You don't lead everyone the same way (and why the little grid is not a law)

Two "you handle it" that turned out very differently

One Monday, the same boss says the same sentence —"you handle it, I trust your judgment"— to two people. The first has been at the company for two weeks. The second, for ten years. For the veteran, that sentence is a vote of confidence: she grabs the problem and solves it better than he would have. For the newcomer, the same sentence is a push into the void: he does not know where to start, he guesses, he gets it wrong, and he hides the mistake for fear of asking. Same sentence, same boss, same minute. Opposite results.

Now flip the boss. One who "is on top of everything": gives step-by-step instructions, reviews every detail, asks for reports. With the newcomer, it works: it gives him the structure he needs. With the veteran, the same thing is a daily humiliation — she feels watched, stops proposing, and starts browsing job postings from other companies on Sunday nights.

The problem, in both cases, is not the style. It is using a single style for everyone. And since every boss has a comfortable style —one he falls back on by default when he is not thinking— he ends up leading well the part of the team that style fits, and badly everyone else: half over-managed and suffocated, half under-managed and abandoned, both performing below what they could. This article is about how to get out of that default: what the most famous model says about adapting your style, what others that almost no one cites add to it, and why —being honest, as we should— that four-box grid taught everywhere is more useful as a mirror than as a law.

One style is not enough

The idea was organized by two authors, Paul Hersey and Ken Blanchard, first in 1969 —under the forgettable name "life cycle theory of leadership"— and later, rebranded "situational leadership," in the 1977 manual that the training world adopted as gospel. Their claim is deceptively simple and genuinely useful: there is no single best style of leading. The right style depends on where the person is —on this task— according to two things: how much they can do it (competence: knowledge, skill, experience) and how much they want to or dare to do it (commitment: motivation and confidence).

On the boss's side, behavior also has two independent dials: how much he directs (instructs, structures, sets the how and follows closely) and how much he supports (listens, encourages, involves, backs up). Crossing the two dials produces four styles, and the model draws a bell curve that runs through them as the person matures.

It is worth saying up front something the model does not always make clear, and that is its firmest part: those two axes are not a training-manual whim. They come from the Ohio State University leadership studies of the 1950s, which isolated two great dimensions of the boss's behavior —initiating structure (the task: organize, instruct, set the course) and consideration (the relationship: listen, support, care)—. And they are among the best-supported findings leadership research has: the meta-analysis by Judge, Piccolo and Ilies (2004), which pooled more than three hundred correlations from decades of studies, found that both are appreciably associated with outcomes (consideration, with a correlation close to 0.48; initiating structure, 0.29). Put another way: the two dials of situational leadership are well chosen. What remains to be seen —and it is what decides everything— is whether the recipe for combining them survives the same test.

The four styles, and whom each one serves

Directing (S1) — high direction, low support. For those who cannot yet and/or do not dare: the newcomer, or anyone —however expert in something else— facing something genuinely new. Clear instructions, defining the task and the steps, close follow-up, one-way communication. It is not micromanagement for its own sake: it is scaffolding for someone who needs it. A crisis or an emergency, with anyone, also calls for this style.

Coaching (S2) — high direction and high support. For the motivated learner: still needs to be shown the how, but is eager to grow. Here the boss explains the why of his decisions, opens the conversation to two ways, "sells" the task —convinces the person they can do it— and gives feedback that builds. It is the eager fresh graduate, or the experienced person who changed areas and, though they know a lot, suddenly feels clumsy and low on confidence.

Supporting (S3) — low direction, high support. The most counterintuitive. For those who can but do not dare or do not want to: they have the competence, but lost the confidence —made a mistake that left them shaken, burned out from overload— or the motivation —stuck, with no horizon—. They need little instruction and a lot of backing: to be listened to, to be handed decisions, to be reminded how well they did other things. Treating this demotivation with more control is the classic mistake, and it is pouring fuel on the fire.

