Management and leadership, oriented

Management and leadership name tasks, not kinds of people. One framing says management aims at efficiency and order while leadership aims at change, and a leader's job usually includes both, weighted by the case.

The move into the job is the first hurdle. Research on nurses promoted into management shows the pattern: people are often chosen for expertise rather than leadership ability, then meet changes in the work, relationships and responsibilities at once, and many report role ambiguity and self-doubt. The Peter principle is the pessimistic version: people rise until they reach a job they cannot do well. One three-wave study read its results as possible evidence, which falls well short of proof.

Your influence has several sources. French and Raven named five bases of power: legitimate, reward and coercive power come with the post, while expert power (knowledge) and referent power (an emotional bond with followers) come from you. Style is how you use them, and the labels are less settled than they sound. Situational leadership says to match your style to the follower's readiness, but its empirical support is thin. Transformational leadership inspires and develops followers, though critics say it overlaps heavily with other style labels. Laissez-faire is the absence of leadership, which differs from autonomy, where someone has real choice over how to reach a goal.

The trap is letting urgent work decide your week. Say Priya, promoted over four former peers, spends every day covering staffing gaps and solving today's problems. In one qualitative study of nurse managers, that pressure kept daily operations going while supervision and staff development were missed or postponed. Avoiding hard calls costs too. In one paper combining an experiment with two multi-wave surveys, employees who felt a manager had failed to act on an aversive event rated that manager less trustworthy.

Rule: schedule the parts of the job nobody else will do before the urgent work fills the week, and earn the influence your title cannot supply.

Intrinsic motivation is doing work because the work itself rewards you. Extrinsic motivation is working for a consequence outside it, though a reason you have made your own behaves much like interest. Self-determination theory names three needs behind good motivation: autonomy (acting by choice), competence (experiencing mastery) and relatedness (supportive connection). Meeting them is linked to well-being, and frustrating them to pressured motivation and ill-being. The theory's advice to you is to protect those needs rather than try to install interest.

Goals add direction. Federal guidance summarizing Locke and Latham reports that people given specific, hard goals outperformed people told to do their best, and it stresses goals that are difficult but possible. Expectancy theory adds three beliefs a person must hold: effort leads to performance, performance leads to reward, and the reward is worth having. Equity theory adds comparison, since people weigh what they put in and get out against their peers and may cut effort when the ratio seems unfair.

Two famous models need caveats. Maslow's hierarchy says lower needs come first, but later research suggests higher needs can motivate before lower ones are met, and that the order varies across cultures. Herzberg separated hygiene factors such as pay from motivators that build satisfaction. It remains contested, and one Saudi public-sector survey found the reverse of its prediction. Rewards have side effects too: poorly designed incentives can crowd out motivation, especially in prosocial or creative work, though how far rewards undermine interest is itself disputed.

The trap is reaching for a bonus first. Say Tom's reps miss a sales target, so he offers a large bonus, yet they believe the target is out of reach. By expectancy theory's logic, the first belief is broken and the money does little. Check the load as well: theories trace burnout to high demands with too few resources and too little recovery.

Rule: before adding a reward, test the three beliefs and the three needs, then fix whichever one is broken.

Delegating frees you for more complex work and develops people, but it does not move the outcome off your desk. Nursing's definition makes this plain: responsibility for doing the task transfers, while accountability for the result stays with the person delegating. So say what moves, since handing over a task differs from handing over the decision about it. Autonomy should follow what the person has shown they can handle, and more is not always welcome: in one study of Generation Z knowledge workers, empowering leadership raised perceived support and also emotional exhaustion.

Delegate the result, not a script. Federal performance guidance separates activities, which are what people do, from accomplishments, the products or services that result. Micromanagement is supervision detailed enough to shrink someone's judgment, though one review notes most studies examined related ideas, not micromanagement itself. Abdication hands off the work and withholds the support it needs. Between them sit regular check-ins, which federal guidance puts at monthly or at least quarterly. A RACI matrix gives each task responsible, accountable, consulted and informed roles, with only one person answering for completion.

Goals need the same clarity. A SMART goal is specific, measurable, achievable, relevant and time-bound. Outputs count what was produced, while outcomes measure what changed, which outside factors can also affect. Key performance indicators (KPIs) are measures of success you track, and objectives and key results (OKRs) pair a stated goal with results that would show progress. When several goals compete, rank them openly.

