The Case Against Doing One Thing at a Time

A useful fact about how skill develops under uncertainty: people who study three different problem types by mixing them , A, B, C, A, C, B, A, C, B, C, A, B , outperform people who study the same material in blocks , AAAA, BBBB, CCCC. The effect is counterintuitive and robust. It is also contrary to how most operators train themselves and their teams, and it is contrary to how most executive development programs are built.

The research term is interleaving. It sits in the broader family of desirable difficulties, the literature on how a temporarily harder practice produces more durable learning. Robert Bjork's group at UCLA has been running these studies for decades. The finding replicates across domains , math problems, motor skills, medical diagnosis, musical performance. The blocked-practice students feel more fluent during the study session. The interleaved students feel slower, sometimes frustrated. A week later, on a transfer test, the interleaved students outperform by substantial margins. The subjective feeling of fluency is a poor guide to actual skill.

Why this matters for operators

An executive's decisions are interleaved by structure. A morning might include a capital allocation call, a personnel conflict, a product prioritization, a partnership negotiation, a health decision, a family decision. Six decision types in four hours, each with different relevant variables, different appropriate strategies, different emotional loads.

The skill that matters at this tempo is not depth in any one decision type. It is strategy discernment , the ability to look at a decision and, in the first thirty seconds, identify what kind of decision it is and which of the tools in your set is appropriate. This is the conditional knowledge piece of metacognition: knowing when and why to deploy a specific strategy. It is the skill most under-trained in most professional development, because most professional development is blocked.

A typical MBA elective runs for fourteen weeks on a single topic. A typical leadership retreat runs for three days on a single theme. A typical coaching engagement works the same decision domain repeatedly for months. All blocked. All producing fluency in the domain under study. None producing the strategy-discernment skill that an executive actually needs on Tuesday morning.

What blocked practice actually teaches

Blocked practice teaches you to deploy a specific strategy. If you sit for six hours with a single framework , say, the pre-mortem , and work through twenty scenarios, by the end of the day you will be fluent in pre-mortems. You will use them well. You will feel skilled.

What blocked practice does not teach is when to use the pre-mortem and when to use something else. Because every scenario in your six hours was a pre-mortem scenario. You were never given a decision and asked, first, is this a pre-mortem decision, or is it a second-order-thinking decision, or is it a sleep-on-it decision? The choice was pre-made for you by the curriculum.

This is why operators who have read extensively about decision-making often describe a specific frustration: they know the frameworks; they cannot, under pressure, feel which framework fits. The tools are in the toolbox; the hand does not know which handle to reach for. The missing skill is the reach.

The specific practice

In the Academy, Day 2 afternoon includes a session we call the interleaved decision set. Participants work through five different decision types in rapid succession , fifteen minutes each, no warm-up, no theme. Capital allocation, then personnel, then partnership, then operational, then strategic. The facilitator interrupts each participant once during the session with a specific prompt: what are you actually weighing right now?

Most participants feel tired by minute forty. The tiredness is the point. They are doing the specific, expensive work of strategy-selection that their ordinary days do not force them to practice deliberately. By the end of the ninety minutes, something has shifted in how they approach the sixth and seventh decisions of the day , they arrive at the strategy selection faster, with more awareness of why they chose that strategy rather than another.

Outside the Academy, the practice is available to anyone, with minimal equipment. Once a week, set aside ninety minutes. Bring three to five decisions you are currently holding, each of a different type. Work through them in alternating order. Write down, before each decision, which strategy you will use and why. After each, note what you actually did and whether the initial strategy choice survived.

Thirty sessions in, the strategy-discernment skill is measurably different. You will feel it in your normal Tuesday mornings: the first thirty seconds of a decision carry information they did not carry before. You are not deciding faster; you are orienting faster. The deciding itself happens on a clearer ground.

Why the practice feels bad while it works

A specific frustration is structural to interleaving and worth naming. During the practice, you are worse at each individual decision than you would be in a blocked session. You are less fluent. You are making smaller mistakes you would not have made if you had been warmed up on this decision type. The fluency gap is real and felt.

The research is blunt about this. Subjective fluency is a trap. The practice that feels the most productive , the one where you are in flow, clicking through cases, never stumbling , is the one that produces the least transferable skill. The practice that feels the most effortful , the one where you keep having to reach for the right tool, where you stumble, where you second-guess yourself mid-decision , is the one that produces the skill you actually need.

Most operators, left to themselves, will revert to blocked practice. It feels better. They will notice the stumble in interleaved practice and interpret it as evidence that the practice is not working. They will go back to deep dives on single decision types. They will continue to feel skilled in their practice sessions and continue to flounder in their Tuesday mornings. The gap between how skillful their practice feels and how skillful their actual decisions are will quietly grow.

What this implies for program design

Lucidity is built around interleaving in ways that are visible and ways that are not. The visible version: Day 2 afternoon. The less visible version: Day 3 peer pods rotate across four decision types within ninety minutes, never drilling one type. The post-program 90-day practice structures weekly peer calls around alternating decision types. The calibration bank rotates through six domains so that participants never answer three questions in a row from the same category.

The design choice is deliberate and is the opposite of what most executive programs do. Most programs optimize for the session feeling good. The Academy optimizes for the skill surviving the session. The sessions are frequently harder to sit through. That, too, is part of the specific thing we are teaching.

The finding is narrow and old. It survives replication. It is under-applied. Executives who want to make their practice more expensive and more useful at the same time can start by mixing the problems and noticing when the mixing feels wrong. The feeling is the evidence that the skill is being trained. The training is the whole point.