Delegating (S4) — low on both. For the autonomous expert: set the objective, agree on the checkpoints, and step aside. It sounds easy and it almost never is —it is especially hard for the insecure boss supervising someone more experienced than he is—. Delegating well is not abandoning or suffocating: it is letting go with clarity about the what and the by-when, and holding the distance. A detail that gets forgotten: even the most autonomous person needs recognition for what they achieve.

Table — Where each one is, and what it calls for. Adapted from Hersey and Blanchard (1977) and Blanchard (SLII).

One person's journey (and why enthusiasm dips before it rises)

The four styles are not four types of person: they are four moments of a single journey. Ken Blanchard, who kept developing the model on his own in the version he called SLII, made it explicit with four development levels that almost anyone goes through when learning something new — and with a curve every boss should keep in mind.

You start as an enthusiastic beginner: you know little, but you are brimming with drive — it is day one, everything seems possible. It calls for Directing: show me the how. Soon comes the most dangerous and worst-read moment: the disillusioned learner. Reality turned out harder than it looked, the launch momentum deflated, and the task is still not mastered. It is the point where most people get frustrated, perform poorly and —if no one holds them— fade out or leave. It calls for Coaching: direction for the missing competence and, above all, support for the motivation that collapsed. Whoever survives that valley becomes capable but cautious: they know how to do it, but do not quite dare, they doubt their judgment. It calls for Supporting: less instruction, more confidence. And only then do they reach self-reliant: they can and they want to, and what they need is to be left alone — Delegating.

That valley —the disillusioned learner's— is a cousin of the pit of the J-curve that came up when we talked about change: the stretch where things get worse before they get better, and where the temptation to give up is greatest. The boss who does not expect it reads it wrong —"they lost motivation, this wasn't for them"— and lets go exactly when he should be holding on. Knowing the valley exists, and that it is a stage and not a verdict, is the difference between losing someone and forming them.

It is not a one-way staircase

Two nuances the training slide flattens, and that are half the value of the model.

It is by task, not by person. No one "is an S4." The same veteran is Delegating on the process she has mastered for years and Directing on the new system installed last month — the same day, in the same meeting. Reading a person with a single label ("she's senior, leave her alone") is the shortcut that produces half the mismatches: she is delegated what she does not yet master, or directed on what she could already fly solo.

You can go backwards. Competence and commitment are not a staircase that only goes up. The expert who burned out, the star who made a costly mistake and lost her nerve, the middle manager demoralized by a restructuring: all dropped a step and, for a while, need Supporting or even Directing again. Assuming that whoever once reached self-reliance will be so forever, and in everything, is one of the most expensive and least visible mistakes in management. That is where the model earns its keep: not as a recipe, but as a diagnosis that forces the right question —"where is this person, on this task, today?"— instead of the comfortable one —"what is my style?".

The two ways to err: suffocation and abandonment

If each moment has its style, there is one way to get it right and two to get it wrong — and it is worth naming them, because they feel different and do different damage.

Over-leading (suffocation). Giving more direction or more control than the person needs: directing someone who could already decide, closely supervising someone who deserves to be left alone. From the side of the one who suffers it, it feels like distrust. The capable person stops proposing, becomes passive —"whatever, he'll review it all anyway"— and often leaves: not for money, but from suffocation. It is the classic mistake of the founder-boss who "is into everything."

Under-leading (abandonment). The opposite: giving less direction or less support than is needed. Delegating to someone who cannot yet, letting go of the disillusioned learner in the middle of the valley. It feels like orphanhood. The person sinks, hides the problems out of fear, and by the time the mistake blows up it is already expensive. It is the classic mistake of the boss who confuses delegating with disappearing.

What is revealing about this matrix is that most bosses err almost always to the same side — the side of their comfortable style. The controller suffocates almost everyone; the absent one abandons almost everyone. That is why the first step is not to master the four styles, but to know your own default and which error it pushes you toward.

It is not alone: cousins with and without support

It is worth pulling situational leadership out of the solitude the training market puts it in, because it is neither the only "it depends" theory nor the first, and seeing it beside its cousins sharpens it.

