There is a small drawer in the studio. It's the bottom drawer of an old wooden tea cabinet that I keep against the south wall, beside the bench. The cabinet itself is older than I am. The drawer was empty when I moved in. It isn't empty now.
Inside, organized by date in nine small linen pouches, are the nine pieces I made over the last three years and then decided not to ship. Each one has its archive number on the pouch. Each one has a small note clipped to the pouch in my handwriting, explaining what was wrong. I keep them not because I'm sentimental — I'm not, particularly — but because the drawer is the most honest record of our quality standard. Anyone could write a sentence on a website that says "every piece is reviewed before shipping." Almost every brand does. The drawer is the proof that the sentence has teeth.
This is the first time I've written about it. I want to open the drawer here, take out each pouch, and tell you what was wrong. Then I want to tell you what the drawer is actually for.
The rule
Before I open it, I should explain the rule. The rule has changed twice in three years; the current version, in effect since late 2025, is this:
Every piece sits on the photograph tray for fourteen days after it's finished, before it ships. During that time, I look at it at least three times. If, on any one of those looks, I would not be confident giving it to a friend, I do not ship it. I remake it from scratch, with the wearer's permission, or I refund and we begin again.
The fourteen days are important. The first look, the day after I finish, the piece always seems fine — partly because I'm proud of it, partly because the act of finishing is its own kind of approval. The second look, around day seven, I see things I didn't see the first time. The third look, on day fourteen, I am no longer the person who made it; I'm a person walking past it. That's the look that matters.
I ship a piece only if all three looks pass. Most do. Nine, over three years, have not.
Opening the drawer
I'll go in order.
#007 · December 2024. A rose quartz piece for a wearer in Vancouver. By day seven I'd noticed that one of the rose quartz beads had a hairline crack near the drill hole — invisible without strong sidelight, but it was there. I wrote to the wearer, explained, and asked permission to remake. She said yes. The replacement piece — also #007, same archive coordinate — shipped two weeks later. The original is in the drawer.
#023 · February 2025. A Her Tenderness for a mother's birthday gift. The composition was correct but on day fourteen I could see that the moonstone I'd chosen wasn't flashing — it was a fine moonstone in isolation, but next to those particular rose quartz beads, it disappeared. The piece would have looked like a normal pink bracelet without a heartbeat. I remade it with a different moonstone. The original — still beautiful — is in the drawer.
#031 · April 2025. A bracelet for myself, actually, that I made during a slow week and was going to keep. By day fourteen I realized I'd made it as a kind of self-portrait, and I didn't like what it said about how I was doing that month. I didn't ship it because I wasn't going to ship it. But I kept it in the drawer because removing it would mean I didn't have to look at what I'd made. I prefer to look.
#044 · August 2025. A Quiet Strength for a wearer who had specifically asked for "very dark." By day fourteen I felt the piece I'd made was darker than she'd actually wanted — I'd over-interpreted her instruction. I wrote to her, sent her two photographs (the one in the drawer, plus a remake at slightly less intensity), and asked which she preferred. She chose the remake. The over-dark version is in the drawer.
#051 · November 2025. A custom piece designed around the wearer's late mother's favorite stone — green aventurine. On day fourteen I noticed that one of the silver spacers I'd used had a slight burr along the edge — barely felt, would never have caught on fabric. I caught it with my fingernail by accident and immediately knew I wouldn't ship it. The piece was remade in two days. The original is in the drawer, with the offending spacer still in place, as a reminder.
#062 · January 2026. A First Light. On day fourteen the citrine I'd chosen had begun to look slightly more orange than it had on day one — I think because of where it had been sitting in the morning sun. It would not have looked wrong to most people. It looked wrong to me. I remade it with a paler citrine. The original is in the drawer.
