Almost every price inside a growing consumer brand was set once, quickly, using landed cost times a multiple or a competitor's number minus a dollar. The problem is not that the number is wrong. It is that nobody wrote down what it was solving for, so nobody feels allowed to change it.
- Cost-up pricing sets a floor and nothing more. It prices your inefficiency, and it tells you to cut price when your unit costs improve.
- Value-down starts from what the customer buys instead of you. The working test is writing the product page at three candidate prices before choosing one.
- Your real ceiling is set by the substitute set, your own discount cadence, and whether you ever want a retail shelf. Two of those three are self-inflicted.
- On a $60 product, 5% on price adds 9.4% to contribution per order while a 5% conversion win adds 5%. A 15% storewide discount hands back 28%.
- Write one page per product family recording the price, the contribution at that price, the substitutes, the discount policy and the review date.
Ask an operator at a $30M brand how the price on their hero product got set and you usually get an archaeology answer. Someone took the landed cost in 2021, multiplied it by whatever number the category folklore said, looked at the closest competitor, took a dollar off, and typed it into Shopify. That was the decision. Nothing has revisited it since, and by now three people who were in the room have left.
The cost base underneath that number has moved four or five times. Freight moved, your returns rate moved, paid acquisition got more expensive. The price did not move, because at some point it stopped being a number and became part of the brand. That is the real pricing problem at this size, and it has very little to do with whether the number is right. The reasoning walked out of the building with the people who did it, so now the price gets protected instead of reviewed.
There are two honest ways to set a price. You can work up from what the thing costs you, or you can work down from what it is worth to the person buying it. Most brands do a rushed version of the first, tell investors a story about the second, and then defend the result for years.
Nobody decided
your price. Somebody
estimated it.
Two origin stories cover almost every price I have looked at inside a growing consumer brand. The first is cost times a multiple: landed cost of $14, times a number somewhere between two and four, rounded to something that looks retail. The second is competitor minus a dollar, which is really a bet that you and the brand you copied have similar cost structures and similar customers. Neither is stupid on day one. Both are indefensible by year three, because the reasoning was never written down.
The reason this hardens instead of getting revisited is that a public price starts to feel like a promise. Change it and you are not adjusting a variable, you are telling your customers something about yourself. So the number survives a tariff round, a freight spike, a packaging change and two rounds of ad-cost inflation, and the P&L absorbs all of it through gross margin instead.
"A price you cannot explain is a price you cannot defend, so you end up protecting it instead."
The fix is not complicated. It is a single page per product family recording what the price is, what it had to clear, and when you will look at it again. Section seven has the format. Everything before that is how to arrive at a number worth writing down.
Cost-up pricing
charges the customer
for your problems.
Cost-up is the default because it is the only method your ops team can compute without arguing. Take landed cost, apply a multiple that clears your target gross margin, round up. It guarantees a positive number on paper, it survives a board question, and it takes about four minutes.
It has two structural flaws that matter a lot more at $10M to $50M than they did at $2M.
The first is that it prices your inefficiency. If your landed cost is high because you are ordering at a small minimum, cost-up quietly asks your customer to fund your scale problem. Worse, it runs backwards when you fix it. Get your unit cost down 20% by moving to a real production run and cost-up logic says lower the price, which is the opposite of what a brand with improving margins should do with the room.
The second is that the multiple you inherited was calibrated for a different business. Keystone thinking, doubling cost at each step, came out of wholesale, where the retailer absorbed the cost of getting the product in front of a person. In DTC you pay that cost yourself, in media, and it does not appear anywhere in landed cost. A price that clears 68% gross margin can still lose money once shipping, payment fees, returns and acquisition are in the room. That is the whole reason contribution margin beats gross margin as an operating number, and why the real cost of a return belongs in the pricing conversation rather than in a separate ops review.
Cost-up is still useful. Use it as a floor, not as an answer. It tells you the price below which you should not be selling this product at all. It has nothing to say about the price above it.
Start from the thing
your customer buys
instead of you.
Value-down starts with a question cost-up never asks: what does this person do if you do not exist, and what does that cost them? A $34 candle is competing with an $11 grocery candle for some buyers and a $65 department store candle for others. Those are two different businesses with two different products, and the price is how you tell the customer which one you are.
The practical version at this size runs in three steps, and none of them require a research panel.
