EVIDENCE > AFFIRMATIONS

Confidence Is Built Backwards

By Seçil Sayhan, MSc Clinical Health Psychology & WellbeingUpdated July 2026

The takeaway

confidence isn't a feeling you talk yourself into. it's a record of times you kept a promise to yourself.

What’s in this article

  1. The thing nobody tells you about confident people
  2. What Bandura's research actually found
  3. Why talking yourself up usually backfires
  4. The smallest version that actually works
  5. "But what about people who fake it till they make it?"
  6. Self-trust is the asset underneath everything
  7. Frequently asked questions

For about three years I had "I am capable" written on a sticky note stuck to my bathroom mirror. I read it every morning. I felt exactly nothing. The note didn't work because I had the whole machine running in the wrong direction.

The thing nobody tells you about confident people

Watch someone you'd describe as confident. Not loud. Steady. The kind of person who says they'll handle something and you stop worrying about it.

Here's what's actually going on, and it has almost nothing to do with feelings. They have a long history of doing what they said they'd do. To others, yes. But mostly to themselves. They told themselves they'd train, and they trained. They said they'd send the hard email, and it went. None of it was dramatic. It just kept happening.

For years I read that backwards. I thought confident people felt a certain way and then acted from it, and that I was missing the feeling. So I went looking for the feeling. Affirmations. Hype playlists. Visualizing the win. I'd talk myself up before a hard meeting and walk in still shaking.

The shaking wasn't a character flaw. It was honest. Some part of me knew I hadn't earned the claim yet. My nervous system was checking the records, finding nothing, and refusing to sign off. That's not weakness. That's the system working correctly. The fix was never to argue with it louder. It was to give it something to read.

What Bandura's research actually found

Albert Bandura spent decades studying where the belief "I can do this specific thing" comes from. He called it self-efficacy, and the key word is specific. Not "I'm a great person." More like: can I give this presentation, can I stay calm in that conversation, can I finish the thing I keep abandoning at 80 percent.

When he mapped the sources of that belief, he found four. Watching people like you succeed counts for something. Encouragement from people you respect counts for a little. Your physical state matters too, because a body that reads its own racing heart as danger will conclude you're not ready.

But one source outweighed the rest by a wide margin: mastery experiences. Actually doing the task and completing it. Direct, lived proof.

This is the part that reorganized my thinking. The strongest input to belief isn't anything anyone says to you, including yourself. It's the receipt from a completed action. Affirmation sits near the bottom of the list, which is exactly where my sticky note kept landing in practice.

There's a reason for the ranking. Words are cheap and the brain knows it. Anyone can say "you've got this." A finished, slightly-too-hard task can't be faked. It either got done or it didn't. So that's the evidence the brain trusts most, and trusts first.

Why talking yourself up usually backfires

Affirmations don't just fail quietly. For a lot of people they make things slightly worse, and there's research suggesting why. When you repeat a confident statement you don't yet believe, your mind doesn't absorb it. It runs a fact-check. "I am capable" instantly summons every recent time you weren't, and now you're staring at the gap instead of the goal.

The sentence becomes a measuring stick for how far short you fall.

This is the core error in most confidence advice. It treats belief as the cause and action as the effect. Feel ready, then move. In real life the arrow points the other way. You move while still unsure, the action completes, and only then does the belief update. Late. Quietly. Built on something that actually happened.

I'm not against the words. I'm against their position in the sequence. An affirmation before the work is a promissory note with nothing behind it. The same sentence after the work is just an accurate description. "I can have hard conversations" lands completely differently the day after you had one.

So when affirmations don't move the needle, the problem usually isn't the wording. You can't phrase your way past an empty ledger. You can only fill it.

The smallest version that actually works

The practical unit is not a better affirmation. It's a kept promise. One thing you said you'd do, done, on a scale small enough that you won't negotiate your way out of it.

Make it almost embarrassingly small at first. Not "I'll work out an hour a day." Try "I'll put my shoes on and walk to the corner." The size isn't the point. The completion is. You're not chasing a fitness outcome. You're filing one piece of evidence that your word means something.

A few things make this hold:

Keep it specific and finishable in a day, so there's a clean yes or no by tonight. Vague goals never generate evidence because they never clearly finish.

Make it slightly hard. Mastery only registers if the task cost you something. Trivial wins don't get recorded.

When you complete it, pause and actually notice. Don't sprint to the next thing. The brain logs evidence better when you mark it. This is the one moment the words belong: "I said I'd do that, and I did."

Then stack. Tomorrow, the same size or a notch up. You're not building motivation. You're building a record. After a few weeks the record gets long enough that the next hard thing feels less like a leap and more like the obvious next entry.

"But what about people who fake it till they make it?"

Fair pushback. Some people do seem to perform confidence and grow into it. But look closer at what's happening underneath, because it's not the performance doing the work.

