Talent Management

AI is transforming the world, and it starts with the right compute. ARM is the AI compute platform trusted by global leaders, proudly NASDAQ listed, built for the future. Visit arm.com discover.

Hello and welcome to the HBR IdeaCast. I'm Paul Michelman and I'm on the phone today with Peter Capelli, the George W. Taylor professor of Management and Director of the center for Human Resources at the University of Pennsylvania's Wharton School. Peter, thanks for taking the time to talk today.

My pleasure, Peter.

In your March Harvard Business Review article, Talent management for the 21st century, you lay out a case that tells us that our approach to talent management is, well, seriously flawed and we need to get our act together before we dive into the details of your ideas. Will you begin by giving us a simple definition of just what talent management?

Sure. And it is a very simple definition. Talent management is just trying to anticipate what you think your needs are going to be as an organization in talent. The people that you'll need, the skills you'll need, the number of people, and then trying to set out a plan to meet those needs. So it's pretty simple definition, simple task, same kind of task that you see businesses facing in most aspects of their product markets.

Okay, but there's something wrong with the way organizations are going about doing this today. What is it?

Well, I'd say there's two things that are wrong, and the most obvious one is that companies have just given up altogether. Surveyed data and survey evidence we can put our hands on suggests that the majority of companies in the US about 2/3, don't try to estimate at all, don't do any planning at all about what their needs will be in the future for talent. And another 2/3, and it could very well be the same. 2/3 don't have any processes in place to try to meet the demand. So they're not trying to forecast, they're not trying to meet it, they're basically just throwing up their hands. I'd say the other group is the group of older firms. And a lot of the best practice sort of discussions, a lot of the consultants, a lot of the current literature that's still around is stuck in an earlier paradigm. And the earlier paradigm was one where companies knew with pretty good certainty what they were going to be building, what they were going to be selling years and sometimes decades in advance. So this was a period when product markets didn't change much. Business strategies were long term and companies could just work backwards from those strategies and those plans and know with pretty accurate certainty what it is they were going to need going forward. The reason that model doesn't work anymore is because business is way too uncertain for that kind of planning model to work. And the problem with this older approach is that it assumes that you can plan your way around uncertainty, which.

Okay, Peter, at the heart of your suggested new approach to managing the flow of people into and within organizations are the principles of supply chain management. Why don't we walk through the four principles, beginning with the first?

Okay, so I'd say the first place to start with this is basically to recognize what the underlying problem is for employers, and that is uncertainty. It's very difficult to know with any kind of certainty what it is that companies will be doing a few years out, what the demands for talent will be. Therefore, and rather than assuming that we know which is what the old model does, the best thing to do is to be candid about the fact that you know what, we're uncertain about this. We may have a guess, but our guess is quite likely to be wrong. So the first thing to do is to say, well, really, how certain is our best guess of what our talent will be? How certain are we of that? And once you have that estimate, then you can begin to think about how to meet that demand. So let's say in a typical company you might say, let's say we think our need is for 100 new engineers of a particular type next year. How certain are we of that demand? Well, let's say we're not very certain of it at all. Then from there we have to ask a second question. And this is where things get interesting. And the second question, which is a basic supply chain question, is just what happens when we're wrong? What happens if we're not right that we're going to need 100 engineers next year? Well, the first thing we know is there are two ways you can be wrong. You could have too many engineers and you could could have too few engineers, and you can calculate the cost of that pretty accurately. That's not a forecasting problem. That's simply an exercise in calculating what happens if you got people sitting on a bench versus what happens if you have to scramble and run around and find talent. I'd say the big change for most companies now is that going long is actually more costly than going short. A generation or so ago, if you didn't have have enough people with the right talent, you just couldn't get anything done. Now if you don't have enough people inside, you can go outside and hire them. So you Sort of think about how many people do we want to develop inside our organization. And then if we fall short, we can go on the outside and hire talent in sort of just in time. The bigger problem is if you have too many people, you have too much talent. And this is where you can really see the problem with the old model. The old model, people talked about having bench strength and what a great thing it was to have a deep bench of talent. Well, the other way to think about a deep bench of talent is that it's inventory. It's excess inventory if it's sitting on a bench. And the worst thing about human capital inventory is that you got to keep paying it. And it gets up and walks away because it gets frustrated not having anything to do. So going long for most companies ends up being really very costly. So what do we do? Well, let's say we've got our best guess is we're going to need 100 engineers next year. And we know that the cost of going long are much worse than the costs of falling short. Well, if that's the case, we sure don't want to produce 100 engineers because we think there's a pretty good chance that we could have too many. So we end up producing less. Maybe we produce 90, or if our estimate is really uncertain, maybe we produce 80. And on balance we'll probably fall short, in which case we have to go on the outside and hire more folks. But it avoids the problem of having too many engineers, in which case we gotta pay a lot of money laying them off and losing all the investments we've made in them. So that's the first. That's sort of a long winded description of the first principle.

Peter, a quick follow up question on this one. Are there some industries in which the intensity of specialization is so high where this principle doesn't really work well, you know.

So I guess that issue would be to say, what if we can't hire people on the outside market? And it could be that you're so specialized in your skill needs, it could be that the market is so incredibly hot that you just can't find people on the outside market anyway. And that certainly changes your calculation, but it also doesn't obviate the basic idea. And the basic idea is you got to manage uncertainty. So let's say there's no possibility of hiring people on the outside. You still have the problem of managing the uncertainty. In this case, if the work won't get done unless the people are there and there's no ability to hire on the outside market. Then you take your best point estimate, say it's 100 engineers again, but this time you probably better overshoot that estimate. You might want to produce 110 engineers, and maybe you want to produce 120 if you're really uncertain as to how many you'll need. Because if you fall short, you're dead.

