You can technically open an Amazon FBA business for a few hundred dollars. You can't run one on that. Between samples, a first production run, inbound freight, the 2026 fee structure, and a PPC launch budget that actually buys data, a realistic starting budget for a private-label product lands between $2,500 and $5,000 — and the sellers who fail usually didn't underfund the product, they underfunded everything around it. Here's the full 2026 cost picture, line by line, with a worked break-even example so you can test your own numbers before you spend a dollar.
The Startup Budget, Line by Line
| Item | Lean | Comfortable | Notes |
|---|---|---|---|
| Professional seller account | $39.99/mo | $39.99/mo | Fixed; individual accounts suit tiny test volumes only |
| Samples (2–3 suppliers) | $150 | $400 | Includes express shipping and the duds you'll reject |
| Product certifications / compliance | $0–200 | $300–800 | Category-dependent; children's products cost more |
| First production run (300–500 units) | $1,500 | $3,000 | At a $3–6 unit cost for typical standard-size goods |
| Freight to FBA (sea + last mile) | $300 | $700 | Sea freight for volume; air only for a tiny test batch |
| Photography & listing visuals | $150 | $500 | 7 images + infographics; do not skip this |
| Trademark (for Brand Registry) | $250–350 | $350–500 | USPTO filing via a provider; unlocks A+ and brand tools |
| PPC launch budget (first 60–90 days) | $500 | $1,500 | Buying rank and keyword data, not just sales |
| Contingency (10–15%) | $300 | $700 | Re-measures, remakes, inbound surprises |
Realistic total: about $2,500 lean, $8,000+ comfortable. The lean path works if you pick a simple, low-compliance product and do the design work yourself. The comfortable path exists because most first products need one remake and one re-order of creative — budgeting for that is why some sellers survive their first launch and others don't get a second one.
The Per-Sale Fees That Come Out of Every Order
Your startup budget gets you onto the field. The per-unit fees determine whether you can stay on it. For 2026 (rates effective January 15), a typical $24.99 standard-size product pays roughly:
| Fee | Typical amount | Basis |
|---|---|---|
| Referral fee | ~$3.75 | 15% of sale price (category-dependent) |
| FBA fulfillment fee | ~$4.00–5.50 | Size/weight tier; +$0.05 avg for large standard in 2026 |
| Inbound placement (minimal split) | ~$0.30–0.60 | Per unit; higher if you don't split shipments |
| Monthly storage | ~$0.05–0.15 | $0.78/cu ft Jan–Sep; $2.40/cu ft Oct–Dec |
Two 2026 notes worth knowing before your first PO: the average fulfillment fee increase this year was just $0.08 per unit (the first adjustment after a flat 2025), and if your product sells under $10, the Low-Price FBA program discount deepened to an average of $0.86 per unit — the low-ticket economics actually improved. The complete fee picture, marketplace by marketplace, is in our FBA Fees 2026 Breakdown.
Worked Example: Your Break-Even on a $24.99 Product
Say you land on these numbers: unit cost $4.50, freight-in $0.60/unit, sell price $24.99, referral $3.75, FBA fulfillment $4.80, inbound placement $0.45.
- Revenue per unit: $24.99
- Total per-unit costs: $4.50 + $0.60 + $3.75 + $4.80 + $0.45 = $14.10
- Pre-ad contribution: $10.89 per unit (~43.6% margin)
Now the honest part: your PPC launch will burn $500–1,500, and listing visuals and samples (another several hundred) don't ride on any single unit. At $10.89 of contribution per sale, you need roughly 90–140 sales just to cover the pre-launch spend — before you've made a dollar of profit. Most new sellers' first reorder decision, not their first sale, is where the business actually starts working.
Why 300–500 units for the first run: enough to survive a 60-day PPC learning curve without a stockout (which triggers the low-inventory-level fee and kills your momentum), but small enough that a failed product doesn't bury you in disposal fees. Never let a supplier talk you into 1,000 units on a first order.
What New Sellers Waste Money On
- Oversized first orders. The discount on 1,000 units is real; the aged-inventory surcharge on 700 unsold ones (up to $0.30–0.35/unit/month for long-stored stock in 2026) is realer.
- Skipping samples to save three weeks. One bad production run costs more than every sample you'll ever order.
- $0 photography. Phone photos on a white sheet convert worse than a $150 shoot. Your listing is your storefront.
- Launch PPC with no ceiling. Know your break-even ACOS (it's your pre-ad margin — ~43% in the example above) before day one. Ads above break-even during launch are fine; ads above break-even forever are a slow bleed.
- Ignoring the fee stack until month three. Model every fee before you pick the product, including the hidden ones — see our hidden fees audit guide.
Free Tool: Test the Numbers Before You Spend
FBA Profit Calculator
Before you order samples, run your candidate product through the free calculator — 12 marketplaces, current 2026 rates, plus a price ladder and stress-test mode so you can see margin at the sell price you'll actually compete at, not the one you hope for. Two minutes in the calculator beats two months of tuition.
Open the FBA Profit Calculator →Free. No sign-up. 12 marketplaces, 2026 rates.
The Bottom Line
In 2026, starting an FBA business costs about $2,500 on the lean end and $5,000–8,000 done properly — and the fee environment rewards preparation: fee changes this year were modest on the headline but moved quietly on inbound, storage tiers, and inventory health. The winners aren't the ones who spend the most; they're the ones who know their per-unit economics cold before the first PO goes out. Run the calculator, order samples, and let the numbers — not the hype videos — pick your first product.