The mass personalization of change: Large-scale impact, one individual at a time

| Podcast

In this episode of the McKinsey Podcast, Simon London speaks with McKinsey senior partner Bill Schaninger and McKinsey partner Alexander DiLeonardo about organizational change. An edited transcript of their conversation follows.

Podcast transcript

Diane Brady: Hello, and welcome to the McKinsey Podcast. I’m Diane Brady. If you’re like me, you’ve probably binge watched more than a few shows during quarantine, so you already know how streaming services collect data that helps them mysteriously recommend what to watch next. What if we could take those tools and techniques and apply them to organizations? Call it, “the personalization of change.” That’s what we’ll explore in this episode taped shortly before the COVID-19 crisis began. Simon London speaks with partners Bill Schaninger and Alexander DiLeonardo. Bill is a leader in our global organization practice and Alex is a partner in our London office. Here’s Simon.

Simon London: So, Alex and Bill, thank you for doing this. And welcome to the podcast.

Bill Schaninger: Thank you for having us.

Alexander DiLeonardo: It’s a pleasure.

Simon London: Good. So Bill, maybe you could start by framing this up for us. Just recap, what are our core beliefs about organizational change and how does the individual fit in?

Bill Schaninger: Well, look, it’s a great question. So we’ve built over the course of 20 years a basic belief that said change works when you manage equally for performance and health, or in plain language, how you’re going to make money and how you’re going to run the place. As we’ve been able to go and dig deeper into organizations we have an embedded question that we’ll often use.

“If you think about the last year, are things the worse, same, or better?” And what we often find is a tale of two cities. You’ll find groups of employees that really think it’s getting better because they’ve been involved or they feel like they were shaping it.

And others feel like they were merely just on the receiving end of it, of ask and demands for performance. When you build something around an average, you start discovering that very few actual individuals look like that perfect average that you built the plan for.

So if we were to take that same language, “How do we make money and how do we run the place?” And say, “What would that look like on the individual level?” Well, in performance it wouldn’t be, “How do we make money?” It would be, “How do I individually contribute to driving value in this organization? What do I have to do day-in and day-out in my day job?”

And that allows them to have very specific role clarity, but understanding that everything they’re working on matters and how it contributes. And on the health side, right, for the organization, how do we run the place, it’s for every person to go, “How do I have to think and behave different every day to bring this to life?” So that’s the essence.

Simon London: So to be successful you’ve got to go down a level. Yet, you’ve got to conceptualize and plan all these things at the level of the organization, but you’ve actually got to approach people at the level of individuals.

Bill Schaninger: One hundred percent.

Simon London: Now, that’s a pretty tall order, right. That’s a tough thing to do in big organizations with tens of thousands of people across, you know, potentially hundreds of work sites. So just talk a little bit, Alex, about why it’s hard and maybe how we can approach it.

Using technology as a tool for change

Alexander DiLeonardo: Just to start, if you think about what Bill just mentioned, doing this at the individual level is a good forcing mechanism to get very clear on what you’re trying to do even at the organization level, what is it we’re actually asking them to do? And then when does that show up in their day-to-day jobs?

Doing this at the individual level is a good forcing mechanism to get very clear on what you’re trying to do even at the organization level.

Alexander DiLeonardo

And then from there, the actual application or methodology of reaching individuals is—is another hurdle. It’s been hard historically because we haven’t had a way to reach individuals on a one-on-one basis. Through technology now we can.

And finally, we haven’t had a way historically to enable individuals to really exercise choice in how they’re showing up and living the change. Through advances in behavioral economics and things like nudge theory, now we can.

Bill Schaninger: We’ve taken a good deal of, I think, inspiration in insight from places like retail and medicine and education when you’ve allowed large numbers of people to still feel like they can have a completely customized experience.

Simon London: You’re talking about mass personalization.

Bill Schaninger: One hundred percent. We’re still saying we want an overarching plan. But what we’re saying is set the menu and allow people to choose from the menu.

Simon London: So that all sounds great in theory. But maybe, Alex, just bring it to life for us. Well, how does this actually work in practice?

