Leadership influence: what the experienced already know

Authority gets you compliance. It reliably produces the thing you asked for, at roughly the standard you specified, on the date you named. That is not a small thing and plenty of organizations run on it.

It does not produce the work nobody asked for: the problem someone flags a month before it lands, the shortcut not taken on a Friday, the honest answer when that answer is unwelcome.

None of that can be required, because none of it is visible enough to check. The gap between what you can require and what you actually need is this page's subject.

What to take away

The information problem is the real one

Every layer between you and the work applies a filter, and the filter is not malicious. It is people reasonably deciding what is worth your attention, sanding off the parts that would prompt questions they cannot yet answer, and waiting until they have a solution before raising a problem.

The result is a steady, gentle drift toward good news, delivered with the confident texture of fact. It is close kin to groupthink, except nobody had to suppress anything. The filtering happened one reasonable judgment at a time.

By the time a problem reaches you unfiltered, it is usually no longer a problem you can help with. It is an incident.

You cannot mandate your way out of this. An instruction to raise problems early gets sincere agreement and no change, because incentives never concerned the instruction.

What changes behavior is what happens the first few times someone brings you something ugly and unresolved. If that produces visible stress, a rushed intervention, or a question about how it got this far, the lesson lands immediately and permanently: solve it quietly, tell them once it is handled.

Ask what is going badly rather than how things are going. Thank the person for the early warning before doing anything else with it.

Resist taking the problem over. The fastest way to stop receiving problems is to confiscate the first one you are given. Whoever tells you something inconvenient paid a cost, so make sure it was worth paying.

Everything you say is louder than you meant it

The most common expensive mistake is not a bad decision. It is thinking out loud.

The amplification is not something you can switch off by asking people to treat you normally. It is a property of the position, not of your manner, and the friendlier you are the less it helps, because friendliness removes the signal without removing the weight.

Three habits contain it. Label the register explicitly, and do it every time: say that you are thinking aloud, that this is idle curiosity, that you do not want anyone to do anything. Say it more often than feels necessary, because the default reading is always the stronger one.

Second, be careful where you speculate, since the same sentence in a one-to-one and in a room of thirty produces very different amounts of work. Third, ask afterwards. "Did anyone start doing anything because of what I said on Tuesday?" is a question that will surprise you at least once.

What you ask about is what gets worked on

Attention is the most powerful instrument here and the least deliberately used. Whatever you ask about repeatedly becomes what the organization optimizes, regardless of what the strategy document says and regardless of what you intended.

Ask about the number every week and the number gets managed, sometimes in ways you would not endorse. That is the everyday form of a measure turning into a target.

Ask only about what is going wrong and you build a culture that reports upward in the language of crisis. Ask about customers and people start talking to customers.

This works whether or not you are paying attention to it, which is the point. It is worth taking one deliberate look at what you have actually asked about over the last quarter, as opposed to what you would say your priorities are. The two lists are rarely the same, and where they differ, the questions are winning.

What the account is made of

Influence at this level behaves like a balance, and it is built from a shorter list than most people expect.

Predictability. People can only act on their own initiative if they can guess what you would say. A leader whose reactions are hard to forecast gets everything routed back for approval, which looks like being needed and is actually a tax on the whole organization. Consistency is worth more than being right slightly more often.

Accuracy about your own uncertainty. Saying "I think, and I am not sure" is what gives your confident statements any content. If everything is delivered at the same level of certainty, nobody can tell which of your views to plan against, so they hedge against all of them.

Not spending it on things that did not matter. Every override, every reversal, every request that turns out to have been unnecessary draws down the same account. The cost is not paid at the time; it is paid the next time you need something to be taken seriously immediately.

Doing the unglamorous part. Whatever you personally follow through on after the interesting phase is over tells people what is genuinely required, far more loudly than what you said at the start.

Spending it

Influence that is never spent is not being conserved, it is being wasted, and there is a recognizable kind of leader who accumulates credibility and never puts it anywhere.

Spend only in narrow cases: something breaks that only you can see across the whole organization. A standard erodes quietly, and each instance of erosion looks reasonable.

A decision is genuinely unpopular and genuinely correct, and you are willing to still defend it in six months. Protect someone who did right and took a hit; that is the single most watched thing a leader ever does.

What is not worth spending on: preferences about how work is presented, opinions in areas someone else owns and understands better, and the urge to be involved in the interesting problem rather than the one you are needed for.

Where this differs from having no authority at all

Moving work across a line you do not control is its own craft, covered in influence without authority. Read it even if you hold authority: most of what you need done sits outside your reporting line too.

Instructing people who do not report to you quickly shows how little formal position travels sideways.

The deeper difference is the failure mode. Without authority you are ignored, and you know it. With authority you are agreed with, and you do not know it. The second is harder to detect and considerably more dangerous, because it feels exactly like being effective.

Common questions

How do I tell whether people agree with me or are just agreeing?

Look for disagreement in low-stakes places. If nobody ever pushes back on small things, you are not getting agreement on the large ones either. You can also check whether the objections you hear are ever new to you, since a room that only ever surfaces objections you already raised yourself is a room that has stopped contributing.

Is being liked useful or not?

Useful, and not the goal. Being liked lowers the cost of every interaction. It does not survive the first decision that goes against someone, and a leader optimizing for it will delay exactly the decisions that need making. Aim for being predictable and fair, which outlasts it.

What if I inherited a team that already learned to hide problems?

Assume it takes months and that early attempts will be tested rather than believed. The turning point is usually one visible instance where somebody brought you something bad and was demonstrably better off for it. You do not get to choose when that instance arrives, only how you behave when it does.

How much should I explain a decision I have already made?

Give the reasoning, and be clear that it is a decision rather than a consultation. Reasoning is what lets people apply the decision to the twenty adjacent cases you did not think of. Presenting a settled decision as an open question is a separate mistake and a worse one.