The First 90 Days

What Actually Happens When You Start a New Leadership Role

The first 90 days of stepping into a new leadership position are not what most people expect. I spent over a decade watching people bomb through transitions, both by hiring them and by making the mistakes myself. The concept has been written about extensively, most notably in the book by Michael Watkins, but the practical application is where things fall apart for most leaders. The core idea is straightforward: you have roughly three months to establish credibility, learn the landscape, and set your direction before the honeymoon period evaporates. After that, people start forming hard opinions about you, and correcting course becomes significantly harder.

The First 90 Days as a Framework

Watkins' framework breaks this period into three phases: the onset phase (days 1-30), the learning phase (days 31-60), and the momentum phase (days 61-90). Most guides present this as a clean progression. In reality, it rarely works that way. You will be pulled between learning and acting simultaneously, often on the same day, sometimes within the same hour. The real mistake people make is treating the framework as a checklist rather than a diagnostic tool. You cannot simply "complete" each phase. You assess, adapt, and re-assess constantly. The structure helps you understand where you might be going wrong when things feel off-track.

How the Transition Actually Plays Out

Here is what I have seen repeatedly. People arrive at a new role and immediately start solving problems they perceive. They spend their first weeks identifying inefficiencies and pushing for changes. This approach works until it does not, and when it fails, it tends to fail catastrophically. The more effective approach is to separate listening from acting deliberately. In your first two weeks, your primary job is to understand the informal organization. The org chart tells you nothing about how decisions actually get made or who influences whom. I once joined a company where the formal reporting structure showed one team leader as the key decision maker. Within three weeks of sitting down with people across three levels and two departments, I discovered the actual influence map was completely different. The person on paper was a figurehead. The real gatekeeper was someone two levels down with no formal authority over me. If I had gone in pushing initiatives through the formal channel, I would have hit a wall I could not have explained to anyone, including myself. Instead, I spent an afternoon having coffee with the actual gatekeeper before proposing anything. That conversation alone saved me approximately six weeks of political friction and two failed pilot programs.

Common Pitfalls and What Most People Miss

The biggest oversight is not recognizing that your predecessor casts a long shadow. People will compare everything you do to whoever came before you, often without consciously realizing they are doing it. This happens even if your predecessor was terrible. You inherit an implicit baseline, and you need to understand whether that baseline was built on actual performance or just institutional memory. Another counter-intuitive truth: trying to appear confident and knowledgable from day one actively works against you. I have seen leaders damage their credibility faster by claiming expertise they do not yet have than by openly admitting gaps and asking questions. There is a narrow window in the first month where asking "Can you walk me through how this decision used to get made?" reads as intelligent curiosity rather than incompetence. After about 45 days, that same question starts sounding like you have been sleeping on the job. The second common pitfall is underestimating the energy drain. A typical week in the onset phase involves learning context, managing upward expectations, building lateral relationships, and still delivering on existing commitments. The cognitive load is genuinely exhausting. I used to underestimate this and plan aggressive first-week agendas. Now I schedule nothing after 5 PM for the first three weeks and actually stick to it. Your brain needs the downtime to process what you just absorbed. Skipping this recovery time leads to thin decisions by day two or three.

A Practical Onboarding Sequence That Actually Works

Week one: Do not propose anything. Listen. Schedule 15-minute chats with everyone who reports to you if this is a management role, and seek out peers in adjacent teams. Ask the same three questions each time: what is working well, what is broken, and what should I know that is not in any document. Take notes. Most people will give you polite, surface-level answers initially. Pay attention to what they refuse to mention. Week two: Begin connecting the dots. Patterns will emerge from the conversations. Someone mentioned budget constraints twice. Another person described the same project failing in three different ways. These are signal hits. Write them down. Do not act on them yet. Validate them through additional conversations with different people. Week three: Start testing assumptions. Pick one or two early hypotheses and design small experiments to confirm or refute them. I had a case where multiple people hinted that a particular software migration was dragging project timelines. Rather than accepting this at face value, I asked for the last three project schedules and compared them against the pre-migration baseline. The data showed the migration had no measurable impact on timeline. The perception problem actually traced back to a staffing reduction that happened two months later. Acting on the initial assumption would have meant I was diagnosing the wrong problem and wasting resources on a software solution for a people problem.

By week four, you should have a preliminary picture of the terrain. Begin shaping a short-term plan. This does not need to be comprehensive. It needs to be real enough that people start to understand what direction you are heading and can adjust their own work accordingly. Ambiguity beyond six to eight weeks starts to erode momentum.

When This Approach Fails Completely

There are scenarios where a careful 90-day transition strategy is not appropriate. If the organization is in active crisis, if the previous leader was fired for cause and there is a fire to put out, or if the board has given you a specific turnaround mandate, the learning phase compresses dramatically. You may have two weeks, not two months, to make impact. Similarly, in highly technical roles where deep domain expertise is the primary credential expected, spending excessive time listening without demonstrating competence can backfire. Engineers and specialists often judge you by what you produce, not by how well you understand their politics. I learned this the hard way when I moved into a head of engineering role at a startup. My previous approach of extensive listening and relationship-building was read as indecisiveness by a team that needed technical direction immediately. I spent three weeks too long in assessment mode before shipping something tangible. The fix was to pair continued listening with visible technical contributions within the first two weeks.

The Hard Truth About Measurement

You will never know if you executed your first 90 days well based on objective metrics alone. Most organizations do not track transition success systematically. The feedback you get is often distorted by people managing their own expectations of you. This is why maintaining external references during the period matters. Former colleagues, mentors, or board members who are not embedded in the new organization's politics can give you more honest reads on whether you are actually making progress or just appearing to. The framework gives you a structure for thinking through a difficult transition. It does not guarantee success. What it does provide is a vocabulary and a diagnostic lens for when things feel wrong but you cannot pinpoint why. That alone is worth more than most leadership development programs I have sat through.