The Socratic Framework: How Asking Questions Wins Everything

March 15, 2026
5 min · What I Learned From...
Script · Cross-Cutting Theme

COLD OPEN

[SCREEN: Black. A question mark appears.]

NARRATOR (V.O.):
I don't argue. I ask questions.

Each question is designed so the only
honest answer confirms one building
block of my position.

By the time I get to the fifth question,
the conclusion is inescapable.

[Five question marks appear in a cascade,
each one clicking into place like a lock.]

This works in insurance disputes.
In sales. In negotiation. In debugging AI.
In every single domain I've applied it to.

This is the Socratic framework.

ACT 1: THE INSURANCE CLAIM (0:00–2:00)

NARRATION: "A water damage claim. The insurer offered me two options for temporary housing. I chose Option 2 — Fair Rental Value. Their own assessment came back: $14,510 to $19,610 per month. Then the adjuster changed her mind."

[ANIMATION: An email appearing on screen.
The key phrase highlighted.]

ADJUSTER: "Under FRV, we provide a payment
based on the market rental value of your home,
which you can manage yourself. You may retain
any remaining funds if your chosen
accommodations cost less."

[BEAT.]

NARRATOR (V.O.):
I accepted. They ordered the assessment.
Then the same adjuster reversed course.

[A second email appears.]

ADJUSTER: "It is illegal for you to profit
from a claim."

[The two emails side by side. The contradiction
glowing red.]

Instead of arguing, I asked questions.

Storyboard Frame: The Socratic Question Chain

"Did the adjuster make this determination on behalf of the insurer?" Yes "Did I accept it in writing?" Yes "Did the insurer implement it by ordering the FRV assessment?" Yes "Has anything in the policy changed since the determination?" No "So what changed?" "It was a mistake." — The adjuster Each question has only one honest answer. The logical conclusion is inescapable. A determination was made, accepted, and implemented. You can't unwind it because you realized it was generous.

NARRATION: "I consulted friends, lawyers, people who'd been through insurance disputes. Every single one told me I wouldn't win. The consensus was unanimous: insurance companies have teams of attorneys, they've seen every argument, the cost of fighting exceeds the payout. Give up."

NARRATION: "I didn't give up. Not out of stubbornness — out of pattern recognition. The adjuster had created a logical chain she couldn't break without contradicting herself. Every email she sent was a building block in my case."


ACT 2: THE PATTERN IN SALES (2:00–3:30)

[ANIMATION: The question chain dissolves and
reforms in a different context — a text
conversation between an AI agent and a renter.]

NARRATOR (V.O.):
The same pattern shows up everywhere.

In apartment sales:
# From the actual CLAUDE.md — the AI's sales doctrine

## INVERT THE BURDEN OF PROOF

Constantly lower the bar so saying NO
becomes the insane option.

"At a minimum, if nothing else worked out,
 could you make this work?"

"I've got a renter ready to sign.
 Want the commission or should I send them
 across the street?"

NARRATION: "Each question is designed so the honest answer moves the deal forward. 'Are you still looking for an apartment?' Yes. 'Is next month still your move date?' Yes. 'Would you like to see a place that's $200 cheaper than what you clicked on?' Yes. 'Can I set up a tour for Thursday?' By the time you get there, saying no to the tour is the irrational choice. Every yes was a building block."


ACT 3: THE PATTERN IN DEBUGGING (3:30–4:15)

[ANIMATION: A terminal. Code scrolling.]

NARRATOR (V.O.):
Even debugging AI agents follows the
same framework.

"Did the lead respond?" Yes.
"Did the classifier detect the intent?" Yes.
"Did the state machine transition?" Yes.
"Did the response template render?" Yes.
"Did the SMS actually send?"

...No.

[The Twilio logs appear. Rate limit exceeded.
The chain broke at step 5.]

The question chain isolates the failure
to a single component. You don't debug
by reading thousands of log lines.
You debug by asking yes/no questions
until the chain breaks.

ACT 4: THE PRINCIPLE (4:15–5:00)

NARRATOR (V.O.):
The Socratic framework doesn't persuade.
It reveals.

The logic was already there.
The questions just make it visible.

[ANIMATION: The question chain from the
insurance case, the sales conversation,
and the debugging session — all three
side by side. The structure is identical.
Five yes/no questions. A cascade. A conclusion.]

You're not arguing.
You're not getting emotional.
You're asking questions that have only
one honest answer.

And you're building a logical chain
that's impossible to break without
admitting to either incompetence
or bad faith.

[FADE TO BLACK.]

This is the most powerful tool I own.
It cost me nothing to build.
It works in every domain.

Socrates knew. Twenty-four hundred
years later, it still works.

Production notes: Animated question chains that build visually — each question clicking into place with a satisfying lock sound. Three parallel examples (insurance, sales, debugging) shown side by side to reveal the structural identity. Tone: measured confidence, not aggression. The power is in the calm.