All chapters
Brainstorming
Core toolkit - Chapter 9
09

Brainstorming

Generate useful options under pressure by creating buckets first, filling them with specific ideas, stretching with a new lens, then prioritizing what to test or do next.

Read time
9 min
Chapter
09
Level
Core toolkit
What you'll take away
  • Use clear buckets to organize ideas before generating them.
  • Recover from 'what else?' prompts by switching lenses.
  • Prioritize ideas using impact, feasibility, speed, and risk.

In the last chapter, you learned how to estimate an unknown number with a clear path.

Brainstorming is different. Now the interviewer is not asking for a number. They are asking for options.

Prompts that trigger brainstorming
  • What could the client do?
  • What else might explain this?
  • What risks should we consider?
  • What other data could we use if this data is unavailable?

For many candidates, this is the moment the mind goes strangely blank. Not because they have no ideas, but because the prompt suddenly feels wide open. There are too many directions, too little time, and a quiet pressure to sound creative on demand.

The relief
Creativity in a case is not random inspiration. It is organized idea generation.

The intuition

Brainstorming is how you create options with structure.

You are not waiting for a brilliant idea to appear. You are building a few clean categories, then using each category to produce useful ideas. That is the whole move.

If "buckets first" sounds familiar, it should. This is the same skill you built in Chapter 1: clean buckets that overlap as little as possible and cover the main angles, just pointed at options instead of causes.

Scattered

We could do marketing, discounts, partnerships, new products, maybe improve the app, maybe cut costs...

Organized

I would organize ideas into three groups: win back customer traffic, increase average order value, and protect margin. Under traffic, I would consider...

The second version does not depend on magical creativity. It gives your mind places to look. A blank page is hard. A few buckets make the page less blank.

Why it matters

Brainstorming tests whether you can think broadly without becoming messy. The interviewer wants to see:

  • Can you generate multiple relevant ideas?
  • Can you organize them instead of spraying them randomly?
  • Can you go beyond the obvious first two ideas?
  • Can you stay practical and business-relevant?
  • Can you recover when asked, "What else?"
  • Can you prioritize the best ideas instead of ending with a long list?

This matters because consulting work is not only diagnosis. Clients also need choices. Once you know why something is happening, you often need to answer, "What could we do about it?" And the first answer is rarely enough.

The core moves

Brainstorm loopBuckets first. Ideas second.
  1. 1
    PauseWhat type of ideas are we generating?
  2. 2
    FrameWhat 3-4 categories will guide the brainstorm?
  3. 3
    GenerateWhat ideas fit inside each category?
  4. 4
    StretchWhat else, using a new lens?
  5. 5
    PrioritizeWhich ideas are most attractive?
  6. 6
    MoveWhat would we test or do next?
A blank page is hard. A few buckets make the page less blank.
Sentence frame
I will take a moment to organize ideas. I would think about this in [number] groups: [bucket 1], [bucket 2], and [bucket 3]. Within each, I will generate a few options, then I will prioritize based on [filter].

Throughout this chapter, we will keep using the same running case:

A coffee chain's profits dropped 15% last quarter, driven mainly by lost customer traffic in Region B after a new competitor opened nearby.

Now the client asks:

What could we do to recover profit in Region B?

1. Pause: understand what kind of brainstorm this is.

Not all brainstorming prompts ask for the same thing. Sometimes the interviewer wants causes, sometimes actions, sometimes risks, and sometimes data sources. Before you answer, take a short pause and identify the job.

Prompt typesDifferent prompts need different frames
Causes
Why might this be happening?
Use driver buckets.
Solutions
What could we do?
Use action buckets.
Risks
What could go wrong?
Use risk categories.
Data sources
How else could we learn?
Use ways to observe behavior.
Growth ideas
How could we expand?
Use revenue or customer buckets.
Cost ideas
How could we reduce cost?
Use cost-driver buckets.
A clean pause
Let me take a moment to organize the options. Since the goal is profit recovery, I would group ideas by revenue recovery, margin protection, and customer retention.

2. Frame: create buckets before ideas.

This is the move that separates a strong brainstorm from a messy one. Buckets before ideas, every time.

Option frameProfit recovery in Region B
Recover profit
Recover traffic
bring customers back
  • local offers
  • office partnerships
  • commuter promotions
Increase AOV
raise spend per visit
  • breakfast bundles
  • premium drinks
  • app upsell
Protect margin
avoid buying volume with bad economics
  • targeted discounts
  • labor scheduling
  • waste reduction
Profit recovery can come from more customers, more spend per customer, or better economics per order.

You could also frame the same brainstorm by time horizon or customer journey.

Time horizon

Immediate moves, medium-term improvements, and longer-term strategic changes.

Customer journey

Awareness, consideration, purchase, and retention.

There is no single perfect frame. The best frame is the one that fits the prompt and helps you generate useful ideas without overlap.

3. Generate: make the ideas specific enough to be useful.

Once you have buckets, fill them. For Region B:

Idea mapSpecific options by bucket
Region B profit recovery
Recover traffic
lost customers
  • targeted winback offers for lapsed loyalty members
  • morning-commute promotions near affected stores
  • office or gym weekday coffee bundles
Increase AOV
more spend per visit
  • coffee plus breakfast bundles
  • higher-margin seasonal drinks
  • app-based upsell offers for frequent customers
Protect margin
better economics
  • avoid broad discounting
  • improve labor scheduling in slower dayparts
  • adjust pastry production to reduce waste

Notice what makes these better than generic ideas. They are tied to the business, the region, the likely customer behavior, and the profit goal.

