Start With The Right Amazon Model

Private label delivers the highest profit margin on Amazon, wholesale the fastest cash flow, and arbitrage the lowest entry capital. Private label is the only model that builds a sellable asset over the long term; dropshipping is the riskiest model and the most fragile one in terms of Amazon policy. The right answer is not in the model itself, it is at the intersection of the capital you have, the time you can commit, and your risk tolerance.

The practical thresholds are as follows. Starting private label with a budget under $3,000 usually turns into an unfinished project, because you cannot fund product cost, customs, launch advertising, and the first reorder at the same time. The $5,000 to $15,000 range is the realistic entry band for private label. In wholesale you can start with $5,000 of initial inventory, but with a 10% to 20% net margin you need volume for meaningful revenue. Arbitrage can be started with $500 to $2,000 for learning purposes, but it is a model that converts time into capital and is labor intensive to scale.

In the Mentoreis consulting process the first session is dedicated to model selection, because this decision determines every decision that follows: which product you will look for, which company structure you will set up, how much cash you need to hold, and when you will turn a profit. A seller who picks the wrong model can lose money even with the right product.

Private label means having a manufacturer produce a product under your own brand and selling it on Amazon. The appeal of the model is in margin and ownership: the listing you compete on is yours, you do not share the Buy Box, you set the price, and brand value accumulates over time. A healthy private label brand built over 2 to 3 years can sell at a multiple of 3 to 5 times its annual net profit; no other Amazon model offers an exit like that.

The realistic capital table looks like this. For a typical first product, a production run of 500 to 1,000 units costs $3,000 to $8,000, sea freight and customs $800 to $2,500, photography, video, and brand design $500 to $1,500, the PPC budget for the first 90 days $1,500 to $4,000, and trademark registration plus company formation $800 to $2,000. That adds up to a band of $6,000 to $15,000. The critical point is this: spending that entire amount on the first order is the most common mistake, because when the product works there is no cash left for the second order that arrives 45 to 60 days later, and running out of stock resets the ranking you built.

The timeline matters just as much. Product research and supplier selection take 3 to 6 weeks, the sampling process 3 to 5 weeks, production 25 to 45 days, and sea freight plus Amazon warehouse check-in 35 to 60 days. So in most projects the first sale happens 4 to 6 months in. The return on the investment, meaning the break-even point, usually falls 6 to 12 months after the first sale. In this model patience matters as much as capital. Trademark registration and Brand Registry are not optional, they are part of the model, because they unlock A+ Content, Sponsored Brands, Brand Store, and Vine access.

Wholesale means buying products from well known brands in bulk, from an authorized distributor or from the brand itself, and selling them on the existing Amazon listing. Unlike private label there is no new listing, no new brand, and no launch process; the product already has demand, reviews, and ranking. That is why the cash cycle is much faster: with the right product, sales start within days of inventory check-in and the first profit typically shows up in the first 30 to 60 days.

The economics of the model are built on volume. The wholesale purchase price is usually 40% to 60% of the retail price, the Amazon referral fee is 8% to 15% in most categories, the FBA fulfillment fee is added based on product size, and what typically remains is a 10% to 20% net margin. You can start with $5,000 of initial inventory, but meaningful revenue generally requires $20,000 or more in working capital. In exchange, because the same product is ordered over and over, the model is predictable and scaling it is less risky than private label.

The real difficulty in wholesale is not finding products, it is getting an account approved. Serious brands do not open accounts for random sellers; they ask for a US registered company, an EIN tax number, a resale certificate, a professional website, and often a physical address. Getting 3 to 10 positive responses out of 100 brand applications is normal. The second difficulty is the buy box: if there are 10 sellers on the same listing your buy box share drops and price competition erodes the margin. That is why in Mentoreis wholesale analysis we review the seller count per product, whether the brand owner sells directly, and the last 90 days of price history together; if the seller count is above 6, the product is usually eliminated.

Arbitrage means buying a product at a discounted price and selling it on Amazon at a higher price. Online arbitrage sources from internet stores, while retail arbitrage buys from physical stores like Walmart and Target. The biggest advantage of the model is the low barrier to entry: you can start with $500 to $2,000, and there is no production, no supplier negotiation, no brand, and no launch process. Sales usually start within the first 2 to 4 weeks.

