It Takes 66 Days, Not 21. Why You Quit Too Early

Share

You start something new. A habit. A side project. A new way of working.

The first week feels great. You're motivated. You're doing it.

Then somewhere around week three, it gets hard. Nothing has changed yet. You miss a day. Then two.

And you think: "Twenty-one days. That's what they say it takes. I did twenty-one days, and it didn't work. Maybe this isn't for me."

Here's the problem. That twenty-one day rule is wrong.

[Watch this on youtube or listen on Spotify podcast]

When researchers actually measured it, the average was sixty-six days. For some people, it took over two hundred and fifty.

So you didn't fail. You quit on day twenty-two of a sixty-six day journey.

And it's not just habits. James Dyson built over five thousand prototypes before his vacuum cleaner worked. The team behind Slack failed with a video game before they found a twenty-seven billion dollar product hiding inside it.

One kept going. One changed direction. Both were right. Because they knew when to push and when to pivot.

Today I'll show you five simple rules to stop quitting too early, and to know when quitting is actually the smart move.

This is The Alpha Drive. Agile thinking for business and life.


WHERE 21 DAYS CAME FROM

So where did twenty-one days even come from?

In the 1950s, there was a plastic surgeon called Maxwell Maltz. He noticed something about his patients. After a nose job, it took them about three weeks to get used to their new face in the mirror.

He noticed something similar in patients who had lost an arm or a leg. They kept feeling the missing limb for about three weeks.

In 1960, he wrote a book called Psycho-Cybernetics. And in it, he said these things seemed to require — and this is the important part — "a minimum of about 21 days."

A minimum. Not the full time. Not for everything.

But the book sold millions of copies. Self-help authors quoted it. Then they quoted each other. And somewhere along the way, "a minimum of about twenty-one days to get used to a new face" turned into "twenty-one days to build any habit."

Nobody had actually tested it.

Then, in 2009, a researcher at University College London, Phillippa Lally, and her team did.

They asked ninety-six people to pick a new habit. Something simple. Drink a bottle of water with lunch. Eat a piece of fruit. Go for a run before dinner. And every day, for twelve weeks, the participants reported whether they did it, and how automatic it felt.

The average time to reach automatic? Sixty-six days.

But the average hides the real story. The range went from eighteen days to two hundred and fifty-four days.

Simple habits, like drinking water, came quicker. Harder ones, like exercise, took much longer.

So the honest answer to "how long does it take?" is: longer than you think, and different for everyone.

And that changes everything. Because if you expect twenty-one days, then on day twenty-two, when it still feels hard, you think something is wrong with you.

If you expect sixty-six, day twenty-two is just… day twenty-two. You're right on schedule.

And this is exactly how agile teams think about any change. Not one magic number. Short cycles, honest measurement, and patience across several of them.

So let's turn this into five rules.


RULE 1 — Expect the dip

Seth Godin wrote a short book called The Dip. His idea is simple. Almost everything worth doing follows the same curve.

At the beginning, it's fun. You're learning fast, and you can see progress. Then comes the dip: a long, boring stretch where you put in effort and nothing seems to happen.

Most people quit in the dip. And that's exactly why the people who get through it win. The dip is what keeps the competition small.

You see this everywhere. The gym in January is packed. By February, it's half empty. Not because people changed their minds, but because the fun part ended and the dip began.

In business it's the same. A new marketing channel, a new product, a new process for your team. The first weeks are exciting. Then results are slow, people complain, the old way starts looking comfortable again.

So here's the action. Before you start anything, write down: "Around week three, this will feel hard and pointless. That's normal. That's the dip."

It sounds silly, but it works. When the dip arrives, it's no longer a surprise. It's a checkpoint you predicted. And a predicted problem is much easier to push through than an unexpected one.

Think about learning a language. The first weeks are amazing: hello, thank you, how much is this coffee. You feel like a genius. Then comes grammar. Verb forms. Cases. And for weeks you feel like you know less than before. That's not you being bad at languages. That's the dip. Everybody who speaks a second language went through exactly that valley.

The people on the other side didn't have more talent. They just didn't take the dip personally.


RULE 2 — Missing a day doesn't kill it

This was one of the most interesting findings in that London study. Missing one opportunity to do the habit did not really affect the process. People who missed a day here and there still built the habit.

But that's not how most of us think. We think in all or nothing. "I missed Monday, so the week is ruined. I'll start again next month."

And that's the real killer. Not the missed day. The decision to give up because of the missed day.

A simple rule that helps a lot: never miss twice. Missing once is life. Missing twice is the start of a new habit, the habit of not doing it.

Agile teams know this very well. A sprint where you finish eight out of ten tasks is not a failed sprint. You review what happened, you adjust, and you start the next one. Nobody throws away the whole project because one sprint was messy.

So treat your habits and your business experiments like sprints. A bad week is data, not a verdict.

The same goes for a team. If you introduce a new process and people forget it twice in the first week, that's normal. Don't announce that the process failed. Ask what got in the way, fix that one thing, and continue. Change doesn't happen in a straight line, for people or for companies.

Ask: why did I miss it? Was I too tired in the evening? Then move it to the morning. Was it too big? Then make it smaller. Which brings us to rule number three.


RULE 3 — Make it small enough to repeat

Remember the study. Simple habits, like drinking water, became automatic faster. Bigger, harder ones took much longer.

So if you want something to stick, the first version should be almost embarrassingly small.

Not "I'll run five kilometers every day." Just "I'll put on my running shoes and go outside." Not "I'll write a book." Just "I'll write one paragraph after my morning coffee."

It feels too small to matter. But a small thing done sixty-six times beats a big thing done six times.