Tannenbaum and Schmidt's continuum (1958). Almost a decade earlier, in Harvard Business Review, two authors drew the same intuition as a dial rather than a grid: a line running from "the boss decides and announces" at one end to "the team decides within limits" at the other, with selling, consulting and co-deciding in between. The underlying question —how much authority to keep and how much to cede— is identical; the drawing, more honest, because it admits it is a continuum and not four boxes.

Goleman's six styles (2000). The best known in companies. In Harvard Business Review, on a sample of more than 3,000 executives, Daniel Goleman described six styles —coercive, authoritative, affiliative, democratic, pacesetting and coaching— and found that the most effective leaders do not have one favorite: they switch styles "the way a golfer picks a club," according to what the hole demands. Goleman adds two things situational leadership underestimates: that some styles —the coercive, the pacesetting— poison the climate if overused, and that the ability to switch styles rests on emotional intelligence —self-awareness and empathy—, not on a technique you memorize. With a caveat that science itself forces: emotional intelligence is measured and sold as a panacea, but its relationship with performance is more modest and more conditional than executive self-help promises. The meta-analysis by Joseph and Newman (2010) found it useful above all in emotionally demanding jobs, and thin or null in the rest. The good boss who switches styles exists; that he does it because he has "a lot of emotional intelligence" is, in part, marketing.

House's path-goal theory (1971). One of the most studied "it depends" theories: the leader adjusts his behavior —directive, supportive, participative or achievement-oriented— according to the task and the person, to clear each one's path toward the objective. House reformulated it in 1996 in light of the criticisms, and the meta-analysis by Wofford and Liska (1993), over some 120 studies, left an uncomfortable moral for the whole field: much of the research that tested these theories was methodologically weak, and the support was partial. It is the recurring pattern — the intuition, powerful; the measurement, elusive.

Leader–member exchange (LMX): the best-tested cousin. And here appears the one that really has backing. Leader–member exchange theory (Graen and Uhl-Bien, 1995) starts from a fact every boss knows and few admit: you do not have the same relationship with every person on your team — you build different bonds, with different degrees of trust and autonomy, with each. Far from being a flaw, that differentiation is the empirically solid version of situational leadership's intuition: the meta-analysis by Gerstner and Day (1997) found that the quality of those differentiated bonds is associated with performance, satisfaction, commitment and even with who stays and who leaves. With a detail that foreshadows the grid's problem: the boss and the follower agree only moderately on how they perceive that relationship. Even the best-tested model trips on the same stone — the boss's view is not the truth.

The thread that runs through sixty years and rival maps is a single one: there is no single best style. They disagree on the cartography —and not all have the same empirical support—, but they agree on the terrain.

The fine print: why the grid is not a law

Now the uncomfortable part, which this series does not skip. It is worth separating two things that usually go together and are not the same. The materials of situational leadership —the two dimensions, the idea of differentiating— have, as we saw, solid backing. What almost does not is the specific recipe: that a given level of "maturity" corresponds to a given style, and that combining them with that formula improves performance. When researchers tested the recipe —not the intuition, the recipe— the results were, to put it generously, thin.

Robert Vecchio tested it in 1987 with 303 teachers from fourteen schools: the prescription held above all for new employees —the Directing corner— and grew flimsy for the rest. Blank, Weitzel and Green (1990) went to the heart of the matter and put the central assumption to the test —that the follower's "maturity" moderates the relationship between the boss's behavior and his effectiveness—; the data, plainly, did not support it. Norris and Vecchio (1992) replicated with mixed results; Fernández and Vecchio (1997) did not find the promised pattern when looking at different jobs; Thompson and Vecchio (2009) tested three different versions of the model and found only partial support. As early as 1983 and 1997, Claude Graeff had explained why it was to be expected: the internal logic is inconsistent and the concept of "maturity" is too vague to be measured seriously. And the most telling fact is the one that is missing: more than half a century on, situational leadership still has no meta-analysis to support it — a rarity for a model so widely taught.

