Your Amazon Pack Size Should Not Be Your Retail Pack Size

Most brands list the pack they already sell. The single bag from the grocery shelf, the six-count from the club store, whatever the website's top seller is. It is the cheapest decision available, and on Amazon it is often the one that keeps a brand from earning a margin.

We say the same thing on nearly every audit call: you're probably not going to sell in the same pack size on Amazon that you do in retail or on your own site. Amazon charges by the unit that ships, so the pack you choose sets the fees, the ad cost per order and the price you can hold. Build the pack for the channel's economics, which usually means an Amazon multipack, and a sub-30% product can move above the floor without a cost cut.

Why the retail pack fails on Amazon

A retail pack is designed for a shelf, a price point and a case count. A shopper in an aisle picks it up, compares it to the one next to it and pays a few dollars. Amazon's costs don't care about the shelf.

Several of them scale with the number of shipments, not with the value of what is inside:

  • Fulfillment. The fee is set by size and weight tier. A single light item and a three-pack built as one sealed unit often fall in the same or adjacent tiers, so the per-unit fee drops as the count rises.
  • Advertising. A click costs the same whether the order has one unit or three. A larger order spreads the ad cost across more product. If your ad cost per order is a fixed amount, a bigger basket lowers the share of revenue it takes.
  • Everything fixed per shipment. The fixed pieces (fulfillment, the ad click, inbound handling) shrink as a share of a bigger sale. The referral fee doesn't help you here, because it's a percentage of the price, and in food it can step up as the price rises.

For low-ticket food, this matters more than almost anything else on the listing. A single $6 bag leaves very little after Amazon's fixed charges. A three-pack at an appropriate price leaves much more per shipment.

The math, on a hypothetical SKU

Take an illustrative product with a unit price of $8 and landed cost of $2.50. Use rounded fees for the example, not a quote, because your category and size tier will differ.

Single 3-pack
Price $8.00 $22.00
Landed cost $2.50 $7.50
Referral fee (15%) $1.20 $3.30
Fulfillment (illustrative) $3.50 $4.25
Left before ads $0.80 $6.95
Pre-ad gross margin 10% 32%

The single sits in the red zone. The three-pack clears the 30% floor with the same product and the same cost per unit. The only thing that changed was how many units ship together. Pull your own referral rate and fulfillment fee for your category, because both change by product and price tier, and run this table for your top five SKUs.

Two things that table leaves out will decide whether the pack works for you.

The cost of building the pack. The $7.50 above is three units and nothing else. A carton, a shrink band, a label and somebody's labor all belong in landed cost. On this example the floor sits at $6.60 left per pack, so the three-pack has $0.35 of room. Add $0.75 of assembly and it falls to 28%, under the floor. If your product already ships in a case of 12, packs of 3, 6 and 12 are easy pulls, and anything that breaks that case adds labor. We ask brands to estimate unit volume on each new pack before they commit, because that number sizes the labor.

The referral line in food. The table uses one flat rate to keep the arithmetic readable. Amazon's published fee table sets a lower rate for Grocery & Gourmet items at or below a price line and a higher one above it, and the rate applies to the total sales price. Beauty, Health and Personal Care works the same way, at its own line. So a pack that crosses the line pays the higher rate on every dollar. Say the rate stepped up seven points, purely for illustration: a $16 pack would pay $1.12 more than it would at the lower rate. Look up the line for your category before you set the price, and either stay under it or clear it by enough to cover the step.

What is an Amazon multipack, and how is it different from a bundle?

An Amazon multipack is two or more units of the same product sold as one item, and a bundle is two or more different products sold together. Amazon's product bundling policy draws that line directly: bundles "don't include multi-packs, which consist of two or more identical products." Amazon's imaging standards add a third name, the variety multipack, for multiples of different products packaged together. That's the assorted flavor box most food brands would build.

The distinction has practical consequences. Amazon's UPC guidelines treat a single and a pack as separate sellable units, each needing its own product identifier, and in most categories a multipack needs its own UPC plus an item package quantity. A physically packaged bundle takes the referral fee of its primary product's category, must be accepted as a single unit for returns, and still falls under the fair pricing policy.

How many units should an Amazon multipack hold?

Hold the smallest count that clears your margin floor, then move up only if the extra dollars per order survive the drop in conversion. Margin percentage stops improving quickly, while dollars per order keep growing, so the two numbers answer different questions. This table runs the same illustrative SKU with an assumed ad cost of $4 per order, at the same flat referral rate.