Pick the reference set on purpose. Write down the three things your customer is actually choosing between, with their real prices, from their point of view rather than yours. Most brands list competitors. Customers compare across categories. Somebody buying a $90 recovery product is often choosing between that and a massage, not between that and a rival brand.
Write the product page at each candidate price before you pick one. This is the test that saves people. Draft the hero copy, the guarantee, the photography brief and the shipping promise at $44, at $58 and at $72. If you cannot write a page you believe at $72, you cannot charge $72, and now you know exactly what would have to be true: better photography, a longer guarantee, a real material claim, a size that changes the per-use math. Price and page move together or the price fails.
Then read the arithmetic, not the reaction. Conversion rate is the wrong scoreboard for a price decision, because a higher price is supposed to convert worse. Section five is the math that tells you how much worse you can afford, and the conversion versus margin tipping point calculator runs it against your own numbers in about a minute.
Where you can genuinely test price on live traffic is narrower than most vendors imply, but it is not zero. New products, new sizes, new bundles and new geographies all launch without a price history attached, which makes them the cleanest place to learn what your customer will pay. Read those launches for contribution per order rather than conversion, and read them for long enough to see the refund rate settle.
Three things set
your ceiling, and
two are your fault.
Operators talk about price ceilings as though the market imposed them. Some of it is imposed, and most of it you built yourself, usually without deciding to.
| Anchor | What it does | Who set it |
|---|---|---|
The substitute set What the customer buys instead | Sets the range the price has to live inside to read as sane | The market, mostly |
Your discount habit Depth times frequency | Sets your real price. List price becomes decoration | You did |
The channel you want next Retail, wholesale, marketplace | Forecloses prices that cannot survive a wholesale margin stack | You did, in advance |
Discounting is the anchor operators underrate most. If you run 20% off six times a year and your customers have learned the calendar, your price is not on the tag. It is the last promo, and everything in between is a waiting room. Brands in that position often think they have a pricing problem when what they have is a promotional cadence that already answered the question for them.
The channel anchor is the one that takes options away before you know you want them. If a shelf is anywhere in your plan, your DTC price is functionally your MSRP, and it has to survive being cut in half on the way to a retailer. Price your DTC product where the wholesale math cannot work and you have chosen a single-channel business by default. The margin math on a first wholesale account is worth reading before you set a price you intend to keep, not after a buyer asks for your price list.
One more constraint that is genuinely external: your input costs are not stable and have not been for several years. What tariffs do to unit economics is the version of this that has hit most brands recently, and the brands that handled it best were the ones who already knew what their price was solving for.
A five percent
price rise beats a
five percent CRO win.
Take a $60 product with an ordinary DTC cost stack and compare three scenarios.
| Line | Baseline | Price +5% | 15% off |
|---|---|---|---|
Price collected | $60.00 | $63.00 | $51.00 |
Landed COGS | $18.00 | $18.00 | $18.00 |
Pick, pack, ship | $8.00 | $8.00 | $8.00 |
Payments at 2.9% plus 30c | $2.04 | $2.13 | $1.78 |
Returns and reship allowance at 3.5% | $2.10 | $2.21 | $1.79 |
Contribution per order | $29.86 | $32.66 | $21.43 |
Versus baseline | Baseline | +9.4% | -28.2% |
Read the middle column first. Five percent on the price adds 9.4% to contribution per order, because almost none of your variable costs move with price. Payments and the returns allowance scale a little. Freight, labor and COGS do not care what you charged.
Now the part that reorders your roadmap. A five percent conversion improvement adds five percent of contribution, because it produces five percent more orders at the same margin per order. The five percent price rise adds 9.4% on the same number of orders. Roughly twice the effect, and it ships in an afternoon instead of a quarter. The break-even is generous too: you can lose 8.6% of your orders at the higher price and still hold total contribution flat.
The right column is the same physics pointed at your foot. A 15% storewide discount hands back 28% of contribution per order. To break even on it you need 39% more orders, which almost no promo delivers. That gap is why the discount habit from section four is a pricing decision rather than a marketing one. It is the same tradeoff that sits underneath conversion rate versus profit margin and underneath choosing between AOV and conversion rate.
Run your own numbers before you argue with mine. The stack varies enormously by category, and unit economics by category shows how far apart apparel, supplements and hard goods sit on exactly these lines.
You do not have
a price. You have
a ladder.
