New Year Vision to February: Identity, Not Discipline

by
Table of contents

🌏 閱讀中文版本

February 1st, 2026. 9:47 a.m.

You open Notion.

That polished goal list is still there.

Gym plan. Rust study. Reading list.

What’s running through your head:

  • Gym membership still has 11 months on it…
  • That Rust book is open to page 30…
  • More “new year, new me” bullshit?

The question worth asking is:

Why do we always know what to do — and still can’t do it?

We assume the gap is discipline. Or method.

That’s understandable.

Most people adjust the surface layer.

They change behavior.

Dan Koe says: change identity.

Identity Is the Engine

Dan Koe’s core point is simple.

Key Insight: Changes at the identity layer are more fundamental than changes at the behavior layer.

If you want a certain outcome, you have to already be living the life that produces it — before you get there.

An engineer who never writes tests suddenly tries to add unit tests.

They can’t — not because they’re lazy.

Not because willpower ran out.

But because their identity is “the person who ships features fast.”

Writing tests, for that identity, means “slow” and “unprofessional.”

Identity shapes behavior. And behavioral evidence loops back to reinforce identity.

If you’re still grinding through on sheer willpower to change —

it means your identity hasn’t changed yet.

You’re still playing the role of “someone who wants to be better.”

Not “someone who already is.”

That’s where the gap lives.

Three Things I Keep Coming Back To

Dan Koe covers a lot. But there are three things I keep returning to.

Here’s how each one shows up in engineering teams I’ve observed.

It’s Not About Willpower — It’s About Identity

This is the core insight.

We tend to assume: Behavior → Result → Identity

The actual logic is: Identity → Behavior → Result

When your identity hasn’t changed, your behavior is a performance.

And a performance isn’t sustainable.

Practically speaking: when you ask yourself “what should I do?” — ask first: “What would my ideal self do in this situation?”

Then go do that.

Don’t ask “can I pull this off?”

Ask “is this what that person would do?”

It’s Not Laziness — It’s Protection

This sounds counterintuitive.

We assume procrastination means lazy.

Adlerian teleology points to something different:

Behavior often serves a psychological purpose.

I watched a colleague delay writing a design doc — always stopping to reorganize IDE settings first.

Last sprint, he’d pushed off three architecture discussions.

On the surface: avoidance.

After a real conversation, it turned out the delay was driven by fear of evaluation — of being judged and found wanting.

Staying in a stable salaried role might itself be a way of pursuing safety and reducing that exposure to judgment.

Your subconscious is protecting you.

It’s afraid of the risks that come with change.

So it uses procrastination to maintain the status quo.

Real change comes from changing the goal itself.

Not “I need to finish this report.”

But “I want to become someone who faces feedback directly.”

Once the goal is reframed that way, some of the friction becomes easier to identify — and easier to adjust.

It’s Not Truth — It’s Psychological Safety

When identity is threatened, people fight or flee.

That’s why people cling to political positions, religious labels, job titles.

Not because of truth.

Because of psychological safety.

When you try to change your identity, the old one activates its defenses.

It says: “You might pay a price for this.” “What will people think?”

These voices aren’t facts.

They’re the old identity protecting itself.

The identity formation loop:

Goal → Perspective → Attention → Action → Automation → Identity → Defense → New goal

You can’t skip the “defense” phase.

You can only understand it, and keep moving.

My Take: From OKR to Identity Engineering

As engineers, we’re comfortable using OKRs to manage projects.

We rarely use identity to manage ourselves.

That’s where I find Dan Koe’s framework most valuable.

Compared to Atomic Habits

James Clear’s Atomic Habits puts systems above goals.

Dan Koe extends that into a one-day vision reset.

Clear gives you identity-based habits and small wins. Dan Koe scales that into a full-day reset protocol.

If you have only the system — no identity — the system breaks down under pressure.

If you have only the identity — no system — identity stays abstract.

Both are necessary.

In long-term behavior change, identity sets the direction; the system holds the execution steady.

The priority order is: Identity > System > Goal

That ordering sounds clean, but in practice, all three layers fight each other.

I tried to push identity change through the system alone — 5 a.m. wake-ups, gym three times a week, 30 minutes of reading every day.

I held it for 11 days, then stopped.

Applied to Career Transitions

This matters especially during career transitions.

I’ve watched senior engineers get stuck in an “executor” identity.

They know technical skill matters.

Precise implementation is a solid foundation.

But the market rewards judgment more.

Practically speaking, when you’re preparing for a transition:

Don’t just write “I know Python.”

Write “what kind of business problems I can solve.”

The first is behavior. The second is identity.

That’s what employers are buying.

The common friction point for senior engineers is translating technical depth into assessable judgment and business context.

Identity reconstruction is the shift from “executor” to “problem definer.”

A friend in tech recruiting told me: “‘I know Python’ on a resume — I skip it. ‘I refactored a 50,000-line payment system’ — that I open.”

Connected to OKR

OKR can become your “game quest.”

  • Objective: your vision.
  • Key Results: your win conditions.

But remember: OKR is a tool.

Identity is the engine.

Without the engine, the tool has a harder time sustaining long-term change.

Executable Next Steps: One-Day Start + Daily Iteration

Enough theory.

Now — how?

Below is an engineer checklist adapted from Dan’s morning/day/night protocol, with step 5 incorporating small wins and an evidence log.

Step 1: Define Your “Anti-Vision”

Take a piece of paper.

Write: “If five years from now, nothing has changed — what will I have lost?”

Be specific.

  • Health?
  • Opportunity?
  • Self-respect?

Make the fear concrete.

Fear is a powerful driver.

An anti-vision is useful for triggering awareness — but long-term it needs to connect back to your vision and daily evidence.

Step 2: Define Your “Minimum Viable Vision”

Write: “What does an ideal ordinary day look like for me, three years from now?”

Don’t write “I want to be CEO.”

Too abstract.

Write a specific scene:

  • What time do you wake up?
  • What kind of work are you doing?
  • Who do you eat with?
  • What do you do in the evening?

Then write the corresponding identity statement.

For example: “I’m an engineer who focuses on architecture design.” “I’m a PM who can own a full product cycle independently.”

Step 3: Set an “Interrupt Autopilot” Reminder

Set 3 random reminders on your phone.

Random times.

One line: “What am I avoiding right now?”

When the reminder fires, stop for 10 seconds.

Answer honestly.

No judgment.

Just observe.

Step 4: The “Three-Horizon” Evening Review

Before bed.

Ask yourself three questions:

  1. One year from now — what capability do I want to have?
  2. One month from now — what project do I want to have completed?
  3. Tomorrow — what are my 2-3 most important actions?

Write only actions.

Not vision.

Vision was handled in step 2.

Step 5: Build the Evidence Log

Turn your goals into an evidence log.

The evidence log is observable behavior that supports the new identity.

  • Boss: next quarter’s performance review.
  • Daily tasks: read 10 pages a day.
  • Rules: no phone after 11 p.m.

Each time you complete one, give yourself a small reward.

Not a material reward.

The satisfaction of clearing a level.

Closing: Identity Is the Signal, Behavior Is the Output

Changing identity doesn’t happen overnight.

It accumulates one day at a time.

Like refactoring legacy code.

You don’t refactor today and wake up to a clean codebase tomorrow.

You make it a little better every time you touch it.

A year later, 4,000 lines becomes 3,000 lines.

Not from any single refactor — from the daily accumulation.

Dan Koe’s framework, at its core, is an identity refactor.

From “someone who wants to change” — refactored to “someone who already has.”

It’s hard.

But it’s worth it.

Next January, which identity’s checklist will you open?

Sources