There is an exception that half-rescues it, and through a revealing door. Thompson and Glasø (2015, 2018), on 80 supervisors and 357 followers, showed that the model predicts better performance only when the boss and the follower agree on the diagnosis of "maturity." When they do not agree —and very often they do not—, the advantage disappears. And that relocates the problem exactly where it was hidden.

And there is the deepest crack, the one the very document that inspired this article has the honesty to point out: the whole model depends on the boss judging the other's "maturity" well — and that judgment is precisely where bias lives. The halo effect (they came with a brilliant recommendation, so we assume they are competent), the mirror bias (they resemble me, they look confident, so I suppose them more autonomous than they are), the stereotype about who "looks" senior by their age, gender or background: all disguise themselves as a reading of maturity. A model that tells you to treat each person differently according to your judgment about them is worth exactly what your judgment is worth — and every honest boss knows that his judgment about people is the least reliable instrument he has.

Wisdom survives the criticism. The false precision of four tidy boxes does not.

The conclusion is not to throw out the grid, but to use it for what it is good for: a mirror and a vocabulary, not an algorithm. Its lasting gift is the question it forces you to ask —am I treating this person as they are, or as my default makes me?— and the humility it should force right after —and how sure am I that I am reading them right?—.

How to read readiness without fooling yourself

If the model's weak point is that everything depends on the boss's judgment, then the real work is not memorizing the four styles: it is learning to read better —and to distrust your own reading—. Four disciplines help.

Separate "cannot" from "will not." Competence and commitment are different things and call for opposite responses: the one who cannot is taught; teaching the one who will not more only insults them. Confusing the two is the most common and most expensive diagnostic error, because training the demotivated or haranguing the incompetent are the two ways of spending energy without moving anything.

Co-diagnose, do not guess. The evidence is emphatic on one point: the model works when the boss and the person agree on where they stand. And the cheapest way to reach that agreement is astonishingly simple and astonishingly rare — asking. "On this, are you comfortable for me to let you run with it, or would you rather we go through it together a couple more times?" turns a unilateral judgment —where bias lives— into an agreement. The person does not always self-assess well; but the conversation almost always brings the two readings closer.

Look at behaviors, not impressions. "Looks confident," "came recommended," "he's one of mine" are not data: they are the doors through which bias enters. Data are other things: have they solved something similar before, and how? What do they do when a problem appears that was not in the script? Readiness is read in past behavior on comparable tasks, not in the first impression or in the resemblance to oneself.

Review, because people move. No reading is permanent: the learner matures, the autonomous one burns out, the task changes. Every management diagnosis has an expiry date — and the boss who does not review it ends up leading the person his colleague was a year ago.

The difficult conversation

"I am the way I am; you can't change your style." Fred Fiedler, another classic, would even grant you that style is hard to move — and for that reason proposed the reverse: place each boss in the situation that fits him. But a boss of a single style only leads one type of person well. Stretching your repertoire is not betraying yourself; it is widening how many people you can lead without breaking them.

"Adapting your style is faking it, or manipulating." It is the opposite. Treating people in different places the same is not fair: it is convenient. Giving scaffolding to the one starting out and air to the one already flying is not falseness — it is the minimum asked of anyone who leads people and not interchangeable parts.

"This is soft psychology, not management." The most expensive management mistakes are of ill-fitting style: the suffocated senior who resigns, the abandoned rookie who fails and leaves, the capable but demotivated one no one listened to in time. It is paid in salaries, in turnover, in recruiting and in lost months. There is nothing soft about that bill.

You don't lead everyone the same way because not everyone is in the same place — and the same person, tomorrow, may be in another. The four-style grid is not a law of physics nor a machine that decides for you; it is a mirror that forces a question autopilot never asks: this person, on this task, today, do they need more hand or more air? Answering it well, often, and with humility about what you think you see, is almost the whole craft of leading.