Your First FBA Shipment: How Many Units to Send (2026)

By Bernardo Campelo, Forbes Business Council E-Commerce Leader, Amazon SPN Certified prep provider, and Founder of PrepVia.

The most common question a new seller asks before the first shipment is also the one with the worst answers on the internet: how many units should I send? Forum wisdom says 100, because 100 is a round number. A YouTube guru says 500, because the factory gave him a price break at 500. Someone else says 25, to test the waters. All three are answering with a number when the question requires a formula.

I run PrepVia, an FBA prep center in Miami, Florida. We receive first shipments every week, and we watch the same two failure patterns play out from the receiving dock. Pattern one: the seller who sent 600 units of an unproven ASIN, is still paying monthly storage on 480 of them in month five, and is now researching removal orders. Pattern two: the seller who sent 40 units, sold out in nine days, and watched the ranking momentum he had just spent launch money building evaporate while his reorder sat on a boat.

Both mistakes were avoidable with basic arithmetic applied to honest assumptions. This post is that arithmetic: the formula, a full worked example, the sane range for a first shipment, what each direction of error costs at 2026 fee rates, and a three-phase plan that removes the guesswork from shipments two and three.

Start From Sell-Through, Not From a Number

Every magic number you will read in a seller forum was right for exactly one product: the one the person posting it sold. A $12 phone accessory that turns 8 units a day and a $95 kitchen appliance that turns 0.7 a day should not receive the same first shipment, yet "send 100 units" treats them identically. The only input that actually drives the answer is expected daily sell-through, and everything else in this post is built on it.

Estimating sell-through for a product with zero sales history is uncomfortable, but it is not guessing. Look at the review velocity of the competitors you plan to rank near, use a sales estimator on their listings, and check how many sellers split the demand on the exact ASIN if you are doing arbitrage or wholesale. That gives you a number for an established listing.

The Formula, With a Full Worked Example

First shipment quantity = estimated daily sales x (restock lead time + buffer).

The lead time is not your supplier's production quote. It is every step between "I need more units at Amazon" and "Amazon shows them as available." Here is what that chain looks like for a typical domestic reorder, with realistic 2026 durations.

For a domestic wholesale product, that chain totals roughly 20 to 36 days. Call it 29 for the example. Add a buffer for the things that go wrong, because something always does: a supplier stockout, a carrier delay, a slow FC check-in week. A 30 percent buffer on 29 days is about 9 days, giving you 38 days of coverage to buy.

Now the whole calculation, start to finish, for a product whose established competitors sell about 8 to 10 units a day:

Round to your case pack. If the supplier packs 24 per case, 120 units is five cases and the math is done. The rounding direction matters less than the fact that you arrived at the number from your own lead time and your own discounted forecast, not from a stranger's product.

Run the same formula across different forecasts and you can see how the answer moves:

Why 50 to 200 Units Is the Sane Range for a New ASIN

The formula will occasionally spit out a number outside 50 to 200. The range exists to catch it. A first shipment has a second job the formula does not capture: it is an experiment, and experiments have a right size.

Below 50 units, the experiment cannot produce a clean result. If the launch works even moderately well, you sell out before the data means anything, and a nine-day stockout during launch damages exactly the ranking signals you were paying to build. You also pay the fixed costs of a shipment, the setup, the boxes, the transit, over so few units that your per-unit logistics cost balloons.

Above 200 units, you are compounding forecast error. A forecast for a brand-new ASIN is routinely wrong by a factor of two in either direction. Being wrong by 2x on 120 units means 60 extra units, roughly two months of extra coverage, absorbed painlessly. Being wrong by 2x on 500 units means 250 units aging toward the 181-day surcharge while your capital sits in a fulfillment center instead of your next purchase order. The cash flow math of FBA punishes trapped inventory harder than most new sellers expect, because the money you locked into units 201 through 500 is the same money you needed for the reorder of the product that actually worked.

The Cost of Missing High vs Missing Low

Both directions of error have a price tag. They are not symmetric, and knowing which side hurts more should shape how you round.

Read the ranking row twice, because it is the asymmetry that matters. Storage fees are a bill: annoying, quantifiable, survivable. A stockout during launch is a reset. Amazon's algorithm reads a new listing that stops selling as a listing that stopped deserving traffic, and the reviews, ranking, and conversion history you were accumulating stop compounding. You can buy your way out of extra storage. Buying your way back to launch momentum costs far more than the storage ever would have.

The deeper fee math on both sides, storage bands, the 28-day threshold mechanics, and how days of supply is actually calculated, is in our low-inventory-level fee explainer.