The trap is trusting one number. Say Ana's support team is scored on tickets closed per week. Tickets get closed fast and reopened under new numbers, and the figure improves while customers wait. Goodhart's law names that pattern: once a number becomes a target, it stops telling you what it used to. One remedy is several imperfect measures that pull in different directions, which makes gaming any one of them harder.

Rule: when you hand work over, state the result, who decides what, and the next check-in, and pair every target with a measure that would expose gaming it.

Feedback earns its keep by ending in a next step. In one study summarized in federal guidance, feedback alone during appraisals did not improve performance, while feedback followed by goal setting did. The usable kind is observed, prompt and specific. The situation-behavior-impact format keeps you there: when it happened, what the person did, what it caused.

"You are abrasive" describes a person and invites a defense. "In Tuesday's client call you interrupted Lena three times, and she stopped raising objections" describes something the person can change. Praise works the same way, because naming what worked tells someone what to repeat.

Popular recipes for delivery are less settled than they sound. The feedback sandwich wraps criticism in praise. One review of feedback models calls it one-way and educator-centered, yet one randomized comparison with 69 dental students found no overall difference from an ask-tell-ask approach. Another idea pairs directness with personal care, and the authors of one interview study of medical residents suggest that feedback is shaped by shared goals and relationship quality, not information alone.

Coaching and directing are both tools. Non-directive coaching helps someone find their own answer and is widely treated as the standard, though the question is debated, and some people get more from direction. The GROW model moves through goal, reality, options and will. In one study of new nurses, adding it raised performance scores, while turnover intention and commitment did not differ.

The trap is saving it all for the annual review. Federal guidance asks for check-ins monthly or at least quarterly, development talks kept apart from reviews, and records of goals and feedback so that ratings bring no surprises. When performance stays poor, it says to look first for a root cause such as a skill gap, motivation or a poor fit, then offer support. Federal agencies then state the expectations and the problem, give a chance to improve, and take action. That is one employer's process, described for study; other employers follow their own rules.

Rule: describe what you saw and what it caused, agree on the next step, and never let a rating be the first time someone hears it.

Start with a job analysis: the tasks, the competencies they need, and which of those must be present on day one. A structured interview then has three parts: the same questions in the same order, a common rating scale, and agreement in advance on acceptable answers. Unstructured interviews are attractive because they are loose and conversational, but two interviewers often rate one candidate differently, and they predict job performance only low to moderately well.

Behavioral questions ask what someone actually did in a past situation, and situational questions pose a realistic dilemma. Work samples show the work directly but cost more, so they pay most when a competency is critical at entry. Validated trait scales can predict performance, while type instruments such as the Myers-Briggs Type Indicator are contested: critics say they force continuous traits into categories, and in one older study only 47% kept the same type over five weeks, though defenders cite other studies with acceptable reliability.

The trap is trusting a good conversation. Say Sam and Rae warm to a polished candidate within minutes. She shares Sam's hometown (similar-to-me bias), and her confident speaking earns credit for problem solving she never showed (the halo effect). If each rates alone from notes before they talk, then explores the gaps, the disagreement becomes evidence.

US rules shape the questions. Asking about disability before an offer is prohibited, and questions that reveal race, religion, national origin, age or sex can serve as evidence of intent to discriminate. This is general information for study, not legal advice.

The same biases follow people into reviews. Recency bias lets the last weeks outweigh the year, and central tendency puts everyone mid-scale. Calibration sessions compare ratings against shared standards. Forced ranking caps how many people may get each rating, which pushes managers to draw distinctions. A 2006 federal review argued that performance rarely follows a bell curve and that ranked colleagues may cooperate less, but that review is not current federal policy: a July 2026 federal manual has agencies work within government-wide rating caps.

Rule: agree on the scoring standard before any candidate arrives, rate before you discuss, and keep evidence all year.

Psychological safety is a shared belief that you can ask a question, admit a mistake or challenge the plan without paying for it. Its effect on performance is debated: in one study of 104 South Korean sales and service teams, it did not directly affect team effectiveness, only through learning behavior and efficacy. It also differs from comfort. Edmondson's grid crosses safety with accountability, so high on both is the learning zone, and safety without standards is the comfort zone. In one interview study, research-team leaders described building safety by modeling humility, seeking input from junior members and asking what led to a missed deadline before assigning blame.