When someone "acts as if" and it sticks, it's because the acting got them to take the action. They walked into the room they were scared of. They asked the question. The bravado was a delivery vehicle for a mastery experience, and the experience is what stuck. Strip out the doing and the posture collapses inside a week. We've all met the person who's all talk. The talk never converted to evidence.

The other honest objection: what about failure? If confidence runs on completed actions, doesn't a missed promise tear it down?

This is where the small scale earns its keep. A blown promise on a tiny commitment costs almost nothing, and it teaches you to set a more honest target tomorrow. The danger is the oversized promise you break loudly and often, because then your brain logs the opposite lesson: my word doesn't hold. That's how people end up unable to trust themselves with anything. Not from one big failure. From a hundred small commitments quietly abandoned. The repair runs the same direction. Small. Kept. Repeated.

Self-trust is the asset underneath everything

Confidence sounds like a soft topic until you notice what's actually being built. Every kept promise is a deposit into self-trust, and self-trust is the thing that lets you commit to anything bigger than today.

The person who has kept small promises for a year can make a large one and believe it, because they have a track record telling them they follow through. The person who's broken every promise to themselves can't commit to a five-year plan no matter how inspiring it is. Some part of them already knows the odds.

This is why I get uneasy when confidence gets sold as a mindset, a morning routine, a recording you listen to. Those treat it as a mood to summon. It isn't a mood. It's infrastructure, and you lay it one brick at a time, mostly on days you don't feel like it.

The good news in that: it's available to anyone, and it doesn't require you to be a certain kind of person first. You don't need to feel ready. You need one small thing you said you'd do, and the willingness to do it before the belief arrives. The belief comes. It just comes last.

This backwards-built kind of self-trust is most of what we work on inside NextGen at marsa.ai — turning it from a feeling you wait for into a system you run.

Confidence is a record, not a mood. It's built backwards: you do the hard thing first, and the belief shows up later, standing on evidence you actually made.
if you want to build this backwards on purpose — evidence, kept promises, a self that updates — that's what NextGen is for ($247). the full system
Explore NextGen →

Frequently asked questions

Do affirmations ever work?

They can, but not in the order most people use them. Repeating a belief you don't have yet tends to make the gap more obvious, and research on self-statements has found they can leave people who already doubt themselves feeling worse. Where words help is after the action. Once you've actually done the hard thing, naming it ("I said I'd do that, and I did") reinforces a record that already exists. The affirmation describes the evidence rather than trying to manufacture it.

What exactly is a "mastery experience"?

It's the direct experience of attempting a specific task and completing it, especially one that cost you a little effort or nerve. In Bandura's research it's the single strongest source of self-efficacy, ranking well above encouragement or watching others. The important detail is that it has to be genuinely yours and genuinely finished. Watching a video about doing the thing doesn't count. Almost doing it doesn't count. The completed action is what your brain logs as proof.

How small should the first promise be?

Smaller than feels worthwhile. The instinct is to start with the goal that matters, but a big promise you break teaches your brain that your word is unreliable, which is the exact opposite of what you want. Start with something you're almost certain to finish today, like a five-minute version. The point isn't the result. It's filing one clean piece of evidence that you do what you say. You scale up once the record is established, not before.

What if I break the promise?

On a small commitment, very little is lost, and you've learned to set a more honest target tomorrow. That's why scale matters. The real damage to self-trust doesn't come from one big failure. It comes from a long pattern of small commitments quietly abandoned, which trains your brain to expect that your word won't hold. If you keep breaking promises, the fix is almost always to make them smaller and more specific until you can keep them again.

How long until confidence actually changes?

There's no fixed number, and anyone promising one is guessing. What's predictable is the mechanism: each kept promise is a deposit, and confidence rises with the length of the record, not the size of any single win. Most people notice something shift within a few weeks of consistent small completions, because the next hard thing starts to feel like the obvious next entry in a list rather than a leap. The work is in the consistency, not the intensity.

Isn't some confidence just personality or how you were raised?

Temperament and early experience matter, and pretending otherwise would be dishonest. Some people do start with more baseline steadiness. But self-efficacy is task-specific and built through action, which means it's available regardless of where you started. You're not trying to become a confident type of person in general. You're building proof that you can do specific things, one at a time. That's a system anyone can run, and it doesn't ask you to feel ready first.

About the author

Seçil Sayhan is a behavioral scientist and the founder of MARSA.AI. Trained on both sides of her field — a BA in Business Management, an MSc in Clinical Health Psychology & Wellbeing, a diploma in neuroplasticity, and advanced training in Lifestyle Medicine from Harvard University — she has spent the past decade helping 7,000+ people across 12 countries rewire the systems running their lives. Behavior is one science — whether it moves a person, a market, or a machine. See the full bibliography at marsa.ai/research.