Alexander DiLeonardo: So we were working with a global steel manufacturer, think 25,000 frontline employees showing up every day making steel coils, steel rods—from start to finish and shipping them out.

This organization has existed for a long time. They had some financial goals they were trying to achieve in the near term. Actually increasing their profitability and operating margin on the front line in terms of how quickly and with, you know, how few resources they could get those steel coils out. The problem statement was, “How do we get 25,000 people all aligned to this goal?” One of the ways was by having frontline employees in different pockets of the organization contribute ideas to increased efficiency.

There was a lot of variability in how manufacturing was being done. And so they needed to collect these ideas and then redistribute them so that best practices were kind of applied universally. All of the individuals in this organization had either a personal or a company smartphone. And we built an app that when they would log in, the first thing they would be able to do when they were invited to log in was actually look at what was the overall transformation goal. And they could see their unique role, whether they were an operator, an individual contributor, or a people manager.

So for the operator, that individual contributor said, “Hey, you actually have some great knowledge in your job that’s not getting out there. This is the role that you play. These are your jobs to be done that contribute to this rolled-up vision which is quite substantial.”

The next thing that we did is we said, “Understanding that you want to contribute these ideas to the rest of your workforce, what might be making it hard? Do you have structural hurdles? You know, do you not have the capabilities to articulate this or the time to talk to your management and share those ideas?

“Or do you think this is just not a good idea? You’re not bought in?” And then based on what those answers would be, they would actually configure a journey in the mobile application to say, “I actually want some reminders. And I want some help. And I want it Monday morning on my commute.”

So people who had structural hurdles would get help problem solving how to remove them. People would get skill- and will-building exercises so they could build their confidence and they could understand why this mattered.

And as they went, they would actually check in and share how it was going, whether they liked the content they were receiving. Whether they felt their behaviors were actually changing.

And so there were analytics that ran in the background which continuously optimized effectiveness of the interventions and experience that people were getting.

Finally, they would contribute ideas, those bottom-up innovation ideas that they wanted from the beginning. We actually piloted with 25 people, just 25 people. Over the course of two months without any top/down communication broadcasts, without any pushing, it has scaled to 5,000 employees that are engaging with this application every day.

We’ll eventually get to the 25,000 employees. But that two-month scaling, completely viral kind of methodology of rolling this out was pretty tremendous. And we saw behavior change. Two to four weeks, we saw a 10 percent improvement in the desired behaviors, that idea sharing, that driving toward efficiency. And we actually saw the frontline manufacturing costs come down by another 10 percent, just in this first pilot period.

Simon London: Right. Now as you’re talking about that, I’m thinking to myself, “Well, an app is fine. But an app doesn’t sound like a killer idea.”

Bill Schaninger: Yeah, look, it’s a great question. Because you could say, “Well, is that it?” Is it just an app? The mere fact that they were going down the road of the app, that they were going down the road of saying, “We are literally going to put it in your hands.” When you think about the substantive and symbolic nature of literally putting it in their hands.

As soon as something works, tell us. As soon as something doesn’t work tell us. How awesome is that? So just the modus operandi is so different. I mean, it truly is empowering the people. And I don’t say that lightly. You know, they got to say what works and what didn’t. They put it on the social wall. They put their own street cred behind an idea by putting it on the social wall and say, “Hey, I tried this. This is different. And it works.” We know that most of these things fail largely because it was, like, people from the center would come and tell you what to do. And we’re saying, “No, we’re going to start and flip that on its head.”

Simon London: So, yes, Bill, all right, I take the point. Actually, you know, customizing when things are pushed to me and so on actually is meaningful in itself. But just talk a little bit more, Alex, about the content. What is actually delivered to me, customized by role, who I am, how I’m viewing this program?

Alexander DiLeonardo: So I think there’s one layer which we’ve already touched on which is we’re customizing content based on role and the profile, where you fit in the organization. And that’s pretty straightforward. But potentially the more sophisticated layer of customization is around unique performance barriers.

They obviously are trying to do something. And we think about those different items that are standing in their way—there are three. The first is this sense of being not allowed. So are there structural hurdles that they experience in their job where they actually just feel like they can’t do the thing that they want to do because of those structural hurdles.