The real cost is not in capital, it is in time. Finding one profitable product means scanning dozens of products, and each product requires its own invoice, its own box, its own labeling, and its own FBA shipment. An experienced arbitrage seller spends 10 to 20 hours a week on sourcing alone. The margin math is also tighter than it first looks: on a product bought for $20 and sold for $35, the Amazon referral fee is about $5.25, the FBA fulfillment fee $4 to $6, and inbound shipping plus labeling $1 to $2, leaving $2 to $4, meaning a 6% to 11% net margin.

The most critical risk sits on the legal and operational side. When an IP claim comes in, Amazon asks the seller for an invoice showing the product was bought from an authorized source, and a retail receipt does not stand in for that document. For that reason, selling restricted brands can lead to a listing being taken down or to an account review. Arbitrage is valuable as a learning tool: it teaches Seller Central, FBA operations, profit math, and the returns process with real money. But it does not scale and it does not produce a sellable asset. Mentoreis generally positions this model as a starting point, not as a destination.

Dropshipping on Amazon is technically permitted, but in the form most sellers practice it, it is a policy violation. Amazon's Drop Shipping Policy is clear: you must be the seller of record, every invoice, packing slip, and outer package in the shipment must carry your name, and no other retailer's invoice or box may reach the customer. In other words, ordering from Walmart or AliExpress and letting that box go to the customer is direct grounds for account suspension.

In practice this model breaks in three places. First, delivery control: when the supplier ships late your late shipment rate rises, and accounts that cross the 4% threshold come under risk. Second, inventory synchronization: a product that runs out at the supplier stays live on your listing and you end up having to cancel the order. There is a common mix-up here: seller initiated cancellations do not count toward the order defect rate (ODR), they go into a separate metric, the pre-fulfillment cancel rate. The threshold for that metric is 2.5% and it is measured over a 7 day window only, so at low order volume even a few cancellations are enough to cross it. Third, margin: because you are a middleman, the typical net margin stays between 3% and 8%, and a single wave of returns can wipe out the monthly profit.

The healthy alternative is the set of models that build the same low inventory logic on legitimate ground. In wholesale, shipping directly from the supplier to the Amazon warehouse; arranging direct shipment to FBA in manufacturer agreements; or using Amazon's own FBM structure with your own warehouse. Mentoreis does not take on dropshipping projects; instead we recommend moving to a wholesale or arbitrage setup with a similar budget, because reopening a closed Seller Central account takes months and often never happens at all.

Once the model is chosen correctly, what determines profit is execution. Most projects that lose money on Amazon fail not because of the wrong model, but because six basic line items were never calculated.

First, the true unit cost. A margin calculated without adding freight, customs duty, the FBA fulfillment fee, the category referral fee, storage, return processing, and the advertising share on top of the product cost is imaginary. Mentoreis starts every project with the assumption that at least 30% of the selling price will be total Amazon cost. Second, the cash cycle. The Amazon payment cycle runs every 14 days, but a critical change came in 2026: funds are now released 7 days after delivery confirmation rather than at shipment. In practice, the time between the order and the money landing in your bank account rose to between 14 and 27 days on FBA. Production and shipping, meanwhile, require 60 to 100 days of upfront financing; a growing Amazon business can run out of cash even while it is profitable. Third, inventory planning. Two to four months of inventory is ideal. The old long term storage fee was removed and replaced by the aged inventory surcharge, which starts at 181 days and is charged every month rather than twice a year; that is why cost accumulates much faster on slow moving inventory. Inventory beyond 12 months wipes out the profit entirely on most products. At the other end there is a low inventory penalty: cutting inventory too far triggers a separate fee, so the target is not the least inventory but the right inventory.

Fourth, certification and compliance. Documents such as FDA, CPSIA, CE, FCC, and OEKO-TEX are mandatory depending on the category, and a product shipped without them gets blocked at the warehouse. Fifth, tax and company structure: forming a US LLC, obtaining an EIN, state level sales tax nexus, and the export records on the Turkish side have to be planned from the start, independent of the model. Sixth, the launch and advertising budget; a project that sets money aside for inventory but not for advertising ends up with a full warehouse and zero sales. In Mentoreis consulting we put these six line items into a single table, and a project is only started when that table holds up.