Okay, Peter, great. So principle two, as you note in the article, is adapt to the uncertainty in talent demand.

And I think what this one means is that we have to find ways to be responsive. And by the way, this is exactly what people who study business strategy are telling us, that the new environment demands a strategy, responsiveness, adaptability. And some of these things apply just straightforwardly from supply chain management. You see companies, for example, that have bottlenecks in the supply of talent. The secret agencies in the US Government, for example, for a while have had backlogs in getting people through security clearances of sometimes up to two years. If you look at that, you say, well, if this is a supply chain problem, what would we say about this? Well, of course you'd like to fix the bottleneck, but they can't typically fix it. It's not within their control. Well, if that's the case and you're developing a backlog of people about two years long who are just sitting on their hands waiting to get the security clearance, what that means is you've hired too many people. The pipeline of talent can't be any bigger than the narrowest part. That is where the bottleneck is. And if you've got people sitting on their hands for two years waiting to get through a bottleneck, the problem there is you're just hired too many people. And I think there are other principles. Maybe the biggest one is simply the idea of using portfolios to balance out some of the uncertainty. So, for example, succession planning, which is very popular in companies, but basically it's the idea of bringing planning down to individual level. This person will be in that job in five years. That's highly uncertain. Both the person has to be around and the job still has to be around. Be there in five years for that to work out. And we know this stuff just goes out the window. Anytime there's a big restructuring, a new CEO comes in. So rather than doing that, you see companies now moving toward more of a portfolio approach. Some people call this a talent pool approach. Rather than planning for a person to go to a particular job, we look at a group of roughly similar jobs and a group of people who might staff those jobs. And over Time. There could be some bouncing around in the people up and down, in and out. There could be some bouncing around of these jobs. Some go away, maybe a couple of others come in. But with a big enough group of people and a big enough group of jobs, pretty good chance we'll make a match between them.

Peter, as we think about talent management in terms of the language of supply chain management, how do we keep in mind that people are a lot more complex than say, flat screen TVs?

Well, some of this will come up in a second in the fourth principle. But I would say the big issue is we recognize this on the inventory side, and that is that you can't park people on a bench in a supply chain the way you can if this was, you know, producing widgets or something like that. You just park them and leave them there and off they go. In this context with people, you park them on the bench and they pick up and leave and go someplace else because they're tired of sitting around. So I'd say the basic principles apply more or less straightforward to people as well. I think the big difference comes when we're thinking about this question of inventory.

Okay, so let's move on to principle three, improve the return on investment in developing employees.

So the big problem that employers have now is that they make an investment in employees, then the employees leave. All the investment has gone out the door with the employees. This is maybe the biggest problem, a huge problem for the economy, because it certainly appears that companies are making fewer investments in their employees. They're very reluctant to make the investments because they're afraid they're just going to lose them. And the way the labor markets are structured and employment relationships work, it's a hard problem to solve. So there are some ways to solve this problem. We're basically thinking about how do you structure things in a way to recoup the investment. And certainly we want to rely on much more on work based learning. It's cheaper because the employees are contributing along the way. And I think the biggest innovation is to find ways for the employees to begin to pay for some of the investments themselves, sharing some of the costs of development. It's a fair thing to do because the employees seem to be getting all the benefits from development. Now. As soon as you invest in them, they can pick up the investments and take them someplace else. And the employer finds they pay for the investment. And they also got to pay a wage increase to keep the employee from leaving. So some ways companies do this is they sort of ask for volunteers. They Say, look, we got a development experience here. It's one that might broaden you. It involves a new type of work assignment. And what we find them doing is they say, look, the key here is you got to keep doing your regular job while you do this development assignment. So it ends up being more work for the employee. It's not costing the employer anything because the pay isn't going up and the employee's contribution is not going down. It's staying the same. And if they're doing work of value to the company, it's actually going up. So who's paying for that experience? Well, the employee is, and therefore the employer can afford to make it.

Peter, you also talk about the importance of staying engaged with your alumni.

Well, there are some companies that have gotten creative about this. There's a company called Select Minds that just helps client companies keep track of their alumni, keep them in the loop. Deloitte has got an interesting program for their accountants now where if you leave the company and you're home, let's say with kids, they provide training for you, they pay to keep your accounting credentials up to date, they let you know what's going on in the company, they keep you in the loop so that when you decide to come back to work, you go back to work for them and they can pull you right back in and you sort of hit the ground running again.

Okay, and that brings us to Principle 4. Preserve the investment by balancing employee employer interests.

Yeah, like the other principle, this one starts with a problem. And the problem is that employees now need to have a say in the management of their career. The old days, companies used to say, look, time for you to go take this job in a different country or a different location or move someplace else. Part of your career development. And the employees, frankly, didn't have much of a choice. Now they have a choice. And the choice is they can quit the company and go work for somebody else, which is a huge lose for the employer who's trying to develop their skills and has already made investments in them. So we have to incorporate the employees interests in this process. And most employers have started to do this with these online job boards, internal job boards, companies have. Virtually every big company in the US now posts all its vacancies, or almost all its vacancies. So how we figure out how to balance the interests of the company and the interests of the individual employee here is really the trick. I can't say we've exactly solved this problem, the employer community anyway. But the general approach here is to try to incorporate the employee's concerns and also give them a say over the process so that they're likely to stay put.

Okay. Peter Cappelli, you've given us a great deal to think about here. Thank you very much.

Thank you. My pleasure.

We hope you've enjoyed this week's program. For more great ideas from Harvard Business, please visit our website at www. Harvardbusiness. Org. Thanks for listening.