Ten years is ten years of decisions. Here are the ones we got wrong.
An anniversary page is usually a business congratulating itself and calling it an offer. There is an offer further down and it is an arbitrary number, which we will also say. First the part worth reading: what we changed our minds about, what we stopped selling, and what we are committing to for the next stretch.
Birthday offer ends in
We went quiet for most of a quarter — Volume outgrew the inbox and we did not say so. We lost customers who were entirely right to leave. Now there is a published reply time on the contact page, and when we cannot meet it the page says that instead of the target
We priced in a way that punished small orders — A minimum that made sense on a spreadsheet and made no sense to somebody buying one thing. It quietly told our smallest customers they were a nuisance. It is gone, and what replaced it is on the pricing page in full
We recommended something we did not use ourselves — Once, for a season, because the margin was good. Nothing was wrong with the product and that is not the point. We only recommend what is in the building now, and if you ask whether we use a thing you will get a straight answer
A line that was profitable and wrong for most people who bought it — It suited about one customer in five and the other four could have spent less on something better. A high return rate is usually a business telling itself something it does not want to hear
What we will not start selling — Name the thing customers ask for most that you have decided against, and give the reason. A refusal with a reason attached is worth more than a catalogue, and it stops the same email arriving every week
Run a countdown that resets — If a timer on any page of ours hits zero, the offer ends. A clock that restarts when you reload is a lie with a stopwatch on it
Invent a price we never charged — A "was" price on our site was the real price, for a real period, recently. That is a legal requirement in many places and a habit almost nowhere
Charge you to leave — No cancellation fee, no restocking charge on our own error, no admin fee for closing an account. Leaving should cost what arriving cost, which is nothing
Pay for a review — Not with money, not with a free product, not with an entry into a draw. A review we paid for tells you about our marketing budget rather than about the thing
Use a photograph of a product we do not sell — Including a nicer version of the same thing. It sounds small and it is the commonest quiet dishonesty in retail photography
A published reply time
One working day, and the contact page says so plainly when we are not meeting it rather than quietly leaving the target up.
The same list again next year
Including whatever we got wrong in the meantime. A business that publishes its mistakes once has done marketing; twice is a habit.
Broken commitments go on this page
Not a blog post nobody reads. If we miss one of these, it appears here with what happened, because this is where somebody would come to check.
10% for ten years
Everything, no minimum, one use per customer. It does not stack with anything else because nothing else is running.
Ends 8/27/2026
“They emailed to tell me a thing I had ordered twice was the wrong one for what I was doing, and refunded the second before it shipped. That was seven years ago and it is why I am still here.”
[Customer]
Since year three
“I complained about something properly badly once. They agreed with me in writing, which almost nobody does, and then actually changed it. The version on the site now is the one I asked for.”
[Customer]
Since year one
Suppliers who have been here the whole time
Read the mistakes before the discount
One of them is the reason the returns policy is what it is.
What we got wrongLive preview — every color, word and block is editable after you pick it.