The second is they can’t. So they—they don’t have the time or resources or they don’t have the skills or confidence to actually do the thing that they want to do. And then finally, they won’t. Meaning they actually just don’t have the conviction or buy-in to do whatever it is that their goals are.

And the way that we get at these different constructs of roadblocks is through the mobile application coaching. So we say, “Would you like to learn a little bit more about yourself and receive customized feedback that you can either discard or use to your benefit?” And then they answer 15 questions. And we can get, you know, this very psychologically valid assessment of these different roadblock constructs.

Simon London: Just out of 15 questions.

Alexander DiLeonardo: Just out of 15 questions. And this assessment, we built on years of research. And we actually piloted it with a broad cohort of working adults. And we saw that, you know, when men take it, they answer the same way that women take it. Different ethnic groups answer the same way other ethnic groups answer. So we feel very confident that these questions are, you know, not biased. And they actually get at the items around roadblocks that we’re targeting.

Simon London: So, for me as an individual employee, you can get a pretty good sense of, “Is it that I feel I’m not allowed? Or I can’t. Or I won’t.” And what’s the mix for me personally?

Alexander DiLeonardo: Exactly.

Bill Schaninger: Simon, I think it’s useful to pick that one up, the idea that the blocking mindsets, the barrier mindsets, they really are a state, right. A state being who you are meets the environment you’re in. And certainly if—if the transformation’s working—

Simon London: It will change.

Bill Schaninger: An environment’s going to change, right? That’s why you need to keep checking in on it. And even something as simple as saying, “Well, I’m not allowed.” Knowing which version of it is there for you really drives what action you take. If someone says, “I don’t think it’s my job,” you probably do need to help them with role clarity.

I mean, somewhere along the line they missed it, right. You need to revisit it. But if someone is saying, “Hey, that’s not how we do it around here.” Well, that’s violating a social norm. That is a big deal.

OK, for these people, they’re going to need to see opinion leaders behaving differently before they move a nudge. As opposed to, “I just need to go over my job description.” That was the power of this. But then every time we introduce a new behavior, we better just check in and make sure we understand what’s getting in the way.

Simon London: Say a little bit more the, “I’m not allowed. I can’t.” And a particularly interesting one is, “I won’t.”

Bill Schaninger: We just talked about not allowed. Now, can’t is interesting. Can’t. Sometimes people will say, “Well, I can’t, because I don’t know how. And I actually don’t have confidence I can do it.” And the beauty of that is, it’s someone showing humility to raise their hand and say—and because very few people like to go, “I actually don’t know how to do this.”

And by the way, if you’re a leader and someone’s raising their hand and say they can’t, you darn well better help them. That’s your job at that point, right.

And by the way, if you’re a leader and someone’s raising their hand to say they can’t, you darn well better help them.

Bill Schaninger

The other two are little more interesting. Because they’re saying, “I can’t, because I don’t have the time or I don’t have the resources.”

And when you really get at it, what they’re saying, “It’s a great idea. But your priorities are not my priorities. And my list is different. And I’m solving for my list. Because that comes down to an allocation of my time, my money, and my people.” If you hear that, you have missed it out of the gate about landing the aspiration for how important this is.

And you should stop. You should stop until you land that. Because, you know, we’d like to say the axiom is, “Shouting louder is not going to work.” They don’t buy in to how important it is. Not only will they not give, like, in-role behavior, time and resources. You’re certainly not going to extra-role behavior. So you must address that directly so they actually get that part.

Deeper and often harder to surface—and you almost always have to work your way through—and I’m not allowed or I can’t. And then sometimes you go, “OK, we solved all that. We’re clear on the list. And yes, it’s your job.” And you’re still not seeing change.

Then you get down into the, “I don’t want to.” Classic, classic. What it really comes down to is in the head of the person you’re asking something says, “I am scared. I don’t know what’s going to happen here. I think this is going to mess up my power, my relationships, my influence, my—my own perception of my ability to be successful.” If you’re messing with that, the safest response is always inaction, always.