The trap is reading silence as agreement. Say Jo asks six people whether a launch date is realistic, hears nothing, and learns in week three that two doubted it from the start. Raising a concern carries a social cost, especially upward, so an absence of bad news is weak evidence that nothing is wrong.

Cohesion has its own risk. Janis described groupthink as the drive for unanimity overriding a realistic look at alternatives, so a friendly team is not automatically immune. Structural counters, borrowed from intelligence analysis, include assigning someone to build the strongest case against the prevailing view, bringing in an outsider, and collecting ideas before discussion converges.

Effort and ownership fail in predictable ways too. Social loafing is strongest when individual contributions cannot be identified, so give each person a distinct, visible piece. Tuckman's stages, from forming through storming to performing, give a team shared words rather than a timetable. Storming over roles is expected, so write down norms and roles early.

Differences and distance add more. In one study of hospital teams, more diverse teams reported more task conflict, and trust among members weakened that link. Remote teams need deliberate rules for when to write and when to call, and current federal guidance holds remote and in-person staff in the same job to the same standards.

Rule: ask directly for doubts, make each person's work visible, and write down norms before friction tests them.

Before a discussion starts, settle who decides and how: the leader alone, the leader after consulting, a vote, or consensus. Consensus needs its own bar, since "everyone can live with it" differs from "everyone actively supports it." No mode wins everywhere. In one paper's scenario experiment and field survey, firefighters trusted a leader more for deciding alone during urgent action and for sharing decisions in the calmer phases between, which suggests the right mode can change with conditions. Disagree and commit is the matching habit: argue fully while the question is open, then carry out the result.

Spend deliberation where errors last. Where a choice can be undone cheaply, a quick decision and real feedback can teach more than a long debate. A choice you cannot walk back justifies more checking, such as assigning someone to argue the best case against the prevailing view. A premortem has the team assume the project already failed and write down why, separately, before committing. Collecting views first keeps the opening speaker from shaping the rest.

The trap is the fake consultation. Say Dee has already chosen a vendor, then asks the team for input. Consulting can strengthen people's sense of control, but only if their input is genuinely heard, so say plainly which part of the decision is still open.

Conflict comes in three kinds: task conflict concerns the work, process conflict concerns who does what and when, and relationship conflict is personal friction. Meta-analyses tie both task and relationship conflict to lower team effectiveness, relationship conflict more strongly, yet task conflict can help under some conditions and can harden into personal conflict over time. When two of your reports clash, ask what each needs rather than what each demands. You also hold authority a true mediator lacks, so when they cannot agree or the issue is yours to control, decide and give your reasons.

For meetings, name the job in advance, phrase agenda items as questions, and end by confirming what was decided, who owns each follow-up and by when.

Rule: name the decider, the rule and how reversible the choice is, then leave every meeting with owners and dates.

Large changes often stumble on people and culture rather than the plan. A 2003 federal report drawing on a forum about major mergers noted that many fail because the combining cultures were not fully understood, and that productivity tends to decline right after a merger. So weigh the expected gain from a reorganization against the disruption it brings.

The same report stresses a clear, consistent reason and a picture of what will be better. Top leaders supply visible backing, while a dedicated team runs the daily work. Communication should be early, frequent and two-way, and one forum participant put the time needed at twice the first estimate. Piloting helps: in one federal example, a new executive compensation plan ran for a year with top executives before reaching all senior executives.

The trap is treating every objection as attitude. Say Mara's clinic moves to a new scheduling system and the nurses push back. A review of nursing studies traced resistance partly to thin support, poor training, the speed of change and weak communication, so the complaint may be a bug report. A small group may resist any change, and continuous change can wear down capacity on its own. Check incentives as well, because ratings that still reward the old priorities argue against the new direction.

Structure is a trade-off, as studies of hospital governance show. Functional lines protect professional standards but must integrate work across them. Units built around a service or region own an outcome and respond locally but can isolate specialists. A matrix tries for both through two reporting lines, which can blur accountability.

Influence often runs sideways and up. In studies of school leaders, hard tactics such as pressure won short-term compliance but often cost morale. One theoretical framework argues that managers depend on staff for knowledge and results, which gives staff leverage. In one study of supervisor and employee pairs, political skill built stronger relationships when task performance was high.

Rule: state why, align rewards with the new direction, and check each objection as a fact before you call it resistance.