Single 2-pack 3-pack 4-pack 6-pack
Price $8.00 $15.00 $22.00 $28.00 $39.00
Price per unit $8.00 $7.50 $7.33 $7.00 $6.50
Landed cost $2.50 $5.00 $7.50 $10.00 $15.00
Referral fee (15%) $1.20 $2.25 $3.30 $4.20 $5.85
Fulfillment (illustrative) $3.50 $3.90 $4.25 $4.75 $6.00
Left before ads $0.80 $3.85 $6.95 $9.05 $12.15
Pre-ad gross margin 10% 26% 32% 32% 31%
Left after $4 of ads -$3.20 -$0.15 $2.95 $5.05 $8.15

The two-pack misses the floor and loses money once ads are paid for. The three-pack is the first count that clears, and past it the margin barely moves while the dollars keep climbing.

Those last rows assume every pack costs the same $4 to sell, and it won't, because a $28 order converts worse than a $22 order. So work out how far conversion can fall before the bigger pack stops winning. If a click costs $1 and the three-pack converts at 25%, that's the $4. The four-pack leaves $5.05 at that ad cost against $2.95, so it can absorb up to $6.10 of ad cost per order and still match the three-pack. At a $1 click that's a conversion rate of about 16%. Above 16% the four-pack earns more per order, and below it the three-pack does. Run that once with your own click cost, then check the answer against what each count does once it's live. Our piece on break-even ROAS and ACoS has the rest of the ad math per SKU.

What the right pack depends on

There's no universal number. The pack that works is the one that gets the SKU above its margin floor while staying at a price a shopper will pay in a single click.

Consumables with repeat purchase. Counts of two to four, or a case size, tend to fit. A shopper who has already decided they like the product is happy to stock up, and a stock-up count is a natural fit for Subscribe & Save.

Gifting and variety. A sampler or assorted pack gets its own price and avoids head-to-head comparison with the single flavors. Variety packs also give the brand a product that retail doesn't carry. When we took over a natural snack brand's account, it was built almost entirely on one-pack listings, and launching multipacks and variety 3- and 6-packs on its best sellers was part of the fix.

Heavy or bulky items. Weight-based tiers punish these hardest. Sell the denser size or the case, and keep the single for retail where freight is shared with a pallet.

Launch. Start with the pack that clears the margin floor, not with the one that lets the price look lowest. Cutting price later is easy. Raising it is not.

Should you keep Amazon's virtual multipacks?

Keep a virtual multipack only if the single already clears your margin floor, because it adds no fulfillment savings: Amazon charges the single-pack fee for every unit in it. A virtual multipack is a listing Amazon can create for you on selected FBA brand-owned ASINs. It sells two or more of your singles as one offer, picks them from your existing single-pack stock and needs no new packaging. Amazon's help page is explicit about the fee: "if a customer orders a 2-pack VMP, then you will be charged 2 * the single-pack fulfillment fee."

That fee is where a virtual pack parts ways with one you build yourself. This table puts the two side by side on the illustrative SKU. Amazon prices a virtual pack at the single price times the count by default, and you can add a discount.

Built 3-pack Virtual 3-pack, default price Virtual 3-pack, matched to $22
Price $22.00 $24.00 $22.00
Landed cost $7.50 $7.50 $7.50
Referral fee (15%) $3.30 $3.60 $3.30
Fulfillment (illustrative) $4.25 $10.50 $10.50
Left before ads $6.95 $2.40 $0.70
Pre-ad gross margin 32% 10% 3%

At its default price, the virtual pack earns exactly the single's 10%. It carries three times the dollars and spreads the ad cost across a bigger order, so it's better than selling one single, but it doesn't fix a thin margin the way a built pack does. Discount it to match the built pack and almost nothing is left. Sellers have posted the gap from their own accounts on Amazon's forums: one compared four units at $15.60 in fees as a virtual multipack with $5.77 for the same four as a prepped pack.