Simon London: You mean the safest response for the individual.

Bill Schaninger: One hundred percent.

Simon London: Because they’re facing fear.

Bill Schaninger: Freeze. Freeze, it’s, like, an amygdala response. The more the boss ratchets it up, “I want you to do this.”

Simon London: The more you’re going to freeze.

Bill Schaninger: Now, some flee, right. But most—they don’t have that option because it’s their job. So they will either actively be against you and undermine you, which at least you can see it. Or they’re going to freeze, right. And so that’s why this idea of, “If I can understand where you are emotionally, I have a dramatically higher likelihood of being able help you get across the, the—

Simon London: And the point is—

Bill Schaninger: Rubicon.

Simon London: That every individual employee will be at a different place—

Bill Schaninger: One hundred percent.

Simon London: And have a different mix of these. And just solving for the averages which until now is pretty much all we’ve been able to do is not going to be as effective.

Alexander DiLeonardo: Now we know in real time if it’s working. So what we found in the manufacturing example that I shared is, when you looked at the constructs of, “not allowed, can’t, and won’t,” all of the behaviors shifted positively based on the interventions that we’re providing.

Not allowed and won’t shifted much more positively than the can’t folks. And this was actually a little counterintuitive. Because I know psychology would say that it’s hardest to shift a won’t mindset. If they don’t have conviction of buy-in, it’s a really tough job.

But we were actually doing an OK job of helping them change their behaviors. But the folks that didn’t have the skills or the confidence, we got real-time feedback, I’m talking days, weeks. We knew immediately that our interventions were not working for those folks.

So what did we do? We clicked down one more level of granularity. We broke out the tasks into different subtasks and said, “Here, practice this. Teach this to somebody else. Do it. Build confidence. And come back.” And then we saw the can’t roadblock shifting. So it’s hard to imagine just splitting these people up into the archetypes without also getting the real-time feedback on what’s working and for whom. I think there’s a magic when you mix both.

Simon London: So the thing that’s going through my mind though is, however good your content is, however personalized this is for me—and I think you alluded to this earlier, Bill—in any big organization, there is going to be a broad mass, maybe even the majority of employees, are either cynical, change fatigued. They’ve got other things going on in their lives. How do you overcome that? How do you actually get people to engage with this?

Bill Schaninger: I think the cynicism sits with both frontline employees as well as leaders. You know, you have leaders who are under duress, under time pressure. And if they could treat their employees as homogenous, it made it simpler. Employees, on the other hand, felt rightly that there was no—there was no “I” there, there was no individual there. It was everyone should be treated the same.

And the minute they get an email, they should magically just get it. So I think that cynicism comes in both directions. The data is beyond question. Right, when these things fail, they almost always blow up because leaders lack support, they’re not willing to make the tough calls, the culture doesn’t support it, and/or you never bring the bulk of the place along.

And I think what Alex and the team have really been pushing is trying to balance this tension of the need for scale and the need for customization. But always anchored back to “what’s the point for doing all of it?”

The influence model’ approach

Simon London: So I buy that answer. But presumably you also need to address, like, the formal mechanisms within an organization as well. How are people assessed? How are they compensated? How are the roles defined? You still need that architecture.

Alexander DiLeonardo: Yeah. And I would argue two things. I would argue, one, that the large components of the influence model, all four quadrants—the way we think about it, are embedded in this approach. You’re just hitting them at a different level.

I would also argue though that this approach should be one thread in a broader, kind of laundry list of things that you do in a healthy transformation, which would include what leadership does. You know, hardwiring the change. Whether it’s from a process perspective or reinforcement and compensation mechanisms.

Simon London: Yeah. So it’s not a magic bullet. But having said that, it—it’s pretty powerful and could be a big unlock for how you address change at the level of individuals.

Alexander DiLeonardo: Right. And I would argue that it’s probably going to work more effectively when it is tied to all those other threads of a transformation, as opposed to a stand-alone stream.

Simon London: So just double click, if you don’t mind, this idea that you’re getting constant feedback as the management team, as the team running this change and transformation program. You’re getting feedback through the app and other mechanisms. How does that actually change how you run a program?