The Three-Phase Test: Validation, Traction, Scale

The first shipment is not a one-off decision. It is phase one of a sequence, and naming the phases keeps you from making phase-three bets with phase-one information.

Phase 1: Validation (shipment one, 50 to 200 units)

The goal is not profit. The goal is data: does the product convert, at what price, at what daily velocity, with what return rate? Keep the shipment inside the sane range, watch sessions and conversion in your reports daily, and resist the urge to judge anything before two full weeks of sales.

Phase 2: Traction (shipments two and three)

Now you have the number the first shipment could not give you: real daily velocity. Rerun the same formula with actual sales instead of the discounted estimate, and reorder to hold 30 to 60 days of supply. This is where most of the risk leaves the system, because you are no longer forecasting, you are measuring.

Phase 3: Scale

Once velocity is stable across two reorder cycles, the question changes from "how many" to "how cheaply." Case pack optimization, freight consolidation, the SPD versus LTL decision, and placement fee strategy start to matter, because at volume they move your landed cost per unit more than any supplier negotiation.

Placement Fees When the Shipment Is Small

One 2026 fee catches first-time senders by surprise at shipment creation: the inbound placement service fee. When you build the shipment, Amazon offers split options. Accept the Amazon-optimized split, sending your units to five or more fulfillment centers, and the fee is $0. Consolidate to fewer destinations and you pay a per-unit fee that depends on the size tier and the shipping weight of each unit.

Amazon rebuilt this fee table on January 15, 2026, and most of the numbers still circulating in seller forums are two revisions old. Here is the current structure. The minimal-split column is the expensive end, the one you pay when you consolidate everything into a single destination.

Every band is a range on purpose, because the exact amount moves with the destination region. The real number for your shipment appears in Send to Amazon before you confirm it, and Amazon bills it about 45 days after the units are received, on the quantity actually received.

On a 120-unit first shipment of a small standard item, the minimal split costs $16.80 to $38.40. That is 120 x $0.14 at the low end and 120 x $0.32 at the high end. Push the same 120 units into the heaviest large standard band and the fee runs $66 to $228 instead, because 120 x $0.55 and 120 x $1.90 are different animals. The small-shipment math is also different from the math at scale. The optimized split can scatter 120 units into five or more part-filled boxes headed to different states. That raises your per-box shipping cost. It also multiplies the number of check-ins you are waiting on. Sometimes paying the minimal-split fee to send one tight shipment to one FC is genuinely cheaper end to end; sometimes the free option wins. Run both numbers on the shipment creation screen before you click. The full decision framework, including when consolidation beats the fee at every volume tier, is in our guide to avoiding Amazon inbound placement fees.

While you are there, note the check-in timing difference too: a single-destination shipment lives or dies on one FC's receiving speed, and in busy months that spread is wide. We track what receiving actually looks like in Amazon FBA check-in times for 2026.

The First Shipment Checklist, From Zero

Frequently Asked Questions

How many units should I send in my first FBA shipment?

For most new ASINs, between 50 and 200 units. The precise number is your estimated daily sales multiplied by your restock lead time plus a buffer, so a product forecast at 3 units per day with a 38 day restock cycle needs about 115 units. Start from that math, not from a round number someone posted in a forum.

What happens if my first FBA shipment sells out too fast?

You lose sales momentum exactly when Amazon is deciding whether your listing deserves visibility, and the algorithm has a short memory for new products. A stockout during launch resets much of your ranking progress and typically takes weeks to recover from. That is why the formula includes a buffer, and why you should trigger a reorder the moment real velocity beats your forecast.

How much does it cost to send 100 units to FBA?

For a typical small standard-size product, expect roughly 40 to 100 dollars for prep and labeling at a prep center, 15 to 40 dollars in partnered carrier shipping for 2 to 4 boxes, and 0 to 32 dollars in inbound placement fees depending on the split option you choose, because a small standard unit runs 0.14 to 0.32 on the minimal split and nothing at all on the optimized split. That puts total logistics at roughly 55 to 172 dollars, or about 0.55 to 1.72 per unit. Product cost and freight to the prep center are separate line items.

Should beginners use a prep center for the first shipment?

If the shipment is small and you have the time, prepping at home teaches you Amazon requirements firsthand, and that knowledge is worth having. Most sellers switch to a prep center once volume passes 100 to 200 units per month, because rejected shipments and labeling mistakes cost more than prep fees. A good center also catches compliance problems like missing poly bag warnings or expiration labels before Amazon does.