Think Horizontal


Think Horizontal

The case for organizing your company in teams.

The Conference Board Review

No, top executives aren’t omnipotent—not even you. Having authority, a dedicated assistant, and a lovely window view don’t make you an expert on everything on which you need to make decisions. You can rely on your managers, but even you can’t peer inside their hearts to see if their counsel is impartial and comprehensive. 

One answer: teams. In Great Business Teams: Cracking the Code for Standout Performance (Wiley), management consultant Howard Guttman lays out guidelines for bringing together small groups, with members holding each other accountable for results, and best using their recommendations. The ultimate goal is what Guttman calls a horizontal organization, one whose key decisions are made by layers of teams, most of which come together to solve particular problems or answer questions—and then dissolve. “This is all about speed, moving things quickly,” he says. “It’s not about creating bureaucracy—ground rules, secret decoder rings. It’s not about creating teams for the sake of creating teams. I don’t want to be the Johnny Appleseed of teams.”

In working with companies over the years, Guttman has turned a number of executives into advocates for team-based decision-making, and two of them—Larry Allgaier, CEO of Novartis Global OTC, and Grant Reid, president of Mars Drinks—accompanied him to The Conference Board’s New York offices to speak with TCB Review editor-in-chief Matthew Budman.

We’re supposed to be in the middle of a leadership crisis. Where do teams come in?

Howard Guttman: Leadership is no longer about one-person rule. Today’s organizations are too complex and far-flung for the “leader” to make all the decisions. That doesn’t mean that organizations should be ruled by consensus—that’s dysfunctional, because it’s virtually impossible to get everyone to agree on every issue. Effective leaders balance the need for speed and a quality outcome with the need to involve members of their team.

Grant Reid: My thought process around leadership used to be that I would make decisions, everyone would come in, I would give them the charter for the day, and they would all run off. Now I recognize that it’s not about me—it’s about getting the right team together, with the right expertise in that particular area, and getting the best out of that team. No one is as smart as everyone.

To a lot of people, setting up team meetings will unavoidably sound like, well, setting up more meetings. Don’t you two already spend enough time in meetings?

Reid: Strangely enough, I actually spend less time.

Larry Allgaier: It’s definitely less.

Reid: There are lots of meetings I no longer need to go to. If you’re command-and-control, you have to be there to command and control; no major decision, no major recommendation can move forward without you there. When you have teams capable of making decisions, there are a lot of things you can let go. When I took over Mars Drinks, we were meeting globally, as a management team, every month. People were flying in from America and different parts of Europe; it was very time-consuming. Now we meet four times a year. We have scheduled telephone conversations with the whole team once a month so everybody’s clear on what they’re doing, and then we get on with it. 

Allgaier: High performance definitely goes with fewer and shorter meetings, because people are accountable—they know what they need to do, they come to the meeting, and they’re in and out.

Guttman: But remember that this is not Pleasantville. We’re not all sitting around a table agreeing on things. That’s not the way it works. I was with my own core group of consultants this Saturday, in my office, discussing some contentious issues we had to deal with as a firm. In the end, we assigned a subteam to handle the issue—and I’m not on it. That tells me that it’s working; that’s what’s supposed to happen. In an old-fashioned model, where people believe that because I’m the leader I’ve got to be “in the game,” no doubt I would be part of that subteam.

Reid: When I was fairly new as head of sales and customer care for Mars Snackfood US, I went to a planning session, and I figured there’d be six or seven people there. I walked in, and there had to be thirty-five people in the room. And I thought, “Well, OK, it’s a big meeting, but at least all the key people are here, so we’ll get to a decision.” The customer asked about a very small thing—maybe about, say, changing a package size from four ounces to four and a half ounces—and I thought, “Well, we’ll answer that and move on.” You’ve never seen thirty-five people look at their shoes so quickly. Nobody was clear on who decides what the size should be.

That’s what this is all about: “Who’s the point person on that? Joe, it’s you—when can you get back to the customer?” “Well, I’ll need to do a bit of work—I’ll need some help from Fred from marketing. We can get back to you in two days; the three of us can meet; we’ll set it up offline.” The issue’s handled, and we move on, instead of us all sitting around.

This may be easy to see and even to implement once a company is already horizontal, with teams in place. But what about at first? How do you convince executives—who all believe that they’re indispensable—that these meetings are worth their time?

Allgaier: When you set up these teams, you strive for the minimum number of people involved.