Alexander DiLeonardo: The analogy that I go to is how we develop software. So from the beginning of when we were doing computer science and developing software, we originally followed the quote, unquote “waterfall methodology.” So you create the vision, you collect requirements, you do a detailed design.

Would you like to learn more about our People & Organizational Performance Practice?

You build whatever it is. You test it. And then you roll it out, and in a lot of ways, that’s the way we’ve been running the transformation office for the last 20, 30, 40, even more years. What this gives us the opportunity to do is to shift to an agile development model.

Now, we can say, “Let’s get started. Let’s see what happens after two weeks. And let’s measure at each iteration.” We can measure things like how many people are involved? Do people like the content? Are they learning? Are they changing their behaviors? And are bottom-line outcomes actually shifting?

Imagine being able to have a control room where you have a window into that every day, every week, every month. And then you take things out that aren’t working. You put new things in where you need them. You see—you know, an archetype or a segment of the population emerge that’s very healthy.

How do we illuminate them and actually elevate them so they’re sharing with other people?

Simon London: Do you need different skills, like, in that control room? Do you need different people? Do you need behavioral scientists and analytics people in the way that you might not have done? Certainly not with the same level of intensity before?

Alexander DiLeonardo: What we foundand we’ve been doing analytics in organization and HR now for almost a decade—what we’ve found is that a lot of the core skills that were relevant before are still the dominant skills. You still need to know a lot about the content.

You need IO [industrial-organizational] psychologists that actually understand the theory of how humans react to environments and change their behaviors. You need business buy-in. You need business sponsorship. And you need, you know, HR professionals that are data savvy, that can drive the process, that can create change in the organization.

So I would say keep the core skills that we’ve relied on. Just change the methodology.

If I could build on that. You need enough skill in the room to be brutally clear on what performance impact you’re looking for and be able to disaggregate that down to the individual level. Many leaders are really good at waving their hands towards a big number.

They may not be capable of breaking it down, disaggregating. So you need that help. Someone with the courage enough to set the big number, someone capable of disaggregating it. That’s for one. You need someone who can say what kind of skills does that require to be able to do it. No doubt.

You also need someone who brings enough due care that the environment you’re going to create in pursuit of helping someone change their behavior is done thoughtfully. We’re not interested in creating undue pressure. We’re not interested in creating a contagion that generates fear, right?

So I do think there is a skill there. And probably does have more of an IO background to it, at least from a thoughtfulness standpoint. Right, because, you know, we hear a lotta things called nudge. Unfortunately, when you look at a lot of it, it’s more like nag. It’s a reminder email sent to everyone. Here’s a good rule of thumb, if it’s sent to everyone, it’s not a nudge.

Simon London: Right. So if done badly, this just turns out to be personalized nagging? Or not even very personalized.

Bill Schaninger: Broadcast nagging. Right. I mean, and then, look, I’m being just a little dismissive. But just understanding. What really motivates a person? Do they need to feel part of the group, in which case you should tailor it to how the group is doing? Do they need to win desperately? Show them a league table. Do they need to live up to what they’ve said? Remind them of the commitment they made.

That sort of tailoring does require skill and thoughtfulness. And that’s the kind of thing we’re getting at.

Customization and data privacy

Simon London: Now, the elephant in the room here, and we haven’t really addressed it yet, is data privacy, surveillance. So I find the steel worker—the app is on my phone, I’m starting to interact with it. What am I buying into? What’s happening to my data? Am I worried that, you know, the boss is looking up whether I’m using this?

Alexander DiLeonardo: Good data-handling practice, and I think this aligns a lot with sort of the guidelines that are set out by GDPR, has three points. One is optionality, the second is transparency, and the third is protection. Optionality means that you are not forcing anybody to participate in this.

Good data-handling practice ... has three points. One is optionality, the second is transparency, and the third is protection.

Alexander DiLeonardo

They are asking for help and they are getting help. And if at any time they find it not helpful or they don’t want to participate, they can actually choose to wipe all their data and disengage from the platform. The second is transparency, which means we’re going to be very up front with what information we’re asking, why it’s helpful and how it ties to the broader strategy. And then the third is protection.