#079 · May 2026. A custom piece for a wearer in Stockholm. The bracelet itself was correct, but the silk I'd strung it on had a tiny knot in the middle of the thread itself — a flaw in the silk, not in my knotting. The thread was structurally sound; the bracelet would have been fine to wear. But the knot wasn't mine, and I don't ship knots I didn't make. The bracelet was unstrung, the stones reused, the silk discarded, and the same piece restrung on a fresh length. The original silk (with the offending knot) is in the drawer, taped to a small card that just says not mine.
#102 · October 2026. A Slow Sunday for a wearer who'd asked for a piece that "looks like dawn." By day fourteen I could not honestly say that what I'd made looked like dawn. It looked like late morning. I'd been working too fast that week. I remade it slower. The original is in the drawer.
#118 · February 2027. The most recent. A Twilight Hour. By day fourteen I noticed that the amethyst I'd chosen had a small calcite vein running through it that I'd missed when stringing — visible only at certain angles. The piece would have read as flawed to anyone who held it in the right light. I remade it. The original is in the drawer.
Nine pieces over three years. Across roughly 1,400 pieces shipped in that span, that's a non-ship rate of around 0.6%. Each of those nine pieces was eventually remade and shipped; in seven of the nine cases the wearer never knew there had been an original. In two cases — #023 and #044 — I explained and asked permission. In neither case did the wearer ask to see the original. Why would they? They had the one they wanted.
What the drawer is for
The drawer is not for me, particularly. I know what's in it. I don't need the proof.
The drawer is, in a way, for you — for the person reading this who is trying to decide whether a small studio in Hainan, China, that you've never heard of and have no friends who've bought from, can be trusted with the kind of object that, if it goes wrong, will be the wrong kind of memory of a moment that mattered. The drawer is the only honest answer I can give to that question.
If I tell you "we hold every piece for fourteen days and refuse to ship anything that doesn't feel right," you have to take that on faith. If I tell you "there is a drawer in the studio with nine specific pieces in it that didn't pass, and here is what was wrong with each one," you have evidence. The first is a sentence. The second is a thing in a drawer.
I think evidence is what we owe wearers we'll never meet.
What it costs
I'll be honest about the cost, since this is the journal where I try to be honest about money.
A SENMOMO bracelet costs roughly $42–$58 in stones and materials, depending on the recipe. Labor — my time — adds another $30–40 per piece if I count it at a market rate, which I usually don't, but should. So the all-in cost of a piece I don't ship is around $80, plus the opportunity cost of the time I spent on it (an hour and twenty minutes, on average), plus the cost of making the replacement (another hour and twenty minutes).
Across nine pieces in three years, the drawer represents about $720 in materials I'll never recover, plus 24 hours of labor on pieces nobody is paying for. Less than $300 per year. Not much. The fact that it's not much is what makes the discipline easy. If non-shipping cost the studio $50,000 a year, I would be making different decisions. At $300 a year, the discipline is essentially free.
The harder cost, honestly, is the conversation I have to have with myself on the day I decide not to ship. There's always a voice that says: This is fine. It's better than what most studios ship as their best work. Just send it. That voice is not wrong, exactly. The piece would probably make its wearer happy. The wearer would never know what I'd seen.
But I would know. And the drawer is the place where the things I would know live, instead of being out in the world pretending I didn't see them.
What I do with the drawer
I leave it closed, mostly. I open it about twice a year — once at year-end, when I'm reviewing the work, and once whenever I'm tempted to lower the standard. The second visit is the more important one. Standing in front of an open drawer of nine pieces I once decided not to ship is the cheapest way I know to remember why I started the rule.
Eventually — at some point I haven't decided yet, probably when we hit piece #999 and Chapter One closes — I'll do something with the pieces in the drawer. They might go to wearers from the Founding 200 list, as gifts, with the full story attached. They might be donated to a museum if one ever asked. They might just stay in the drawer for as long as the studio exists. I genuinely don't know.
What I do know is that the drawer doesn't lose anything by holding them. They're protected, dry, and accounted for. And every time someone writes to me with the slow careful question of can I trust this with my mother's birthday gift, I have an answer that isn't a sentence. It's a thing.
Now you know about it, too.