A box with your brand on it sits in an Amazon warehouse. Someone three states away types in what they want, clicks buy, and Amazon picks that box, packs it and ships it while you are asleep. Which product would you want inside it?
Compare that with how your income works now. You trade hours for a paycheck that someone else decides. Ask for more and you get a speech about budgets. Meanwhile the price of everything climbs, your savings sit still, and the friend who started selling online two years ago just put down a deposit on a house.
Amazon controls 38% of US e-commerce, so your buyers are already there and already searching. Your side of the work is choosing the product, having a manufacturer make it under your own brand, and getting the inventory into the warehouse, then looking after the listing and the advertising that feeds it.
Every category fills up over time. The sellers who find a gap and collect early reviews make it painfully hard for anyone who arrives later. The niche you spot this month could belong to somebody else by next season.
Startup costs run $0-$500 and first profit tends to land in 1-3 months. The income range here is $1,000-$10,000 a month, and $1,000 is the honest number to aim at first. Today, search Amazon for a product you already own and understand, then write down the price and the review count of every listing on the first page.
FBA is Amazon's fulfillment service. You send your products to Amazon's warehouses, and Amazon handles storage, picking, packing, shipping, customer service and returns. You focus on finding products, creating listings and marketing. Amazon handles the logistics.
You create your own branded products by sourcing generic items and adding your brand. This is the approach that scales furthest and builds the most value.
You buy branded products at wholesale prices and resell them on Amazon as an authorized reseller of someone else's brand.
You buy discounted products from retail stores and resell them on Amazon for profit.
The same as retail arbitrage, with your sourcing done from online retailers instead of physical stores.
You know the models now and roughly which one you would run. Everything after this hangs on a single choice, and here is where you make it.
Product research is where fortunes are made or lost. Spend 80% of your early time here.
The 3-star reviews are where I would start. What do buyers of your chosen product keep complaining about that you could fix with one change? That fix is your reason to exist on the page.
Your brand printed on a supplier's sample box is the moment product research stops being homework. You are now someone who owns a product, with an MOQ, a margin and a listing that carries your name. Ask two suppliers for samples before you choose.
You have a product in mind now, but a product you like on a screen is still a hunch. Turning it into stock you own means dealing with the factory that makes it.
Hello, I am looking for a manufacturer for [product description].
Please provide: FOB price for different quantities, MOQ, Production time, Sample cost and shipping.
I am building a long-term partnership and plan to scale significantly.
Would you notice a 3% defect rate before your customers did? An inspection at $200-400 is how you make sure you do.
Your listing is your salesperson, working 24/7. Put real effort in here.
For your first product, pay for professional photos. They directly affect your conversion rate.
Keywords decide who finds your product.
Process:
- Use Helium 10 Cerebro or Jungle Scout Keyword Scout
- Enter competitor ASINs
- Export the keywords they rank for
- Pick out high-volume, relevant keywords
- Group them by search intent
Keyword types:
- Primary keywords: high volume, highly relevant (use in title)
- Secondary keywords: medium volume (use in bullets/description)
- Long-tail keywords: lower volume, specific (use in backend)
Listing optimization
Title formula: [Brand] + [Primary Keyword] + [Key Feature] + [Secondary Keyword] + [Size/Quantity]
Example: "BRAND Kitchen Knife Set 5-Piece - Professional Chef Knives with Ergonomic Handles - Stainless Steel Cooking Knife Block Set"
Bullet points:
- Lead with benefits, support with features
- Use all 5 bullet points
- Include relevant keywords naturally
- Answer common customer worries
- Keep each bullet 150-200 characters
Product description:
- Tell your brand story
- Expand on your bullet points
- Use HTML formatting (available via Brand Registry)
- Include long-tail keywords
Backend keywords:
- 250 bytes for search terms
- No commas, no repetition
- Include misspellings, synonyms
- Spanish keywords for the US market
Amazon PPC advertising
Your listing is live and nobody has seen it yet. Paid placement is how you buy those first impressions while your organic ranking is still nothing.
Pay-per-click advertising is essential when you launch a new product. You cannot rank organically without some early sales velocity.
Campaign types
Sponsored Products:
- Most important for you as a new seller
- Appear in search results and on product pages
- Three match types: Broad, Phrase, Exact
Sponsored Brands:
- Require Brand Registry
- Banner ads at the top of search
- Great for brand awareness
Sponsored Display:
- Retargeting and competitor targeting
- Lower priority while you are starting out
Your launch PPC plan
Week 1-2: Research phase
- Automatic campaign with a $20-30/day budget
- Let Amazon find the keywords that convert
- Gather data on what works
Week 3-4: Optimization phase
- Export converting keywords from the auto campaign
- Create a manual campaign with exact match
- Add negative keywords for irrelevant terms
- Raise bids on high-converting keywords
Ongoing:
- Target ACoS (Advertising Cost of Sale): 25-35%
- Review and optimize weekly
- Add new keywords, remove poor performers
- Adjust bids based on performance
What to budget
For a new product launch:
- Month 1: Expect negative ROI on ads (you are building velocity)
- Month 2-3: Break-even to slight profit
- Month 4+: Profitable campaigns supporting organic rank
Budget 25-30% of revenue for PPC at first, falling to 10-15% as your organic rank improves. Could you watch $20-30 a day go out for a month with little coming back, and keep your nerve? That first month is meant to look like a loss, so decide now how long you will give it.
