that in two months with AI.’”

Patrick: It’s funny you should mention this. I happen to have been doing something very similar. The leadership program I mentioned—the one that failed terribly—is no longer used by Deloitte, but I still have all of that content. One of the things I wanted to experiment with was: what if I created a chatbot that ingested that material and could provide advice as an instructor or facilitator from that course?

I mocked something up really quickly. It took me about an afternoon to create a solution and test it out. As I started testing, I realized instantly what some of the strengths and limitations were.

I think that mindset of getting to something quickly—just trying something out—is really important. AI is incredibly powerful, but it’s not actually that complicated to develop solutions anymore. Things are getting easier with agentic AI and these no-code, low-code platforms.

They’re becoming the modern equivalent of building a website—going from writing raw HTML in Notepad to using Squarespace or another user-friendly platform. That’s where we are now.

Instructional designers need to embrace that, because that’s where you’re going to learn and grow. That’s where you’re going to drive value as an organization.

Nolan: And that’s where you’re really heeding your own advice—if talent is going to be the leveler, then be that talent that helps level your organization. They’re looking to you. It’s a great opportunity to answer that call.

Closing Thoughts

What a great way to end. Patrick, thank you so much for investing your time with us and sharing your lessons. I think this was a really great conversation from my perspective. So often, we get caught up in the nuts and bolts of things and don’t take a step back to consider how we feel about doing some of these things quickly—what the emotion is behind it—and how we might change the way we approach the projects we have to complete.

Thanks for adding that perspective.

Patrick: Of course. Yeah, happy to be here. I mean, like I said, I think learning is where culture lives, and learning is really important for everybody. So, as instructional designers and learning professionals, it’s a powerful opportunity. I hope, for folks who are listening or watching, to partner with them on that.

Nolan: Yeah, well—everyone, Patrick Peterson is on LinkedIn. He is not the famous NFL athlete Patrick Peterson, so don’t confuse the two of them. This is Patrick Peterson with Deloitte, not with the Arizona Cardinals or wherever he is by the time this comes out. Well, thank you so much, Patrick. I look forward to possibly having you on sometime in the future.

Patrick: Wonderful, thanks for having me.