Guttman: And while this makes sense intellectually, if your organization prides itself on inclusion, it can feel off-putting and exclusive to make one’s ability to contribute the new standard for meeting attendance. So you’re really trying to change people’s belief that their value to the organization is tied to the number of messages they’re copied on or how many meetings they are asked to attend.

Which is why, if you’re going to start from scratch, ideally you have to work with the highest-level team so they get that point. And then it makes it easier to get traction down below.

Allgaier: I think more people find it a faster, more fulfilling way to operate, and they make the transition fairly quickly. It’s just a better way to get more things done. And if you do have a player who’s not going to get there, they self-select out of the team model. You end up by default getting the right kind of players around you over time.

Guttman: The first challenge, though, is around people being willing to put it out there on the table.

Presumably, all companies already have some institutional teams set up—maybe with thirty-five people, but still.

Guttman: Many times we will go into a company and look at brand teams or category teams, and we’ll ask how many people are on the team, and they’ll say, “Well, we’re not sure, maybe twenty-five or thirty”—and right off the bat you know it’s dysfunctional. They don’t have any ground rules; they want to think of everybody who matters; they don’t want people to feel left out.

On what teams are you currently serving?

Reid: I’m part of the Mars Inc. global management team. There’s my own management team at Mars Drinks, which I’m the leader of. And then there are multiple subteams that I’m on, depending on what the issue of the day is and whether I have particular expertise in those areas.

Allgaier: Similar for me: My boss has a leadership team for three business units, so I’m part of his team. Then there are various small councils.

Is there a danger of an organization setting up too many teams? It seems as though one could wind up delegating a lot of routine day-to-day decisions to teams, to the point where almost everything becomes deliberative and subject to discussion.

Guttman: The goal is to have the fewest number of teams. Some teams are a permanent part of the organization—freestanding teams that must remain intact by virtue of their charter. Then there are project teams that come and go, based on needs that arise. What would be dysfunctional is an organization that created team after team after team. People would drown in the confusion.

Reid: You can’t keep setting up teams that do nothing, because the first thing that happens is that teams sit down and say, “OK, what are we trying to achieve here?” And if there’s no real goal, then each of the team members will want out. It’s a self-governing process. Back in the ’80s, when I was working in Europe, you’d have all these teams that were more like bagel clubs—everybody would come in, have a nice chat.

Guttman: Teams in search of an agenda.

Reid: Everybody was pleasant to each other; nobody really called anybody out; if somebody said something that didn’t make any sense, it would be noted sagely. The difference now—certainly in my organization—is that people ask, “What are we trying to achieve here?” And if you can’t answer that question, then chances are that team doesn’t have a purpose.

Allgaier: I agree with Grant: If there’s no need for a team, it should quickly go away. The lifetime of a team can be very specific: You set up a workstream, they meet, they’re clear, they knock it out, and then it dissipates.

Guttman: The only time it’s different is when they’re institutionalized, intact teams—the leadership team or the top team of an individual function. And a company should have the fewest institutionalized teams possible. Other than that, based on particular needs, you should continually be pulling together the right players, determining their agenda, reaching closure, and moving on.

What issues should not be assigned to teams?

Guttman: Teams are not appropriate when you are dealing with everyday, unilateral, technical, or functional calls. You don’t need to have the marketing people getting input from the finance people; there’s typically no value added there.

Should everyone on a team be at the same organizational level? If not, won’t people naturally defer to their supervisors?

Allgaier: You want to set up an environment in which people provide expertise and bring certain things to the table. It shouldn’t go by level.

It shouldn’t, but how does that work in practice?

Allgaier: We mix it up so that not everyone is at the same level. Sometimes it adds value to have a more junior person on the team. When you have a high-performance team, people recognize value no matter what level in the organization it comes from.

But that junior person is still sitting in the room with her boss.

Guttman: There’s no question that instinctively, people think hierarchically. When a person starts a new job, the first thing they’re thinking is, “To whom am I reporting, and what are the expectations for me?” When you put people from various levels on a team, as Larry said, the people on the team should be there because they add value. And the reality is that the people higher up on the ladder often know less about a given issue than the people on the ground. So the question of being deferential is antithetical to the whole notion of working as a high-performing team. If you had players on the team who were deferential, it would mean that they weren’t the right players, and it would also indicate that the higher-level team members might be inadvertently sabotaging those people’s ability to contribute by taking up too much airspace.

