Thirty to forty percent of retail promotions lose money. Not break even. Lose money. That is the Boston Consulting Group's number, and bundles are one of the easiest ways to land in that bucket without noticing.

Here is the trap. A merchant builds a bundle, slaps on a 25% discount because it feels generous, and watches AOV go up while gross margin quietly goes down. The order looks bigger on the dashboard. The bank account doesn't agree.

Why most bundle pricing is guesswork

Ask ten Shopify merchants how they priced their last bundle and eight will say some version of 'it felt right.' That's not a strategy. That's a coin flip with extra steps.

Bundling done well is not a discount. It's a packaging decision. You are changing what the customer buys, not just what they pay. Get that distinction wrong and every bundle you build slowly bleeds margin, one order at a time.

A useful anchor: a 5% price cut needs an 18.7% jump in volume just to keep profit flat, according to margin-structure research widely cited in pricing literature. Most stores never check whether their bundle actually clears that bar. They just assume bigger basket means bigger profit. It doesn't, automatically.

The three bundle pricing models, and when each one works

There isn't one right way to price a bundle. There are three, and picking the wrong one for your catalog is the single most common mistake.

1. Fixed discount bundling

You take the sum of individual prices and knock off a flat percentage, usually 10 to 20%. Simple to build, simple to explain, easy for customers to evaluate against buying items separately. The risk: if your margin on the anchor product is already thin, a flat 20% off can push that specific bundle underwater.

2. Price point bundling

You set one clean number for the whole bundle, say $49, regardless of what the individual line items add up to. This works best when you're selling a 'kit' or 'starter set' story rather than a math problem. Customers stop comparing unit prices and start comparing the bundle to a mental benchmark of what a kit like this should cost.

3. Mixed bundling

You offer the bundle at a discount but also keep every item available individually at full price. This is the model with the best data behind it: mixed bundling generates 25 to 35% more revenue than pure bundling, because it captures both the shoppers who want convenience and the ones who only want one item.

A bundle discount is not a thank-you gift to the customer. It's the price of getting them to buy three things instead of one. If it doesn't earn its keep in incremental units, cut it.

The margin math nobody skips (but should check twice)

Before you publish a bundle, run this: total bundle price minus total cost of goods, divided by bundle price. That's your blended margin. Compare it to your normal single-item margin. If it drops more than 4 to 6 percentage points, the discount is too deep for that specific combination of products.

Here's a simplified example of how the same 20% headline discount plays out completely differently depending on what's actually in the box.

Same discount. Wildly different outcomes. This is exactly why a single '20% off bundles' rule applied store-wide is lazy pricing, and lazy pricing shows up as a margin problem three months later, usually right when you're trying to explain Q3 numbers.

What the data actually says about bundle discounts

Strategic bundling drives roughly 20% sales growth and 30% profit improvement when priced correctly, per McKinsey research on retail bundling. Notice the order there. Sales growth first, profit improvement second, and only 'when priced correctly.' That qualifier is doing all the work.

On average order value, brands running structured bundle programs report 20 to 35% AOV gains. Some real numbers from smaller DTC brands: one skincare label added 20% to AOV with a three-product bundle, a pet food brand saw 15% growth from a subscription bundle. These aren't outliers. They're the median outcome when the discount is calibrated to the actual margin structure instead of picked because it sounded good in a Slack message.

Anchor pricing matters too. Apple One's family tier saves subscribers roughly 30%, its premier tier closer to 43%. Both numbers are big enough to feel like a deal and small enough that Apple isn't giving away the store. That's the target zone for most product bundles: enough discount to change behavior, not so much that you're subsidizing customers who would have paid full price anyway.

Three pricing mistakes that quietly wreck margin

Discounting the anchor product instead of the accessory. If your bundle pairs a hero product with a low-cost add-on, apply the discount to the add-on's list price, not the hero's. The hero is what sells the bundle. Discounting it teaches customers your flagship product is negotiable.

Ignoring shipping and fulfillment cost per bundle. A bundle that ships in three boxes instead of one just ate part of your margin before a single discount was applied. Price the logistics reality, not the theoretical one.

Never revisiting bundle prices after a cost increase. Suppliers raise prices. Freight goes up. A bundle margin that worked in January can be gone by June if nobody rechecks it. Set a quarterly reminder. It takes ten minutes and it's the cheapest insurance you'll buy all year.

When bundling is the wrong move entirely

Not every catalog should bundle. If your products already sell at a 60%+ margin and move fast on their own, a bundle adds complexity without adding much. You're solving a problem you don't have.

Bundling earns its place when you're sitting on slow-moving inventory, when your AOV is stuck below your shipping break-even point, or when customers keep buying two specific products together anyway and you haven't packaged that behavior yet. Look at your own order data first. Don't copy a competitor's bundle just because it looked good on their homepage.

What to prioritize if you're starting from zero

1. Pick one bundle, not five. Test the model with your best-selling pair of products before building a whole bundle catalog.

2. Calculate blended margin before you calculate the discount. Work backward from the margin floor you're willing to accept, then figure out what percentage off gets you there.

3. Use mixed bundling, not pure bundling. Keep items purchasable individually. You lose nothing and gain the customers who don't want the bundle.

4. Set the discount between 10 and 20%. This range shows up again and again in the data as the sweet spot between 'feels like a deal' and 'still profitable.'

5. Review the numbers after 30 days, not six months. If a bundle isn't moving units at that margin, either raise the price slightly or replace the products in it.

Pick one bundle. Run the math first. Launch it. Check the margin in 30 days, not next quarter.

If you want to see how this plays out with a specific format, the mechanics of mix and match bundles and the psychology behind why bundling works on customers are worth reading next, along with how volume discounts solve a related but different problem: rewarding quantity instead of variety.