We sell concentrated cleaning sheets, the kind that replace a two-litre plastic bottle of detergent with something that fits in a drawer. There are two of us, we have no staff, and every dollar in the business is ours.
Our product is made overseas, shipped here by sea, and sits in a warehouse in New South Wales until somebody orders it.
That last part sounds like an operational detail. It is actually the thing that shapes almost every commercial decision we make, because cleaning products have a use-by date – and I had no idea that was the case until we were already committed.
It is less dramatic than it sounds. Nothing rots. Nothing separates or leaks or smells bad.
The active ingredients lose potency. In a detergent that shifts protein, starch and grease, much of the work is done by enzymes. They are biological catalysts rather than stable chemicals, and they get weaker over time. Fragrance usually turns before anything else does. Anything water-soluble also hates humidity, so a warm, damp warehouse ages stock faster than the label ever assumes.
The result is a product that looks perfectly fine and cleans slightly worse. No customer could pick it. No photo would show it.
The chemistry is not the deadline. The printed date is. That took me far too long to work out.
Retailers and customers go by what is on the box. It does not matter whether the product has genuinely faded, whether your warehouse is climate-controlled, or whether you have any evidence at all that the thing still works exactly as it did. Once that date is close, you have a hard sell-by date, and everyone in the chain treats it as absolute.
So you are not managing a quality problem. You are managing a printed number, and it does not negotiate.
Every founder learns to watch the runway, the months of cash left before the business runs out of money. You know that number, you know what burns it, and you can slow it down. Cut spend, delay a hire, take less yourself. The runway responds to your decisions.
The second clock does not. Stock ages at the same rate whether you sell it or not. Being disciplined buys you nothing. There is no lever that slows it and no restructure that makes it go away. It runs while you sleep and it runs while you are still working out your positioning.
I have watched a lot of founder talks. Nobody mentioned it.
Order quantity comes first. Overseas manufacturers set minimums, and the per-unit cost falls sharply as the order climbs. The spreadsheet says order more. Shelf life says order less. Neither answer is safe, so it becomes a bet, and you place it long before you have any real demand data to guide you.
Pricing is where the orthodox advice starts to creak. Hold your price, the thinking goes, and let the brand earn its premium. I agree with it. But it quietly assumes patience is free. When the stock has a date on it, waiting has a cost, and the nearer that date, the less freedom you have to be principled about anything.
Discounting is the ugly one. Discounting into demand is a strategy. Discounting into a deadline is something else, and buyers feel the difference even when they cannot name it. Every cut teaches your customers that the real price is lower than the label.
There is also a version that feels like a win and is not. You discount hard enough to clear the lot, then realise you have converted stock back into roughly the cash you paid for it.
Nearly every accelerator, book and founder interview tells you to go slowly at the start. Talk to customers. Iterate. Do not scale before you know what people actually want.
It is good advice. It also assumes the cost of waiting is close to zero, which holds if your product is software and does not hold if your product is on a pallet. Founders with a physical product get told to be patient while holding an asset that is quietly depreciating on a schedule set by chemistry and a date printed by a factory.
Those two instructions do not fit together, and nobody flags the collision.
What helped me was treating shelf life as a financing cost. It behaves like interest. It accrues whether or not you are doing anything useful, it compounds against your worst inventory rather than your best, and it belongs in the model as a line item rather than as a background worry.
Ask for the shelf life before you ask for the price. Not after the sample, not after the quote. First.
Then ask what drives it and under what storage conditions the number was reached. A two-year date on a wool dryer ball and a two-year date on an enzyme detergent are different kinds of promise, and a date reached in a stable lab means less in a shed in February.
Then work out how many units you can sell inside that window at your current rate rather than your hoped-for rate, and order against the honest number even though the unit cost will sting. The margin you give up buying small costs less than the margin you give up clearing stock later.
And build the date into your reporting from the start.
This article on the startup metrics every founder needs to know is a good reminder of that, but for anyone selling something physical I would add one more:
Weeks of cover measured against remaining shelf life. If the first number is larger than the second, you do not have an inventory position. You have a deadline
None of this is a reason to avoid building a product business. I would do it again, and I am glad we did. I would just like to have known I was signing up for two clocks instead of one, and that only one of them was ever going to listen to me.
- Bree Uebergang is the co-founder of Filthy Clean, an Australian eco-cleaning brand.
