Amazon FBA vs. FBM

Amazon FBA vs FBM is not a philosophy debate — it is a per-ASIN math problem, and sellers who answer it at the account level instead of the product level leave margin on the table in both directions. FBA (Fulfillment by Amazon) trades fees for the Prime badge and operational leverage. FBM (Fulfilled by Merchant) trades your own labor and shipping costs for control and, on the right products, materially better unit economics. Most brands doing $1M+ on Amazon should be running both. This guide works through the full cost stack on each side, what the Prime badge actually does to conversion, where capacity limits change the equation, Seller Fulfilled Prime as the middle path, and a decision framework based on size, velocity, and margin.

What FBA Actually Costs: The Full Fee Stack

FBA pricing looks simple on the rate card and gets expensive in the line items sellers forget to model. The referral fee (typically 8–15% of sale price depending on category) applies identically to both models, so set it aside. The FBA-specific stack:

Fulfillment fees. Charged per unit, banded by size tier and weight. A small standard-size unit under 4 oz runs in the low $3 range; large standard items climb through roughly $4–$7+; large bulky and extra-large items start north of $9 and scale with weight. Apparel carries a premium, and units priced under $10 get a discounted rate. The size tier boundaries are cliffs — a package that measures 0.1 inches into the next tier pays the next tier’s rate on every unit forever, which is why we dimension-audit every catalog we take over. Check current rates against the published schedule in our FBA fee reference.

Inbound placement service fees. Charged per unit when you ship to a single inbound location and let Amazon distribute inventory across its network — roughly $0.21–$0.68 for standard-size units depending on size and destination. You can avoid it by splitting shipments to multiple destinations yourself, which trades fee for freight complexity.

Monthly storage. Around $0.78 per cubic foot for standard-size from January through September, jumping to roughly $2.40 in October–December. Oversize rates differ, and low-inventory-level fees can hit standard-size products you chronically understock.

Aged inventory surcharges. Inventory sitting 181+ days accrues escalating surcharges that climb brutally past 271 and 365 days. Slow movers in FBA are a compounding tax, not a parking fee.

Returns. Amazon accepts returns generously on your behalf, and a meaningful share come back unsellable. High-return categories effectively pay a hidden 2–5% of revenue here.

Model all of it per ASIN before comparing. Amazon’s own revenue calculator and the fee preview report get you most of the way; our FBA fee calculator guide walks through the traps in both.

What FBM Actually Costs: The Honest Self-Fulfillment Number

Sellers chronically undercount the FBM side because half the costs do not arrive as invoices.

Shipping. Your negotiated carrier rates versus Amazon’s. This is the crux: Amazon’s fulfillment fee on a 1 lb standard item is often cheaper than any rate you can buy for a single residential parcel, because Amazon’s carrier leverage is unmatchable at small parcel sizes. At 20+ lbs, dimensional pricing, or freight-class items, the equation flips hard — Amazon’s large bulky fees frequently exceed what a decent 3PL or your own dock can ship for.

Storage and labor. Warehouse rent per cubic foot (typically far below Amazon’s Q4 rate), pick-and-pack labor at $1.50–$4.00 per order for most 3PLs, packaging materials, and WMS software.

Customer service. FBM means you own shipping-related customer contacts, returns processing, and every “where is my order” message within Amazon’s 24-hour response SLA. That is real headcount at volume.

Performance risk. FBM exposes you to Late Shipment Rate, Valid Tracking Rate, Cancel Rate, and on-time delivery thresholds — metrics FBA sellers never think about. A bad carrier week can put an FBM account into account health jeopardy, and that risk has a cost even when nothing goes wrong.

Amazon FBA vs FBM: The Economics Side by Side

The pattern in the table is consistent: FBA wins on small, fast, standard-size; FBM wins on big, slow, high-value, or fragile-margin products. Almost no serious catalog is 100% one or the other.

What the Prime Badge Does to Conversion

The Prime badge is the reason FBA survives its fee stack. Over 180 million Prime members filter or default to Prime-eligible offers, and the badge signals delivery speed, easy returns, and trust in a single glyph next to your price.

Sellers who have run the switch both directions typically report conversion lifts in the 20–60% range when a listing gains Prime, with the biggest swings on commodity products where a shopper has five equivalent choices and picks the one arriving Tuesday. On differentiated products with no close substitute, the penalty for FBM is smaller — the buyer wants your product specifically and will tolerate 4-day shipping to get it.

Run the sensitivity math before deciding anything: if FBM saves you $3.10 per unit in fulfillment cost but the listing converts at 9% instead of 14%, you did not save money — you traded velocity, and velocity feeds organic rank, which compounds. Conversion is also a ranking input, so the FBM discount shows up twice: once in the session-level CVR and again in slowly eroding keyword positions. This is why we default new competitive launches to FBA regardless of the fee math, then revisit once the product holds page-one rank.

Capacity Limits Change the FBA Math

FBA is not an infinite warehouse. Amazon assigns monthly capacity limits based on your IPI score, sales forecast, and its own network constraints — and every Q4, sellers discover their limit will not hold their Black Friday forecast. You can bid for additional capacity through the Capacity Manager (a reservation fee offset by performance credits as the extra space generates sales), but you cannot count on it.