In the agile and startup world, there's a name for this: the minimum viable product. You don't build the full thing first. You build the smallest version that lets you learn something, and then you grow it.

Same with a new business process. Don't redesign your whole sales system at once. Start with one thing. For example, every new lead gets a reply within one hour. That's it. Do that for two months. Once it's automatic, add the next piece.

And attach the new thing to something you already do every day. After I pour my coffee, I write one paragraph. After I close my laptop, I put on my shoes. After a client call, I send the summary email. The old habit becomes the trigger for the new one, so you don't have to remember it or find motivation for it.

Small doesn't mean unambitious. Small means you'll still be doing it on day sixty-six.


RULE 4 — Set your review date in advance

Now, here's the honest part. Not everything deserves sixty-six days. Some ideas are just bad. And pushing through with a bad idea isn't persistence. It's stubbornness.

So how do you know the difference?

Let's go back to the two stories from the start.

James Dyson was frustrated with his vacuum cleaner losing suction. He had an idea for a vacuum without a bag. He built a prototype. It didn't work. He built another. And another. Five years, and five thousand one hundred and twenty-seven prototypes.

Each one taught him something. Each one was a little better. And the evidence kept saying: this is getting closer. So he kept going. Today his company is a household name.

Now the other story. In the early 2010s, a small team led by Stewart Butterfield was building an online game called Glitch. They worked on it for years. But the game wasn't growing. Not enough players. So they shut it down.

But while building the game, the team had created an internal chat tool to talk to each other. And they noticed: that tool was the thing they actually loved using.

So they pivoted. That chat tool became Slack. In 2021, Salesforce bought it for about twenty-seven point seven billion dollars.

Dyson persevered. Slack pivoted. Both were right.

The difference was not motivation. The difference was what the evidence was saying.

In his book The Lean Startup, Eric Ries calls this the "pivot or persevere" decision. And the key idea is that you make it on purpose, at a planned moment, based on data, not on how you feel on a bad Tuesday.

So here's the action. Before you start, write down three things.

One: how long you'll give it. For a personal habit, maybe two or three months. For a business experiment, maybe four to six sprints.

Two: what success looks like. Not "it feels good". Something you can measure. Ten new leads. Five kilos lifted more. Three paying customers.

Three: the date you'll review it.

If you watched my video on building a decision-making system, this is the same idea: set your exit rule before you start, while your head is clear.

Then, until that date, you don't ask "should I quit?" You just do the work. And on the review day, you look at the numbers and make one calm decision: keep going, change direction, or stop.

This one rule removes most of the drama. You stop quitting in the dip, and you stop wasting years on something that's clearly not working.


RULE 5 — Judge the trend, not the day

On any single day, progress is almost invisible. One workout doesn't change your body. One sales email doesn't change your business. So if you judge day by day, you'll always feel like nothing is happening.

But zoom out to two months, and you see the real picture.

So measure two kinds of things.

First, the effort. Did I do it? How many days this month? That's what you control.

Second, the result. Am I lifting more? Are more people replying? Is the process faster? That's what you're aiming for.

In the first weeks, look mostly at the effort. Results lag behind. In the study, the habit became automatic gradually, slowly at first and then more and more, until it stopped feeling like effort at all.

And this is exactly how we approach automation projects at NodeHighway. When a team switches to a new process, the first weeks are always a bit bumpy. People forget, the old habits pull back. So we agree upfront what we're measuring and when we'll review it. Then we let the trend speak, not the first bad week.

Here's a simple tool for your own life. Take a calendar, a paper one or an app, and mark every day you did the thing. Don't judge it. Just mark it. Once a month, look at the whole page.

That page will tell you more than any single day ever can.

And for a business, do the same with one simple number. Replies to your emails. Bookings per week. Hours spent on admin. Write it down every Friday, and put the weeks side by side. Don't overreact to one bad week, and don't celebrate one good one. Look for the direction.

Because the trend is the truth. A single day is just noise.


OBJECTIONS

Before we wrap up, let me answer the two objections I hear most.

The first one: "Sixty-six days? That's even more depressing than twenty-one."

I get it. But think about it differently. Twenty-one days was a promise that set you up to fail. Sixty-six days is an honest expectation. And you don't need motivation for sixty-six days. You need a habit small enough that you can do it on your worst day. That's rule three.

The second one: "What if I spend two months and it turns out to be the wrong thing?"

Then you've spent two months learning something, instead of two years wondering. That's what the review date is for. You'll never lose more than one cycle. And honestly, most people don't lose time on the wrong things. They lose it by starting the right things and quitting in week three.

Think of everything you started in the last few years and dropped after three weeks. A course, a sport, a newsletter, a new tool at work. Now imagine if you had given just one of them a fair, planned sixty-six days, with a clear review at the end. Whatever the answer would have been, you'd know. Instead, most of us live with a drawer full of half-started things and a quiet feeling that we're "not the type" who finishes.

You are the type. You just had the wrong number.


CLOSE

So let's put it all together.

Expect the dip, because almost everything gets harder before it gets easier.

Missing a day doesn't kill it. Just never miss twice.

Make it small enough to repeat, so you're still going on day sixty-six.

Set your review date in advance, so you decide with data, not with mood.

And judge the trend, not the day.

Twenty-one days was never the rule. It was a minimum, from a surgeon, about getting used to a new face.

So if you're on day twenty-two of something right now, and it feels hard… good. You're exactly where you're supposed to be.

I'm Alex. This is The Alpha Drive.

If this was useful, subscribe. Consider it day one.

And if you want to go deeper on making the pivot-or-persevere call, my text on building a decision-making system is right here.