The Pricing Framework for a New SKU in an Existing Line
This comes from a specific job: a $2M DTC brand I ran wanted to launch a second SKU. Same customer, same channel, new product. The team wanted to "test a few price points" and see what stuck. That's not a pricing strategy. That's a guess with a spreadsheet attached. If you're figuring out how to price a new product line for a DTC brand, the answer starts before you touch a calculator.
What the problem actually looks like
You've got an existing hero SKU. It sells at $48. Margins are healthy. Now you're adding a second product, maybe a bigger size, a bundle, a new flavor, a premium version.
The team debates price for two weeks. Someone says "match the flagship." Someone says "price it higher, it's premium." Someone pulls up three competitor sites and averages their prices. Nobody looks at contribution margin. Nobody checks what it does to average order value. The launch date arrives and the price gets set by whoever argued loudest in the Slack thread.
Three months later, the new SKU is either cannibalizing the hero product or sitting flat because nobody understood why it costs more.
Why it happens
Pricing a new SKU inside an existing line is harder than pricing your first product, not easier. With your first product, you had no anchor. Now you have one, and every customer who's bought from you already has a mental price map of your brand. A new SKU either fits that map or it fights it.
Most teams treat the new SKU as an isolated decision. They cost it, mark it up, done. But price isn't just about covering cost plus margin. It's about where the SKU sits relative to the rest of the line, what job it does for the customer, and what it does to your blended economics across the whole catalog.
The other failure mode: teams import a price from a competitor without importing the competitor's cost structure, fulfillment model, or margin targets. A competitor priced at $65 might have 20% better manufacturing costs than you. Copying their number means copying their margin without their advantage.
What we do about it
Here's the actual process I run, step by step.
- Start with landed cost, fully loaded. Product cost, packaging, freight, duties, pick and pack, payment processing. Not just COGS. I've seen teams forget payment processing fees (2.9% plus $0.30 adds up fast at volume) and wonder why margin doesn't match projections.
- Set a minimum margin floor before you think about market price. On the DTC brand I scaled from $100K to $3M, our floor was 65% gross margin on any new SKU. Below that, paid acquisition math doesn't work once you factor in a $22-28 CAC on cold traffic.
- Map the new SKU against the existing line, not against the market. Draw a simple ladder: cheapest SKU, hero SKU, new SKU, most expensive SKU. Ask what job each one does. If your hero is $48 and the new SKU is a bigger size, price it using a per-unit discount curve, not a flat markup. Customers do this math in their head even when they don't do it on paper. A 2x size for 1.7x the price reads as a deal. A 2x size for 2x the price reads as no reason to upgrade.
- Model cannibalization honestly. Before launch, forecast what percentage of hero SKU buyers will shift to the new SKU instead of buying both. On that same brand, we assumed 15% cannibalization on a new bundle SKU. Actual was 22%. We caught it in week three by tracking SKU-level AOV, not just blended AOV, and adjusted the bundle discount from 18% off to 12% off.
- Price test with real money, not surveys. Survey pricing (Van Westendorp, willingness-to-pay questions) tells you what people say. A live A/B test on price, run through Shopify with two audience segments at $200-300 ad spend each, tells you what people do. Run it for at least 500 sessions per variant before reading results.
- Set the price, then protect it. Once live, don't discount the new SKU in the first 60 days unless data says the price is genuinely wrong. Early discounting trains your list to wait for a deal, and you'll never get a clean read on true demand.
The most common mistake
Pricing the new SKU based on cost-plus math alone, with no reference to the existing line's psychology. I've watched teams calculate a perfectly reasonable margin, land on a number like $54, and never ask what $54 does sitting next to a $48 hero product. Six dollars might not clear the mental bar for "clearly better" or "clearly bigger." Customers don't compare your SKU to your spreadsheet. They compare it to the other thing on your site.
What it costs to ignore
On the infrastructure project I marketed, a $2.2B build, pricing mistakes get buried in massive contracts and nobody notices for years. DTC doesn't give you that cover. Bad SKU pricing shows up in 30 days, in your Shopify analytics, in plain numbers.
Underprice a new SKU and you train customers to expect that price forever. Raising it later costs you reviews, refund requests, and social posts calling you out. I've seen a brand eat a 40% spike in support tickets after a "quiet" price increase on a SKU that launched too cheap.
Overprice it against the existing line's psychology and the SKU just doesn't move. You'll see traffic to the product page, add-to-cart rates in line with your average, and then a checkout abandonment rate 15-20 points higher than your other products. That gap is the market telling you the number is wrong. Most teams read it as a traffic problem and throw more ad spend at it instead of fixing the price.
Get the cannibalization math wrong and your blended margin drops while topline revenue looks fine on a dashboard. That's the dangerous one. Revenue growing while margin quietly erodes can run for two full quarters before finance catches it.
FAQ
Should a new SKU always be priced higher than the existing hero product?
No. It depends on the job the SKU does. A smaller, entry-level SKU should price lower to capture price-sensitive buyers who wouldn't try the hero product cold. A premium version should price higher, but only if it delivers a visibly bigger, better, or different outcome. Price ladders work in both directions.
How long should I run a price test before committing?
At minimum, enough sessions to get statistical confidence, usually 500+ per variant, and enough time to cover a full purchase cycle for your product, typically 2-4 weeks for most DTC categories. Shorter than that and you're reading noise, not signal.
The real takeaway: price the new SKU against your own line first, the market second, and your cost sheet third. Cost tells you the floor. The market tells you the ceiling. Your existing customers tell you what number actually makes sense in between.