Motivation fades.
Hacks break.Boring work compounds.

Self-improvement that survives contact with real life — systems instead of sprints, sources instead of vibes, honest testing instead of promises. A library, not a feed.

  • 62 articles
  • Six clusters
  • Publishing since 2007
  • Zero stock photography
/01

Start where you actually are

Six entrances. Pick the sentence that sounds like your week, not the topic that sounds impressive.

I know what to do. I don’t do it.

Perfectionism correlates with procrastination at −.03 — essentially nothing. Every common explanation, ranked by the evidence behind it.

Enter

My habits die in week two.

One word hides four different problems: no anchor, avoidance, effort, changed context. Find which one is yours before picking a method.

Enter
/02

Four numbers this site will not let you unsee

Every chart below is drawn from a paper you can open. No illustration is decorative, and none of these figures appear on the sites that give you the same advice.

-0.6-0.4-0.2+0.2+0.40Conscientiousness−.62Impulsiveness.41Perfectionism−.03
r · correlation with procrastination
35of 97 replicatedEach square is one experiment with an original effect.
Open Science Collaboration 2015 · 270 researchers
0small .2medium .5large .8d = 0.0495% CI −0.07 to 0.15 · crosses zero23 labs · 2,141 people · one preregistered protocol
Hagger et al. 2016 · Cohen’s d
0501001502002506621 — the number everyone repeatsmedian 66 · IQR 39–102 · range 18–25496 volunteers · one new habit each · measured daily
Lally et al. 2010 · days to automaticity
/03

Start with these

Three pieces that show what the rest of the library is like.

/01

Why We Really Procrastinate

The longest piece on this site. Delay as short-term mood repair, the neurobiology underneath it, and a protocol for the state you are actually in.

Procrastination
/03

How Long Does It Take to Form a Habit?

The 21-day rule traced to its origin: a 1960 note by a plastic surgeon about patients adjusting to a new face. The first real measurement found a median of 66 days.

Myths
/04

Eighteen years of being wrong in public

In 2015, two hundred and seventy researchers repeated a hundred psychology experiments. Thirty-five of the ninety-seven that had originally reported an effect produced one again. Some of the studies that failed were in books this site once recommended.

This project has been publishing since 2007. Five things it got wrong, by name:

An author who has never been wrong isn’t a sage. He’s someone who never checked.

  • Twenty-one days to a habit.The figure belongs to a plastic surgeon writing in 1960 about how long patients took to stop flinching at a new face. He was not writing about habits at all.
  • Ten thousand hours.A rounded average from one study of violinists, turned into a law by a book the researcher publicly disowned.
  • Willpower as a muscle that tires.Twenty-three laboratories ran the same preregistered test on 2,141 people. The effect came out at 0.04, with a confidence interval straddling zero.
  • Visualising the finished goal.Picturing the outcome reliably lowers the odds of reaching it. Picturing the obstacle does the opposite.
  • The law of attraction.No mechanism, no measurement, no failed prediction it could not absorb.
/05

Latest entries

The feed lives at the bottom on purpose. A library is organised by subject, not by date.

/06

One email. One tested idea.

The newsletter

No fireworks, no funnel, no course at the end.

When something in the evidence changes, you hear it here before it reaches the listicles.

Unsubscribe in one click. We never sell or share the list.