Amazon creates these in waves. Its latest batch was created by October 28, 2026, and went on sale to customers on November 11, 2026 unless the brand deleted it. Selection is automated and there's no way to opt in, so check rather than assume. A few things to look at while you're there:

  • Find them. Filter Manage All Inventory for VMP_.
  • Decide before you delete. A deleted virtual multipack can't be reactivated during the pilot, and the pilot is planned to run through 2027.
  • Watch the single's price. The virtual price follows the single, generally within 8 hours, so a price cut on the single reprices the pack too.
  • Watch the single's stock. The virtual pack draws from the same inventory, and it disappears from the detail page when the single hits zero. If you already sell a built pack at that count, ask why you'd want a second version paying three fees.
  • Supplement brands. Amazon warns that some product types, dietary supplements included, may get flagged as restricted.

Amazon also ran an invite-only fee discount on virtual multipacks, and its page says that pilot ended June 9, 2026. If a new one appears, rerun the table.

The trap on the other side: price parity

Choosing a different pack is not free from risk. Amazon watches prices across the web, and a product that is cheaper somewhere else can lose the Buy Box. We've written on how that plays out in the price parity trap. The relevant point for pack size is this.

Selling a three-pack on Amazon and a four-pack at Walmart used to keep the two apart. It doesn't anymore. Amazon's own selling partner documentation says that it might take the per-unit price of a 6-pack at an outside retailer and multiply it by 12 to set the competitive external price for a 12-pack in its store, "to adjust for the different pack sizes." Price above that number and Amazon can make the offer ineligible for the Featured Offer, which is what most sellers still call the Buy Box. Amazon's wording is "might," and it doesn't say how often it happens, so treat this as a live risk rather than something that hits every listing. What's no longer in doubt is that a count difference alone won't protect you.

The more durable approach is to make the Amazon product different in a way a per-unit calculation can't flatten: a size no retailer carries, a bundle of two flavors, a gift-ready count or a variant with a different ingredient. Those are separate items with separate listings. Some brands go as far as creating an Amazon-only size, for example a 14-ounce bag where retail has a 12. That comparison is harder to automate, and it gives you your own margin math. It's still no guarantee, because the same Amazon document says it can price against products customers might consider equivalent because of "near-identical attributes," so a small change to the size is a weaker defense than a different product.

How should you price a multipack against your own single?

Keep every multipack's per-unit price at or below your single's price on Amazon, and let the per-unit price fall as the count rises. Amazon's Marketplace Fair Pricing Policy lists "selling multiple units of a product for more per unit than that of a single unit of the same product" as a practice that harms customer trust, and the consequences it names run from losing the Featured Offer to removing the offer. Its Pricing Health page lists a further reference price: one derived from "the per-unit Featured Offer price for a smaller bundle of the same product." So the comparison runs inside your own variation family as well as across the web.

In the table above, the per-unit column steps down at every count: $8.00, $7.50, $7.33, $7.00, $6.50. That ladder is what you're protecting. It breaks more easily than brands expect. Drop the single to $6.99 for a sale and three of them cost $20.97, so a $22 three-pack now sits above its own single per unit. Check the pack prices before any price cut on a single, and remember that a reseller holding the Featured Offer on the single can set that number without asking you.

There's a practical ceiling on the outside too. If a retailer runs that same single at $6.49 and Amazon uses it, the matching three-pack price is $19.47. At that price the pack leaves $4.80 before ads, or 25%, and you're under the floor because of a promotion somebody else planned. Open Pricing Health and look at the reference prices on your packs, since that's where Amazon shows the competitive external price it's using.

Shoppers do this arithmetic as well. Amazon requires unit count on many product types and uses it to show a price per ounce, count or fluid ounce on the listing. Set the unit count correctly and name your sizes in one unit. You don't have to be cheapest per ounce. We've seen brands hold top category positions at a higher price per ounce than their competitors.

Which SKU should be Amazon-only, and will it cannibalize retail?

Make your best seller with the thinnest Amazon margin the Amazon-only pack, in a count no retailer or distributor carries. A best seller already has demand and reviews behind it, and that makes it the fastest place to test a new pack. A count nobody else sells gives you a pack no buyer can compare directly, though Amazon can still compare per unit, and it keeps your retail buyers from seeing the same item at a different number.

We don't have a tidy data set on cannibalization, and we'd be suspicious of anyone who says they do. What we can tell you is how we think about it. The shopper buying one bag in an aisle and the shopper buying four online for the pantry are usually on different errands, in the same way that a club pack and a shelf single live side by side. The bigger retail risk is the buyer who searches your brand and finds a lower per-unit price than the one on their shelf. A pack at $7.33 per unit against a $7.99 shelf single reads as a stock-up discount, and that's a defensible conversation. The same single at $6.99 on Amazon is a harder one.