And protection has two parts. One is just good hardware and software and methodology around how you handle data. But potentially more importantly is what do you do with that data and what does it look like when you’re reporting back to organizational leadership.

And so what we’ve done in the steel case, for example, is anything under ten records, that doesn’t actually get surfaced to insights when we’re showing results to leaders. And so we’re offering these individuals the help that they’re asking for. But it is optional. They have transparency into why this matters. And they’re actually getting the protection.

Simon London: So the boss is, in theory at least, not going to know whether I as an individual am even participating in this program or not.

Alexander DiLeonardo: Exactly. And in this case, we’re actually acting as the data controller instead of the data processor. Which means we’re not handling your individual employee data on behalf of your employer at the company. We’re handling your data on behalf of you. You have control over it and you can wipe it at any time. And you’re actually the person to whom we’re accountable.

Simon London: Right. Because that seems to me a very important set of principles. And I can imagine that, you know, some companies frankly may not do this right.

Bill Schaninger: Well, the opt-in part is a big deal, right, if you think about choice. I mean, nudge at its core was intended to be choice architecture, right? And so running through and through this is choice. And I also think it’s an opportunity for the employee just to choose how they’re going to experience the transformation.

What we’re trying to say is, but it doesn’t have to be that way for everyone. You actually do have an option to engage with this differently, to give yourself a more informed, a more involved, a more enlightened way of saying, “Maybe I can thrive in this environment.”

That’s the hope, because you could say the organization’s going do this regardless, right? So why not put your hands on the wheel a little bit more as an employee?

Simon London: So I think an interesting question here is sort of why is this happening now? Is it just the analytics and the delivery mechanism? And all that is available. Or is it actually that something’s changing up in a macro environment that is making this more important to do?

Alexander DiLeonardo: Yeah, there’s certainly a component that is technology evolution, though—many would argue that the tools that we’re using here have been around for five to ten years. But I think from a macro perspective, and I know it’s a cliché, you know, the pace of change has never been as great as it is today. And it will just continue to increase and accelerate.

And when you have that context, where people are not going to be in the same job for their entire careers. They’re going to change jobs maybe multiple times a decade, maybe multiple times a year. With all of that entropy also comes a little bit of opacity.

And so for the worker, I think it’s very exciting that we can help people tie themselves to the overall strategy and the overall goals, whatever they are, of the day, of the week, of the year. Help them understand where they fit in, and then help them diagnose their roadblocks. And essentially, just build a road to be higher performing and to contribute more.

Bill Schaninger: You know, I think there’s an interesting challenge for leaders if you come to grips with the idea that they’re asking their employees to take a less personalized experience at work than they get at home or as consumers.

Their experience as consumers, their experience as patients in a healthcare setting is largely configurable and customizable to them. And we continue to ask employees to say, “You’re just like everybody else.” I mean, it’s remarkable, right?

And I think part of that is leaders have to come to grips with the simplifying devices they’ve used to make their jobs understandable to them have to go away. They can’t keep treating the entire organization as a monolith, as a block.

Changing how we change

Simon London: So we need to change how we do change?

Bill Schaninger: Change how we do change. Employees deserve and are increasingly demanding more. You don’t have the luxury of these things taking five years anymore, three years anymore, right? What we’ve learned from agile ways of working suggest “if you build it, they will come” is not going to work.

Simon London: So is there an element of this that, instead of going through sort of endless five-year change programs, which probably only make it two or three years in before they get replaced by the next five-year change program, actually, we’re getting into more of a continuous improvement, continuous change, but individualized change mindset.

Alexander DiLeonardo: Yeah. What if it’s just called leading your organization.

Bill Schaninger: Right. I mean, all the way back to literally your first question, which is how we make money and how we run the place. Let’s maybe change how we run the place.

Simon London: Yeah. All right. Well, thank you for a fascinating and insightful conversation. Thank you, Alex and thank you, Bill.

Bill Schaninger: Thank you.

Alexander DiLeonardo: Thank you.

Explore a career with us