Why solving 100 cases won't make you structured

Almost everyone preparing for consulting interviews follows the same plan: get a casebook, find a partner, and do as many cases as possible before the first round. Thirty cases becomes fifty, fifty becomes a hundred, and somewhere along the way the hope is that structure "clicks."

For some people it does. For most, progress flattens after the first twenty or so cases. They get faster at the cases they've seen, but a new prompt still produces the same blank pause and a framework that doesn't quite fit.

Volume trains recognition, not reasoning

When you work through a casebook, you're mostly learning what those cases look like. The fourth airline profitability case feels easy because you remember the third. That's recognition, and it's useful, but it isn't the skill the interviewer is testing.

The interviewer wants to see you take a problem you have never seen and break it into pieces that are distinct, complete and ordered by importance. That's a reasoning skill, and like any skill it improves through focused practice on its parts, not just through repetition of the whole.

Three things that work better than another case

1. Master the building blocks first

Every structure you'll ever draw is assembled from a handful of ideas: the profit equation, the difference between price and volume, the customer–competitor–company split, a way to size a market. If any of these is shaky, every case built on it wobbles. Make sure each block is solid before you stack cases on top of it.

2. Mix your practice

Doing five profitability cases in a row feels productive, but the hardest part of a real interview, working out what kind of problem you're facing, never gets practised. Mixing case types forces you to make that call every time. It feels harder and slower. That difficulty is the point.

3. Come back after a gap

A concept you understood on Monday is half gone by Friday. Revisiting it just as it starts to fade, then again a little later, is one of the most reliable findings in learning research. A short review session beats a long cramming session every time.

Getting better at cases isn't about how many you've seen. It's about how reliably you can build a good structure for one you haven't.

What this looks like in practice

Instead of "do three cases tonight," try: one short concept refresher, a mixed set of case openings where you only build the structure, then one full case where you lead and your partner reveals facts only when you ask the right question. Score yourself on structure separately from the maths, so you know which one needs the next rep.

That's the loop lowercase is built around: learn one idea, prove it, use it in a case, and bring it back later. It takes fewer cases than you'd expect.