There is a small, common moment that most of us get exactly backwards. Your child puts on their shoes without being asked. Something in your chest loosens. And you say the thing every parent says: Good job, buddy! You mean it as a gift. But watch what happens over the next few weeks. The shoes stop going on unless you're in the room. The tooth-brushing pauses, mid-brush, so they can check your face. Somewhere along the way, a routine your child was doing for themselves quietly became a performance they are doing for you — and neither of you noticed the switch.

This isn't an argument for withholding warmth. It's a much stranger finding than that. Decades of research on praise suggests that how we say the encouraging thing changes what the child learns from doing the thing. The same two syllables can build a kid who gets ready because it feels good to be capable, or a kid who gets ready to keep a face from going cold.

The difference between praising the child and praising the doing

In a well-known line of research led by Melissa Kamins and Carol Dweck, young children were praised in one of two ways after succeeding at a task. Some got person praise — praise directed at the self: You're such a good girl. You're so smart. Others got process praise — praise directed at what they actually did: You found a good way to do that. You kept trying different ways.

Then the researchers gave the children a setback.

The kids who had been praised for who they were took the failure hardest. They were more likely to judge themselves as bad or not smart, to feel worse, to want to quit. The kids praised for what they did were more likely to treat the setback as information — something about the strategy, not something about them.

The logic is quietly brutal once you see it. If success means I am a good girl, then failure means the opposite is available too. Person praise doesn't just hand out a compliment. It hands out a scoreboard, and the child stays on it whether they're winning or not. This is why the child who is told they're a great helper can crumble so completely on the morning they forget their backpack — you didn't intend to make their character depend on their backpack, but the sentence did.

Why "good job" starts to feel like a leash

There's a second mechanism, and it comes from self-determination theory, the motivation research of Edward Deci and Richard Ryan. In that framework, praise can carry two different messages at once. It can be informational — telling you something true about your growing competence. Or it can be controlling — telling you that someone with power is pleased, and will be pleased again if you repeat the behavior.

Informational praise tends to support the child's own sense of agency. Controlling praise tends to erode it, because the reason for doing the thing migrates outward, from inside the child to inside the adult. Jennifer Henderlong and Mark Lepper, reviewing this literature, argued that praise supports motivation when it's sincere, when it points at things the child can control, and when it doesn't turn into pressure to keep performing.

Good job is not evil. It's just empty enough to be filled with whatever the child suspects. And what a child often suspects — reasonably — is that you were watching, and that the watching was the point. A four-year-old cannot articulate this. They just start looking up after every step.

The most underrated sentence in parenting is not praise at all

Here is the move that changes the most for the least effort: describe what you see, and stop talking.

"You got your shoes on, and you did the velcro yourself."

That's it. No evaluation. No verdict. You are not telling your child they are good. You are handing them a clear, accurate mirror and letting them draw their own conclusion — which, reliably, is I can do this. Adele Faber and Elaine Mazlish built much of their communication work on exactly this, and it maps neatly onto what the motivation research would predict: the child gets the information about competence without the surveillance.

The test I use: could this sentence be true if I felt nothing about it? If yes, it's a description. If it only works because I'm pleased, it's a verdict.

There's a related move for the harder mornings — the ones with no success to describe. Instead of praising an outcome that didn't happen, name the effort or the strategy that did: You were really frustrated with that zipper and you stayed with it. You are not manufacturing a win. You are pointing at the part of the story the child can actually repeat tomorrow.

What this has to do with routines specifically

A routine is supposed to end up running without you. That is the entire point of it. Every morning you spend as the audience is a morning the routine cannot become automatic, because the sequence has an extra step wired into it: do the thing, then check whether Mom's face changed.

So the goal of routine-time language is not to motivate. It's to get yourself out of the loop.

This is why so many parents find themselves trapped: the more enthusiastically they cheer, the more the child needs the cheering. You have made yourself load-bearing. Then, one Tuesday, you're tired, you don't cheer, and the whole thing falls apart — and it looks like the child regressed, when actually the reinforcement you'd installed just went quiet.

The fix is to move the feedback out of your mouth and into the routine itself. A child who can see, without asking, that four of five things are done has already received the information your praise was carrying. She doesn't need your face. She has the evidence.

Your next moves

  • Run a one-day count. Tomorrow morning, silently tally how many times you evaluate your child (good job, nice work, perfect, that's my girl). Don't change anything, just count. Most parents are shocked. You cannot adjust a habit you can't see.
  • Swap one sentence, not all of them. Pick a single recurring moment — shoes on, plate in the sink — and replace your usual praise with a plain description: "Plate's in the sink." One moment. Let the rest stay messy.
  • Rewrite one piece of person praise into process praise, in advance. Take the phrase you say most (you're such a good helper) and pre-write its replacement (you carried that all the way to the kitchen without spilling). Say it out loud once in the car so it's available when you need it.
  • Practice the three-second pause. When your child finishes a step and looks up at you, count to three before responding. Often they'll turn back to the next step on their own. That turn is the whole thing you're trying to build.
  • Make one step of the routine self-verifying tonight. Put the pajamas where the child can see the pile shrink, or the toothbrush where they can see it's wet. Anything that answers did I do it? without a parent in the sentence.

When the routine can tell them

This is the quiet reason visual routines work on parents as much as on children. When a child can see the sequence in front of them and watch it complete itself, the feedback stops arriving as a verdict from someone who loves them and starts arriving as a fact about the world. Rhythm was built for that handoff — a routine your child can read, move through, and finish without looking up to see whether they've earned it. If you've noticed you've become the scoreboard, it might be time to hand the job to something else: rhythm.lumenlabs.works. Your kid gets to be capable. You get to just be glad.