How to Have High Agency: A Step-by-Step Guide for Business Improvement & Strategy Professionals
High agency is easy to admire but hard to practice — here’s how to build it step by step
One of the most common compliments McKinsey partners give in consulting is: “He/she has high agency.”
It’s the mark of someone who doesn’t need constant direction, who moves forward even when the path isn’t clear, and who can transform vague inputs into meaningful progress.
High agency is one of those traits that separates the “cube” from the “ball.” A cube, once flicked, comes to a halt unless pushed again. A ball, once nudged, keeps rolling forward on its own momentum. In business, you want to be the ball.
Everyone understands the idea of high agency. The hard part is implementing it - especially in environments where ambiguity is the norm, stakeholders are busy, and the instructions you get are more like a sketch than a blueprint.
In this article, I’ll break down how to develop high agency step by step, so you can move from simply following instructions to being the person your leaders trust to drive outcomes independently.
Why High Agency Matters
In business improvement and strategy (and consulting more broadly), you rarely get a fully-formed, well-defined task. More often you get a loosely phrased objective:
“Can you take a look at our customer churn problem?”
“We need a page on competitor pricing - what’s the story?”
“See if we can work with that supplier in Indonesia.”
If you treat these inputs literally, you risk circling back with “I wasn’t sure what you wanted” - which frustrates leaders and slows everyone down.
By contrast, someone with high agency can take those rough inputs, frame them into a clear problem, test an approach, and quickly return with an 80/20 answer - before iterating toward a gold-plated version if truly needed.
The 8-Step Playbook for Building High Agency
Here’s a systematic way to approach any ambiguous task.
Step 1: Understand the Task to the Extent Possible
Often, the person giving you the assignment hasn’t thought it through fully. That’s not laziness - it’s just reality at senior levels, where ideas come fast and time is scarce.
Your job:
Write down what you think the task is asking for.
Identify what’s missing or unclear.
Make reasonable assumptions so it makes sense to you.
If you can double-check, great. But don’t expect your manager to give you a perfectly structured answer. High performers are comfortable figuring things out from partial inputs.
Example: If asked to “look at customer churn,” decide whether the task is to analyze reasons for churn, benchmark against peers, or propose solutions. Choose one angle to start, and clarify if needed later.
Step 2: Make a Perfect-World Plan
Sketch what you’d do if you had unlimited data, time, and access.
What’s the ideal analysis?
What would the perfect data set look like?
What would the final deliverable include?
This anchors you in the “gold standard” of how the problem could be solved.
Example: For customer churn, the perfect plan might involve a regression on customer-level data, detailed churn surveys, and competitor mystery shopping.
Step 3: Simplify to the 80/20 MVP
Now, cut ruthlessly. What’s the 20% of effort that gets you 60-80% of the answer?
Use assumptions instead of perfect data.
Google sources instead of launching surveys.
Build a simple Excel model instead of a Python pipeline.
The goal isn’t perfection. It’s speed + directional accuracy.
Example: Instead of a regression, start with a quick cohort analysis using the data you have. Instead of surveying 1,000 customers, interview 5 sales reps for insights.
Step 4: Stress-Test Your MVP
Once you have a first cut, ask: Where could this be wrong?
Vary assumptions to see which ones swing the outcome.
Sharpen the critical few assumptions.
Ignore the rest for now.
This makes your MVP robust enough to be credible without wasting effort on immaterial details.
Example: If your churn analysis depends heavily on one assumption about competitor pricing, pressure-test that assumption by checking a few price points online.
Step 5: Share the MVP and Get Feedback
Show your leader the MVP output and ask:
“Is this the direction you had in mind?”
“Here’s what I assumed - do these look reasonable?”
“This is 80/20 - do you want me to take it further?”
This accomplishes two things:
It validates whether you’re on the right track.
It allows your manager to calibrate how much more accuracy is needed.
Often, the MVP is enough to guide decisions.
Step 6: Do the Next 20%
If further work is needed, don’t jump straight to perfection. Ask: What’s the next 20% effort that gives me the next meaningful leap in accuracy or insight?
Maybe it’s gathering real supplier quotes instead of Googling. Maybe it’s cleaning a messy dataset instead of relying on averages.
Deliver again, then recalibrate.
Step 7: Iterate
Repeat the cycle: MVP → feedback → refine.
Each loop adds value without committing you to unnecessary polish. This prevents the classic consultant trap of spending 50 hours on something the client only needed at a directional level.
Step 8: Gold-Plate Only if Needed
If the situation truly demands it - e.g., a board presentation or an investor due diligence - then go for the gold-plated version.
But by this point, you’ll be certain it’s worth the effort, and you’ll have a robust base to build from.
Practical Examples
Example 1: Partner Outreach
Task: “See if we can work with Supplier X in Indonesia.”
MVP: Google Supplier X, check if they export, see if they have existing Western partners.
Next step: Email a contact, test responsiveness.
Gold-plated: Fly to Jakarta, negotiate terms face-to-face.
High agency means you don’t stall because you don’t have a full playbook. You start small, test, and escalate only if promising.
Example 2: Organizing a Workshop
Task: “Put together a client workshop on efficiency.”
MVP: Draft a 3-page outline of topics and a sample agenda.
Stress-test: Ask if the client cares more about cost or customer experience.
Next step: Add detailed exercises and breakout session plans.
Gold-plated: Book venue, professional facilitators, full deck.
Instead of asking “What exactly do you want?”, you push forward with a prototype.
Example 3: Physical Process Improvement
Task: “Improve our warehouse picking process.”
MVP: Walk the floor, observe workers, note inefficiencies.
Stress-test: Compare with benchmarks (items per hour).
Next step: Pilot a new layout in one aisle.
Gold-plated: Roll out redesign across the warehouse with full training and metrics.
Again, you don’t wait for perfect data. You act.
Final Thoughts: Becoming the Ball
High agency is one of the most valuable traits you can develop in a corporate environment. It’s the difference between being a cube that needs constant flicking and a ball that rolls forward on its own.
The steps are simple in theory - but powerful in practice:
Frame the task.
Plan ideally.
Simplify to MVP.
Stress-test.
Get feedback.
Do the next 20%.
Iterate.
Gold-plate if needed.
Master this process, and you’ll stand out as the person who creates momentum where others stall. Leaders will trust you with more responsibility, colleagues will look to you for direction, and your career trajectory will accelerate.
High agency isn’t about having all the answers. It’s about creating answers when the path is unclear.
Be the ball.