Practical consequences: FBA-only sellers carry structural stockout risk they do not control, and slow-movers eat capacity that your best sellers need. This is where a hybrid model earns its keep — an FBM offer on the same ASIN keeps the listing live and the rank fed when FBA inventory runs dry, and overflow inventory sits in a 3PL instead of against your capacity limit. Managing that interplay between limits, restock timing, and offer switching is a core part of what disciplined inventory management on Amazon looks like.

Seller Fulfilled Prime: The Middle Path

Seller Fulfilled Prime (SFP) lets you display the Prime badge while shipping from your own warehouse or 3PL. You get the conversion lift and keep the control. The catch is that Amazon holds SFP sellers to standards most operations cannot meet: a trial period, on-time delivery around 93.5% or better, roughly 99% valid tracking, near-zero seller-initiated cancellations, nationwide one- and two-day delivery coverage on standard-size items, and weekend pickup or delivery capability. In practice you need either a genuinely excellent multi-node 3PL or your own distributed fulfillment network.

Where SFP shines: oversized products where FBA fees are punitive, high-value items you do not want commingled or handled loosely, regulated or fragile products needing controlled packaging, and catalogs with deep SKU counts where FBA storage would bleed you. Where it disappoints: ordinary standard-size goods, where meeting SFP’s delivery-speed bar costs more than FBA would have. SFP is a scalpel, not a default.

The Decision Framework: Size, Velocity, Margin

Run each ASIN through this tree:

1. Is it standard-size, under ~2 lbs, selling 10+ units/day? FBA. Amazon fulfills it cheaper than you can, the Prime badge feeds velocity, and velocity feeds rank. Add an FBM backup offer if stockouts have burned you before.
2. Is it large bulky or heavier than ~20 lbs? Price FBM/3PL first. Quote real freight-inclusive costs against FBA’s weight-scaled fee and the Q4 storage spike. If the FBM cost is 25%+ cheaper per unit, test FBM and measure the CVR hit — for many oversized categories, buyers already expect slower delivery.
3. Is it slow-moving (under 0.5 units/day) or long-tail? FBM, or a token FBA quantity with FBM depth behind it. Aged inventory surcharges make FBA a losing warehouse for anything that sits.
4. Is the margin thin (under ~20% net after referral fees)? The fulfillment delta decides viability. Model both; a $1.40/unit difference on a $22 product is the whole business.
5. Is it high-value, fragile, or regulated? Consider FBM or SFP for handling control, and to avoid commingling and return-abuse exposure.
6. Is it seasonal with a compressed selling window? FBA for the season (badge maximizes the window), with a hard plan to clear or remove stock before aged surcharges and off-season storage bite.

Then re-run the framework quarterly. Fee schedules change every year, your negotiated 3PL rates change, and a product’s velocity at launch is not its velocity at maturity.

A Worked Example

A garden tool: 24 × 8 × 6 inches, 6 lbs, sells for $49, referral fee $7.35, landed cost $14. FBA large standard fulfillment runs about $8.50 plus roughly $0.40/month storage and a $0.35 placement fee — call it $9.60 all-in, netting $18.05 per unit after landed cost and referral fee. A regional 3PL quotes $3.20 pick-pack plus $7.90 average shipping plus $0.45 storage — $11.55 all-in, netting $16.10 on the same basis. FBA nets $1.95 more per unit and adds the Prime badge. Same math on the 36-inch oversized version of that tool flips the answer by $6 per unit in FBM’s favor. That is the whole discipline: no rules of thumb survive contact with a specific ASIN’s dimensions.

How to Test a Switch Without Torching Your Rank

When the math says a product should move from FBA to FBM, do not flip it cold — conversion and rank effects are empirical, and the only trustworthy data is your own listing’s.

1. Add the FBM offer alongside the FBA offer first. Same ASIN, both offers yours. Nothing changes for the buyer while FBA has stock, and you validate your fulfillment pipeline on the trickle of orders FBM catches.
2. Let FBA sell down naturally, then hold FBM-only for 3–4 weeks. Track sessions, conversion rate, and keyword rank on your top 5 terms weekly (Helium 10 or Data Dive both track this cleanly). Sessions should hold roughly steady; conversion is what moves.
3. Do the P&L comparison on contribution margin per week, not per unit. If FBM’s per-unit gain is $2.10 but weekly units fall 30%, FBA was earning more total dollars.
4. Have the reversal ready. Keep a replenishment order staged so you can restore the FBA offer within two weeks if rank starts sliding — rank decay is much cheaper to stop early than to rebuild.

Run the test in a normal-demand month, never inside Q4, and never on your single biggest ASIN first.

Most Real Answers Are Hybrid

The FBA-versus-FBM framing implies a single winner, but the operating answer for most established brands is a portfolio: FBA on the fast standard-size core, FBM on oversized and long-tail SKUs, FBM backup offers on every A-item, and SFP only where the math demands the badge without the FBA fee structure. Getting there requires per-ASIN cost modeling, honest CVR testing, and someone watching the fee schedule and capacity limits as they move. That is exactly the kind of ongoing decision-making our Amazon account management service owns for clients — including the quarterly fulfillment-mix review that most in-house teams never get to. If nobody has re-run your FBA vs FBM math since the last fee change, start with the account management overview or have us run the numbers with you.