Allgaier: If a CEO doesn’t let a younger, more junior person add value, it ultimately sabotages the leader.

Guttman: In one organization, we worked with a person who had taken over a division. He frequently held brainstorming sessions with his team, in which he was always grabbing the marker and writing on the flip chart. When people expressed their points of view, he would always one-up them. He was trying to give the impression he was playing an egalitarian, high-performing team game, but it was his game. He was operating like a king.

Reid: And of course, typically, that same person holding the marker then says, “Give me some feedback”!

I don’t think it’s natural to think in an egalitarian way—you grow up going to school and taking a job, and there’s always a leader who’s responsible for creating the environment. I’ll give you an example from Mars: I worked with Howard back in 1995, when I was much lower down in the company.

Guttman: I was 26. 

Reid: Yes, with a full mop of curly hair. And we tried to put high-performing teamwork in place, four levels down. We were very willing. But because we hadn’t onboarded people further up in the organization, they really weren’t looking for feedback, and we stalled; it wasn’t successful. Unless people at the top are looking to make it work, it won’t. When I took over Mars Drinks, it felt more command-and-control there. I came in and said to people, “OK, what do you think?” “Wonderful, Grant, wonderful.” And even I know I’m not that good! You have to create an environment in which it truly is comfortable to give your boss real feedback. That has changed completely since those early days, and the leaders in place throughout Mars are now some of the best examples of high-performing-team role models that I have met.

Guttman: In the research I did for Great Business Teams, when asked about creating and sustaining high-performing teams, many leaders said one of the more difficult things was to be able to be that receptive, that vulnerable.

Reid: It’s not something you can turn around overnight, because, again, it’s just not natural to give and get all that feedback, especially when you’re the boss. I like to say that the truth shall set you free, but the process shall make you miserable. I spend more time with my direct reports than with my boss, so who’s going to make me better? My direct reports.

Howard, you envision a horizontal organization “ruled by high-performance teams.” But presumably, even decisions and recommendations of a high-functioning team are still subject to veto from the CEO. Larry, how often have you had to overrule the recommendations of a team?

Allgaier: I don’t think I’ve turned over too many decisions—though you’d have to ask my direct reports and see what they say, right? When I have overturned something, it might be really specific to a business issue—where someone recommends killing a project, and I might take a contrarian view to keep that project alive. Usually it’s because I haven’t engaged that team early enough, with enough clarity, about why that project should exist and where it should go. They may be saying it’s going to take this amount of resources, this amount of money, but I haven’t had a deep enough conversation with them about the strategic relevance of that project. It becomes a learning experience.

Reid: If you empower people and are clear that the team is offering a consultative recommendation, and everyone understands how the decision is going to be made, very rarely do you have to overturn their decision. You sit down with them and talk about the pros and cons, and they feel like they’ve been heard. Issues arise when people slide a recommendation under the door and white or black smoke comes out; then people feel like, “I wasn’t really part of it; I don’t understand why that decision was made.” In my experience, if you sit down with people and say, “I’m not going to go with the recommendation for these reasons, and I may be wrong, but here’s what I think is happening,” then ninety-nine times out of a hundred, people say, “OK, I get that,” and they stay engaged.

Guttman: Now, you described an organization “ruled by high-performance teams.”

That’s your phrase. On page 10.

Guttman: That was my phrase? It’s a very good phrase. But I wanted to say that whether or not the leader delegates decision-making power to the team depends on the answers to these questions: “What evidence do I have that the team is ready to step up?” and, “If I delegate, will I get a good night’s sleep?”

Allgaier: Empowerment has to be earned over time. You have to see it.

Reid: There’s a difference between abdication and delegation: You have to trust but verify, and the more the teams get things done right and on time, the more you can delegate, and the faster the meetings are—there’s no telling stories or going ’round the table; people focus on the issues.

Now, within a team, obviously there are different ways to arrive at decisions and conclusions. Which has worked best for your teams: democracy? consensus?

Allgaier: We have something called agree/disagree/commit: You can agree or disagree, but when it’s out there you have to commit and execute with all your energy and passion. It’s not necessarily a consensus—in fact, the majority of decisions are not consensus. The disagreement is important—you’ve got to get that out in the open.

Reid: My feeling is that if you try to make a decision by consensus, you water it down to the lowest level. Virtually none of our decisions are consensus—even choosing where to go for dinner, because I can’t get everyone to agree. The bar is: Can you live with it?

