
You have probably felt it for years. They have probably never heard it. Before anything else, before any explanation of how this works — one question.
7
people
7
different days
10
day window
You think about sending the message. You compose it in your head on the drive home. You decide you'll call this weekend, when things are calmer and you can say it properly. Then Tuesday becomes next week, and next week becomes the vague and permanent country of someday.
Nothing is wrong with you. Appreciation asks you to be sincere without an occasion, to interrupt somebody's ordinary day with something good, and to sit in a few seconds of not knowing how it will land. Most of us decide that's a job for a better version of ourselves, later.
Appreciation kept inside reaches one person.
Expressed, it reaches two.
This experiment does not ask you to feel more than you already do. It asks you to send it — seven times, on seven different days, with something small at stake so that someday never gets a vote.


Researchers at Chicago Booth and UT Austin asked people to write a letter of genuine gratitude, then predict how the recipient would feel. Then they asked the recipients. The senders were wrong in a consistent, very human direction.
"It'll be awkward."
Recipients reported markedly less awkwardness than senders predicted.
"They already know."
Recipients were far more surprised by why they were appreciated than senders expected.
"I'll say it badly."
Senders agonised over wording. Recipients were moved by the fact that someone bothered.
Your expectation of awkwardness is the single best predictor that you'll stay silent. It also happens to be inaccurate.
Expressers underestimated how positive recipients felt, and overestimated how awkward it would be.
Kumar & Epley (2018), Psychological ScienceEverything hard about this happens in your chest, not on this website.
Enrol, and place a small refundable amount behind your own intention. The exact arithmetic is the next section, and it's the most interesting part of the design.
Today, tomorrow, or any day in the next three. Pick the version of you that's ready. If you never choose, day one begins on its own — flexibility, but no someday.
Text, call, email, voice note, handwritten card, or face to face at the kitchen counter. Seven people, seven different days. How you say it doesn't matter. That you send it is the entire experiment.
One tap a day on your private dashboard. Reach seven and your eighteen dollars returns to your card immediately — you don't wait out the remaining days.
Behavioural economists call this a commitment contract: you deposit money you only get back by following through. It works because a possible loss sitting in the present is more motivating than a benefit floating somewhere in the future. We will work considerably harder to avoid losing eighteen dollars than to earn eighteen dollars.
Odysseus had himself lashed to the mast so he could hear the sirens and not jump. Cortés scuttled his own ships. You are wagering eighteen dollars that you will tell seven people the truth. It is the least dramatic and most effective version of the same idea.
The amount is deliberate, too. The larger the deposit, the fewer people are willing to begin. Eighteen dollars is set to be small enough to say yes to tonight and large enough that you'll remember it on day four.
In one randomized trial, roughly half of participants on a 16-week commitment contract hit their goal, against one in ten receiving usual care.
Halpern, Asch & Volpp (2012), BMJSo here is the whole of it, in three lines and no small print.
Where the money goes
$25
Today
One payment through Stripe when you enrol. Nothing recurring, no subscription, no upsell later.
$7
Kept
The actual cost of running the experiment: your dashboard, your daily reminders, the apparatus that holds you to it.
$18
Yours, back
Your own money, held and returned the instant your seventh day is checked in. Not a reward — a refund of a bet you won.
And the unpleasant sentence, stated plainly: if the ten days end before your seventh check-in, the $18 is not returned. That possibility is the whole reason the mechanism works.
Begin the experiment — $25One payment via Stripe. Net cost if you finish: seven dollars.
Seven days of action inside a ten-day window, which means three days of built-in grace. Miss a Tuesday and nothing breaks. These are suggestions, not assignments — but in this order, they tend to build.
The name that arrived in the first ten seconds of this page. You'll rehearse it three times and then send nine words.
The person you've been meaning to thank since roughly 2019. Lead with the delay: "This is years late, and I still mean it."
Someone who does something for you so reliably that you've stopped noticing. Name the specific thing, not the general fact of them.
A teacher, a coach, an old colleague, a friend two cities away. This is the one that tends to make somebody sit down.
The person you live with, who receives your leftovers and your logistics and rarely your gratitude.
Someone you barely know. A neighbour, a barista, a nurse, the person who fixed something. Low stakes, startling effect.
The name you keep flinching away from. Send it anyway. This is where the experiment actually happens.
Ten days from now, seven people know something they don't know today. That isn't a mood. That's a result.
Your private dashboard asks exactly one question: did you express genuine appreciation to someone today? You tap yes. You may add a first name if you want a record of it later.
You are never asked what you said. You never have to show it to anyone. Each day we send one short email nudge — not a newsletter, not a lecture, just a hand on the shoulder.
Your experiment
Day 5 of 10
4/7
expressed
Did you express genuine appreciation to someone today?
Optional: who did you appreciate?
So far
Illustration of the participant dashboard. Once you save a day, the next qualifying check-in unlocks the following calendar day.
In a randomized trial of 958 adults doing a daily gratitude practice for one week — almost exactly this design — the people who wrote to specific individuals showed the strongest gains in gratitude and positive emotion. The people who merely listed things they were thankful for did no better than the control group.
Gratitude lists did not differ from controls on any outcome. Letters to particular people produced the strongest effects.
Regan, Walsh & Lyubomirsky (2023), Affective ScienceWhat we are not claiming
Nobody has run a controlled trial on this specific programme, and we won't pretend otherwise. What has been studied are the parts: that expressing gratitude to specific people lifts well-being more than generic gratitude journaling, that recipients are affected considerably more than senders predict, and that money placed at risk makes follow-through far more likely.
We assembled those three findings into ten days you can actually do. Whether it works on you is, appropriately, an open question. You're the one holding the instrument.
Sample size: n = 1. Investigator: you.

Finish the sentence, then hold onto it. On day ten, come back and find out whether you were right.
I predict that after telling seven people the truth about what they mean to me, I will feel…

The appreciation is already there. It has been there for years. The only thing missing is the sending — and ten days from now you'll either have done it or you'll be reading a page like this again.
Secure payment via Stripe · $18 returns on completion
Start any day in the next three
If you finish, you'll have spent seven dollars and a week of small courage — and seven people will carry a memory of you they didn't have before.
7 people. 7 days.
$18 returns when you finish