So the working band for a multipack is narrow: at or below your Amazon single per unit, at or below the lowest per-unit price Amazon can find elsewhere, and high enough to clear your floor. If you can't find a price inside that band, the count is wrong or the product needs to be different. It's also worth getting your distributor's account list in writing, because a pack you think of as Amazon-only stops being one the moment a reseller or an online account picks it up.

Your own website usually gets more room, though not always. We tell brands they can price their site below Amazon if they want to own the customer, and that a lower price on another marketplace is where the Buy Box goes quickly.

Rules for the new SKU

  1. Give it its own UPC and listing. A multipack is a separate sellable unit to Amazon, with its own identifier and item package quantity. If it's the same product in a bigger count, add it to the single's variation family as a pack-size option. Reviews still pool across pack size and quantity variations, and that gives the new count the single's reviews from day one. A variety pack or a different product stands alone.
  2. Price it for the margin, then check the ladder. Your cost sheet sets the floor. The per-unit price sits under your single, and a shopper can see it on the listing.
  3. Keep the single too. The single is the entry point and the one shoppers search for. It earns its place as a trial purchase even when the margin is thin, as long as ads aren't pushed behind it. We'll often leave a few sub-30% SKUs in a catalog as trial drivers and get the average up to 30-40% with larger packs.
  4. Send the ads to the pack that makes money. If the single is the thin one, point your prospecting campaigns at the multi and let the single catch organic traffic.
  5. Build it so it survives inbound. Amazon's packaging rules require a set to be sealed so it can't be separated and labeled as a single unit, with wording such as "Sold as set" or "This is a set. Do not separate." Amazon will open a master case and sell what's inside as singles, as the FBA fee piece covers. Even stickered packs get split sometimes, and it's worth checking the count after check-in. Food has one more rule: the expiry date goes on the outer pack and on each item inside.
  6. Show the count. Amazon's imaging standards require the main image to show the total quantity delivered, in the packaging, and the title to state it as a pack or count. Shoppers buying the wrong quantity by mistake become returns and bad reviews.
  7. Check your reviews plan. A true multipack inherits reviews. A variety pack on its own listing starts with none, and the Vine program and a launch plan matter here. Amazon's Vine page excludes bundled products. Confirm your pack qualifies before you build the launch around it.

What it won't fix

A pack change helps when the product itself is sound and the cost structure is the problem. It does not rescue a product with a landed cost so high that no pack gets it above the floor, or a category where shoppers will only buy the single. In those cases the honest answer is to look at the cost side, or to decide the channel is not the right place for that SKU.

It also works best with room in your price. If your retail price is already above the category norm, a bigger pack makes the total look steep.

And it can hide a problem as easily as it solves one. Moving buyers into bigger packs raises revenue per order, and sales can look flat while units and new customers fall. We often see that on accounts that have already taken the pack and price lever as far as it goes. Track units and new-to-brand customers next to revenue, per SKU, after any pack change. A per-SKU Amazon profit and loss statement makes that easy to see.

What to do this week

  • List your top SKUs with price, landed cost, referral fee and fulfillment fee.
  • Compute pre-ad gross margin for each as a single and for 2, 3 and 4 counts, with assembly cost in landed cost.
  • Find the first count that clears 30% (or your floor).
  • Look at how competitors price that count in your category.
  • Filter Manage All Inventory for VMP_ and decide which virtual multipacks to keep.
  • Check the per-unit ladder against your single and against Pricing Health.
  • Build the listing as its own item with its own images.

The bottom line

The pack that works on a shelf was designed for a shelf. Run the margin for the same product at two, three and four counts, find the one that clears your floor, and give it its own listing. If your retail pack earns a thin margin on Amazon, the product probably isn't the problem.

If you'd like a second set of eyes on the numbers, send us your top ASINs with their landed costs, and we'll tell you which pack count clears your floor on each one and whether any virtual multipacks are costing you money.

cta_free_trial

Read more blogs

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System
  • Amazon

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System
Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround
  • Amazon
  • Case Study
  • Natural CPG

Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround

Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround
From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround
  • Case Study
  • Natural CPG
  • Amazon

From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround

From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround

Find out what your Amazon channel really earns

Apply for a free trial of our work, or chat with our team. Either way you'll leave knowing your true profit, which is more than your current reports can tell you. And if we work together and miss the targets we set, the sprint to fix it is on us.