Now, some things are going to be unilateral: I’m going to say, “Here’s what the overhead percentage needs to be for the company,” and that’s what it’s going to be.

Allgaier: And each person is accountable. You tell the person who’s on the team, “You have to deliver your function.” It’s their responsibility to go back to their people and get what they need and make sure they fully represent. There shouldn’t be any hiding behind, “You’re just getting my opinion, because I didn’t talk to so-and-so.” People hold people accountable for delivering all the input they need to bring; they can’t say, “My boss feels differently”—it’s their job to cover their boss and get their boss’s input and bring it to the meeting. If not, the complexity can get overwhelming.

Does the CEO play a different role in a horizontal organization?

Reid: For me, it’s not about your organizational structure. It’s about being mentally horizontal. You can have a traditional management structure. But the majority of major things will involve multiple players and multiple teams.

Guttman: If you run a company and have one person at the top operating like a king or queen, there is no enduring organization, and when that person leaves, the thing falls apart like a house of cards. So if you’re trying to create an enduring organization over the long haul, your only real option is to create these high-performing teams and have distributed decision-making—putting power and authority in the hands of teams and their members, provided the conditions are right, the ground rules in place, and the players sufficiently evolved to deliver maximum payoff.

Then there’s the issue of complexity: Some of these companies are so complex that for someone to play a command-and-control game today is about as outdated as leading by divine right. That world doesn’t exist anymore. I mean, Grant, could you be command-and-control even if you wanted to? 

Reid: No. In a global world, you can’t be on top of every situation, every geography.

Now, a few years ago, Mars decided to take the radical step of becoming a horizontal organization. But a company can introduce teams effectively without redrawing the whole organization chart, right?

Allgaier: Sure. We’ve done some of both. We’ll set up a team on a specific workstream that we need, and it lives for a certain amount of time, until it gets a process in place and has a certain effect on the business and then it goes away. We have set up category teams, which was a formal organizational change, and it’s gone very well.

Guttman: In real life, there still is a structure; there’s still a pecking order. But if you have a senior team at the top that plays it horizontally, with peers holding peers accountable, and you can replicate that in each of your functions, you are going to speed up time to market and decision-making; the leaders will be able to spend more time strategizing and less time supervising. It’s a game-changer. Frankly, if you have the right players—

Allgaier: —you can overcome a lot of imperfect organizational setups.

Guttman: They should be able to rise above their circumstances to make things work, or to change things, regardless of what the structure looks like. But as we were saying, it’s counterintuitive for people to be that candid and hold each other accountable. If you put people in a room, they’re not brand new to each other; there are preexisting relationships; they don’t want to alienate each other. Everyone dumbs down their contribution. So this isn’t effortless—particularly if, as you said, you have people at different levels. People will be more, or less, inclined to offer their points of view.

Allgaier: I think that’s a key point. For us, it’s less about what the organization should look like—what matrix—and more about how we operate.

Howard, you say that operating horizontally “requires making private, internal changes—not just organizational ones.” Change advocates always call for people to alter their mindsets. How much can one reasonably expect executives to change?

Reid: In my experience, when you bring people together, everybody’s uncomfortable with change to different degrees. Some say, “OK, I get this,” and others say, “This is never going to work in my lifetime.” Those people frequently self-select and go off and do something else, and for the others, the team will hold them accountable.

Guttman: When we go into an organization and try to shift it, there are three things we’re working with: We’re trying to change mindset, we’re trying to change skill sets, and we’re changing the ways people work. The most challenging thing to do is to have people start to see beyond their little box and look at it as part of a bigger realm. But step one is changing their mindset— redefining for people what it means to be accountable in that organization. Peers need to hold peers accountable.

Speaking of accountability, how do you measure individual performance in a team setting?

Guttman: In the companies that do the best, there usually is a combination of the leader’s feedback and perspective from peers. One of the things a team needs to do is determine what percentage of compensation is going to be predicated on the leader’s input and how much is going to be team-based. It’s got to be both.

Reid: With my main team, we’re structured globally, with both regional GMs and functional heads. The first individual business objective for everyone on the team is the performance of Mars Drinks—growth, earnings, cash, and level of overhead. If the company is down, then everyone on my team is below expectations even before they start doing their own personal objectives.

Does everyone give everyone else feedback?

Reid: I get input from above, input from peers, and input from below, to get a balanced perspective. In some cases, people get twenty pages of feedback, which is very powerful because none of it is my opinion. It’s important to depersonalize feedback—you should see it as a gift, coming from a place of trust.