Most brands at this size have one real price and three accidents. The trial size got priced to be cheap. The bundle got priced by whatever discount felt generous the week it launched. Subscribe and save is at 15% off because somebody picked 15% in 2021. Nobody has looked at the set together.
Look at it as a ladder and the hero price stops carrying all the weight. Your entry product exists to make the first purchase easy, so it should be priced for acquisition and judged on second-order rate rather than on its own margin. Your hero product carries the brand and should be priced where the page can defend it. Bundles are where you raise average order value without touching the number your loyal customers see, which matters because a bundle price increase is invisible in a way a hero price increase never is. The subscription discount is the one to audit hardest, since it is a permanent margin decision usually made once, casually, and it compounds across every future order.
A second size or format is a pricing tool disguised as a product decision. Adding a larger size at a better per-unit price moves your mix without repricing anything, and it reads as generosity rather than as an increase. That is the cheapest structural margin work available to most brands, and it almost always sits with the product team instead of with whoever owns the P&L.
Write the decision
down so it can
be changed later.
One page per product family, in a shared doc that somebody other than you can find. This does not need to be a model and it definitely does not need to be a deck.
Record the price and the date you set it. Note the landed cost that day and the contribution per order at that price, using the full stack from section five rather than gross margin. List the three substitutes your customer actually considers, with their prices. State the discount policy in plain terms, including the deepest promo you will run and how many times a year. Name the review date and the trigger that pulls the review forward, which for most brands is landed cost moving more than five percent.
That page does one useful thing. It converts "we have always charged $48" into "we chose $48 in March against these numbers." The first sentence is an identity and you cannot argue with it. The second is a decision, and decisions get revisited without anybody feeling attacked. Brands that keep this page raise prices earlier, discount less reflexively, and spend a lot less time relitigating the same question in leadership meetings.
Twice a year is the cadence I would hold you to, plus any time landed cost, freight or duty moves materially, plus any time you meaningfully change the product page or launch a new size. When the answer is that the price should move, the execution is its own discipline, and raising prices without losing customers covers the timing, sequencing and cohort work that decides whether it lands.
Your price is the fastest lever you own and the one you have touched least. Somebody guessed it once, under time pressure, with worse information than you have today. You do not owe that guess your loyalty.
Questions operators
ask about setting
prices.
Q: Should I price from cost or from what customers will pay?
Both, in that order, for different jobs. Cost-up tells you the floor: the price below which the product does not deserve shelf space in your own catalog, calculated on contribution rather than gross margin so shipping, payment fees and returns are in the number. Value-down tells you the price. Start from what your customer buys instead of you and what that costs them, pick the reference set you want to sit inside, then check that the page you would have to build at that price is a page you can actually build. If cost-up and value-down give you a range rather than a number, price at the top of the range you can defend on the product page and leave the discount alone.
Q: What gross margin does a DTC brand need?
There is no universal number, and the ones quoted at conferences usually come from categories with cheap freight. What I see across the consumer brands I have operated and advised is that a DTC-led P&L generally needs gross margin comfortably north of 60% to fund paid acquisition at any scale, and brands sitting under about 50% end up dependent on either wholesale volume or an unusually strong repeat rate to work at all. Treat those as bands rather than targets. The number that decides whether you can grow is contribution per order after shipping, payments and returns, because that is the money available to buy the next customer.
Q: How often should we revisit price?
Twice a year on a calendar, plus three triggers. Landed cost, freight or duty moving more than about five percent pulls the review forward. So does any material change to the product page, since price and page are one decision. So does launching a new size or format, which changes your whole ladder rather than one number. Most brands at $10M to $50M have gone two or three years without a real review, which is why the first one usually finds room.
Q: Can I test price on live traffic?
Within limits. Running two prices at once on the same product to the same audience creates support problems and, in some jurisdictions, compliance ones, and the honest answer is that most brands at this size cannot get a clean read anyway. What works is using the launches that arrive without a price history: new products, new sizes, new bundles and new geographies. Price those deliberately rather than by habit, then read contribution per order rather than conversion rate, and wait long enough for the refund rate to settle before you conclude anything.
How much conversion can your price rise afford to lose?
The tipping point calculator runs the arithmetic from section five against your own numbers: your price, your cost stack, your conversion rate. It tells you the order volume you can give up at a higher price and still come out ahead. If you would rather talk through the pricing decision itself, reach me at hello@taylorsicard.com.
Run the tipping point → Or start a conversation →