4. Stretch: handle "what else?" without panicking.

"What else?" is one of the most uncomfortable questions in a case. Often the interviewer is just testing whether you can keep generating under pressure. The move is simple: do not search harder inside the same bucket. Change lenses.

Stretch lensesBackup doors when one frame runs dry
Profit levers
Revenue, cost, margin
Useful for profit recovery.
Customer journey
Awareness, trial, purchase, repeat
Useful for demand and retention.
4Ps
Product, price, place, promotion
Useful for marketing and go-to-market.
Time horizon
Now, next quarter, longer term
Useful for implementation planning.
Stakeholders
Customers, employees, suppliers, partners
Useful for ecosystem problems.
Risk types
Financial, operational, customer, competitive
Useful for recommendation pressure tests.
Data sources
Internal data, external data, experiments, proxies
Useful when information is missing.
When one frame runs dry, open another.
What else response
Another way to think about this is the customer journey. Before purchase, we could increase local awareness with geo-targeted ads near the competitor. At purchase, we could improve speed with mobile order pickup. After purchase, we could use loyalty data to trigger a return offer for customers who have not visited in 30 days.
Blank-mind resetChange the lens and keep moving
  1. 1
    PauseTake one second to reset.
  2. 2
    Name a new lensCustomer journey, operations, time horizon, risks, or data sources.
  3. 3
    GenerateAdd one or two ideas inside that lens.
  4. 4
    Tie backConnect the ideas to the client's objective.

5. Prioritize: do not leave the interviewer with a pile.

Generating ideas is only half the move. The other half is judgment. After you brainstorm, pick the most attractive ideas using a simple filter.

Prioritization filterTurn a list into client advice
Impact
How much could this move the result?
Does it solve the main issue?
Feasibility
How easy is it to execute?
Can the client actually do it?
Speed
How quickly could it work?
Does it fit the urgency?
Risk
What could go wrong?
Could it hurt margin, brand, operations, or customers?
Prioritized close
I would prioritize targeted loyalty winback offers and local morning-commute bundles first. They are specific to the traffic problem, can be launched quickly, and are less likely to damage margins than broad discounting. I would deprioritize chain-wide price cuts because they may recover volume but hurt profitability.

6. Move: connect the brainstorm back to the case.

A strong brainstorm ends by saying what you would test next.

Next analysis
To choose among these, I would first look at loyalty data to identify which customer segments stopped visiting, then test targeted offers in a small set of Region B stores and track traffic, average order value, and margin.

Now the ideas are not floating. They are connected to evidence and action.

What weak looks like

Weak brainstorm · scattered list
AI
Prompt
The coffee chain's Region B traffic fell after a competitor opened nearby. What could the client do to recover profit?
YOU
Weak answer
They could do marketing, discounts, partnerships, new drinks, maybe improve the app, maybe cut costs, maybe close stores, and maybe do social media.
Why it falls flat
The ideas are scattered. Some are revenue ideas, some are cost ideas, some are channel ideas, and some are too vague to be useful. The candidate never explains the categories, does not connect the ideas to Region B's traffic issue, and does not prioritize.

What strong looks like

Strong brainstorm · organized options
AI
Prompt
The coffee chain's Region B traffic fell after a competitor opened nearby. What could the client do to recover profit?
YOU
Strong answer
I would organize the options into three groups: recover traffic, increase average order value, and protect margin. To recover traffic, we could launch targeted winback offers for lapsed loyalty members, run morning-commute promotions near the affected stores, or partner with nearby offices for weekday coffee bundles. To increase average order value, we could bundle coffee with breakfast items, promote higher-margin seasonal drinks, or use app-based upsell offers. To protect margin, we should avoid broad discounts and instead target offers to customers who stopped visiting, while also adjusting labor schedules and pastry production to match the new demand. I would prioritize targeted loyalty winback and morning bundles first, because they directly address lost traffic and can be tested quickly without cutting price across the whole chain.
Frame, multiple ideas, practical specificity, and prioritization all show up.

The answer is creative because it is organized. Not despite the structure. Because of it.

Common traps

The usual ways brainstorming goes wrong
  • Listing ideas without categories.
  • Choosing categories that overlap.
  • Stopping after two obvious ideas.
  • Trying to be wildly creative instead of useful.
  • Giving generic ideas that could apply to any company.
  • Ignoring the client's objective.
  • Forgetting constraints like margin or implementation cost.
  • Ending without prioritizing.

The most common trap is thinking creativity means less structure. In cases, it usually means better structure.

How to use this later in CaseLab

When a live case later asks for ideas, use this chapter as your feedback lens. The move is not to be wildly creative on command; it is to pause, create clear buckets, generate specific ideas, and close by prioritizing the best options.

Later, in a CaseLab rep, review whether your brainstorm was:

  • Structured into clear buckets.
  • Broad enough to cover multiple paths.
  • Deep enough to include specific ideas.
  • Practical and tied to the case.
  • Responsive to "what else?" prompts.
  • Prioritized using impact, feasibility, speed, and risk.
The standard
Do not just ask, "Did I have enough ideas?" Ask: did my ideas help the client see better options?
Up next
Chapter 10 · Business intuition
Next, learn how to make structures, hypotheses, math, and recommendations feel specific to the business instead of generic.
Continue to Chapter 10
Finished Chapter 9?Mark it complete to track your progress through the handbook.
← PreviousMarket sizingUp next →Business intuition