Top people aren’t used to hearing direct criticism of their performance. 

Allgaier: If leaders want to improve and learn and adapt, they learn to invite pointed feedback. As Grant said, you’re going to get better feedback from your direct reports and your team than you will from your boss. They see how you operate; you’re in touch with them all the time. So you have to invite it. What I try to do with people is to proactively give them something I don’t think I’m doing very well. So I’ll call someone in Germany and say, “Listen, I feel like I still don’t understand some fundamentals in your markets; I feel like I should get to your market more often; what do you think?” That gives him license to give me three or four things that he really wants to see from me. Otherwise, there’s natural resistance to giving the boss feedback. You have to go out of your way to enable it.

Now, people I have frequent contact with, in a team environment, are used to speaking up. Last week, the team was pushing me out of certain things, which was really encouraging, and we were talking about how I should spend my time and how they should spend their time. It was all on the table; it wasn’t personal. That’s how we get more done and get things done faster. It’s not personal to say, “Larry would maximize his effectiveness more if he did blank.”

Guttman: The team members need to put their concerns on the table, no matter where the discussion goes.

Allgaier: If it gets a little dirty, that’s good: “Hey, this function is not working; it’s broken; what do we do about it? The function head is in the room—let’s talk.” If it needs to go there, it goes there. That’s the sign of a team that’s on the journey toward high performance.

Guttman: If you were on a football team, and the play you just ran didn’t work, and you go back in the huddle, somebody would probably bring that up! If you care about winning, it’s crazy to not give input to the quarterback or the running back when they’re not making the play. So going back to what Larry said, the most serious thing that can happen is for someone to tell you that you’re not up to the job, based on the evidence they’re observing. And what would next be germane is for them to tell you what it would look like if you were. That would be a high-performing team, one on which people clearly understand that they are being paid to take that level of ownership for the whole team. That’s not something that people instinctively come into an organization thinking about. You have to feel that that team, that organization, is your own family business, and that the name of the company is your last name. You’ve got to play it like something’s at stake. And it’s not easy for people to do that, particularly the further down you go in an organization.

Reid: People usually think about responsibility for their direct reports and for themselves, and that’s where their accountability starts and ends. But in the team model, you take care of your peers, you take care of your boss, and you’re taking care of the enterprise. So it’s not a gotcha game—you’re trying to get to a space where you say, “My peers are giving me this feedback to make me more successful.”

Are there high performers who don’t do their best work within teams?

Allgaier: Definitely. You have people with highly technical expertise on a range of topics; they get pulled in, they help solve, but they’re not natural leaders, and maybe their ability to develop into a high-performing leader is less. It doesn’t decrease their value to the organization at all. In fact, the clarity around what they’re good at and what they’re not good at is a great thing to have, because then you don’t try to stretch them in the wrong areas. You play them to their strengths. You want to keep them as the go-to person for a specific issue. It’s liberating.

Reid: That’s just the word I was going to use. It sets them free. Right now they’re being dragged along to all these meetings, where the goal is fuzzy, so you’ve got to have everybody there because you don’t know what’s going to come up. If you’re a strong individual contributor who works best in your cubicle, that’s a very disenfranchised place to be.

Guttman: You find that all the time with people in R&D.

Allgaier: Some salespeople are great with customers and key accounts, but not necessarily managing a sales organization.

Guttman: There are a lot of people who are technically extremely strong, but they’d rather be lone wolves—sitting in their own offices, doing their own thing. Not everybody is meant to be part of a horizontal team.

Last: How often do teams have trouble reaching closure? Do they ever just pile up unresolved issues?

Reid: Without a high-performance plan, I would say that’s the norm. Issues are never truly resolved. Or you think they’re resolved, and then there’s a hand from the grave.

Allgaier: The more you get through, the more there is still to do. Businesses are complex. The list is always long. That’s the nature of business: fast-changing, with different issues—you’ve got to love that.

Guttman: You just can’t put it in cruise control, unfortunately.

Allgaier: You never reach autopilot. 

Guttman: Nirvana. 

Allgaier: It never happens that you get through six or seven issues and you have two left and there’s nothing else to do. The goal isn’t having nothing to do.

Guttman: It’s about adding shareholder value. You always want to go to a deeper level. It’s never a done deal.

Designed & developed by Greenfield/Belser Ltd.