Getting reviews
Reviews are social proof. They directly affect your conversion rate and organic rank.
Legitimate ways to get them
Amazon Vine Program:
- Available to Brand Registered sellers
- You give free products to trusted reviewers
- Costs: product cost + $200 enrollment fee
- Expect 3-30 reviews depending on category
Request a Review Button:
- Built into Seller Central
- One-click review request
- You can send it 5-30 days after delivery
Product Inserts:
- Include a card in the packaging
- Ask for feedback (never specifically for positive reviews)
- Offer no incentives
- Point buyers to the Amazon review page
Follow-Up Emails:
- Use tools like Jungle Scout or Helium 10
- Automated sequences after delivery
- Put customer satisfaction first
Common mistakes
- Never pay for reviews
- Never offer discounts for reviews
- Never use review manipulation services
- Never ask family or friends to review
Amazon's algorithms detect manipulation. Account suspension is common, and it is devastating for you when it happens.
Running your FBA business
Key metrics to track
Profitability:
| Metric | Definition |
|---|
| Gross margin | Revenue - COGS |
| Net margin | After all Amazon fees |
| TACOS | Total Advertising Cost of Sale (ad spend divided by total revenue) |
Health metrics:
- IPI Score: Inventory Performance Index (keep above 400)
- Account Health: order defect rate, late shipment rate
- Return rate: target under 3%
Growth metrics:
- Organic rank for your target keywords
- Review velocity and average rating
- Your market share in the niche
Inventory management
Reorder formula: (Daily Sales x Lead Time) + Safety Stock = Reorder Point
Example:
| Variable | Value |
|---|
| Daily sales | 10 units |
| Lead time | 45 days (production + shipping) |
| Safety stock | 15 days |
| Reorder point | (10 x 45) + 150 = 600 units |
Common mistakes:
- Running out of stock (kills your organic rank)
- Overstocking (storage fees, aged inventory fees)
- Forgetting seasonal spikes
- Forgetting Chinese New Year (February)
Scaling your business
Horizontal scaling:
- Launch more products in the same niche
- Cross-sell between your products
- Bundle products together
- Create variations (colors, sizes)
Vertical scaling:
- Expand to international Amazon marketplaces (UK, DE, JP)
- Launch your own website with Shopify
- Wholesale to retail stores
- Build a subscription/repeat purchase model
Financial scaling:
- Reinvest 50%+ of profits at first
- Use Amazon lending or Payability for cash flow
- Consider business credit cards for points/float
Reinvesting half the profit feels slow until you see what the other half does at home. If a product settles at $1,000 a month in profit, the share you take home could cover the kids' school shoes and the dentist visit the week they come up, with the rest buying the next batch of inventory.
Common Mistakes to Avoid
Product selection mistakes
- Choosing on passion and ignoring the data
- Skipping competition analysis
- Underestimating regulatory requirements
- Picking seasonal products without a plan
Financial mistakes
- Underestimating total costs (always add a 30% buffer)
- Forgetting that Amazon's fees change
- Poor cash flow management
- Pulling profits out too early
Operational mistakes
- Skipping quality control
- Having no backup suppliers
- Ignoring customer feedback
- Poor inventory management
Marketing mistakes
- Launching without a PPC budget
- Expecting organic rank immediately
- Leaving listings unchanged when the data says otherwise
- Ignoring what competitors do
Your path to a profitable first year
Month 1-2: Foundation
- Complete product research
- Source and order your first product
- $2,000-4,000 invested
Month 3: Launch
- Inventory arrives at Amazon
- Listing goes live
- Begin PPC campaigns
- First sales (likely negative profit)
Month 4-5: Optimization
- Gather reviews (15-30+)
- Optimize PPC campaigns
- Improve the listing based on data
- Approaching break-even
Month 6+: Profitability
- Product profitable
- Organic rank improving
- Begin research for a second product
- Target: $3,000-5,000/month profit
Month 9-12: Scale
- 2-3 products live
- Systems and processes documented
- Steady sales without daily intervention
- $3,000-5,000/month profit
By month six the target on this page is $3,000 to $5,000 a month in profit. If two or three products hold that through month twelve, alongside stock costs and your own bills, your parents' winter electricity and a weekend away for them could come out of it without touching your savings. Document fulfilment and reordering so your second product runs as quietly as the first.
Picture the first month that product pays out. That could be a car payment gone, a weekend trip paid in full, or a little set aside for your parents. It starts small. Then each reorder becomes a reminder that you built something that runs while you live your life.
Is Amazon FBA right for you?
FBA suits you if you:
- Have $2,000+ to invest
- Want to build something with real value
- Are patient (3-6 months to profitability)
- Enjoy research and optimization
- Want semi-passive income in the long run
FBA will frustrate you if you:
- Need money immediately
- Have zero capital to invest
- Cannot handle financial risk
- Want completely passive income
- Hate dealing with logistics and suppliers
Read the second list again slowly. If even one line describes you today, which one is it, and can you change it within six months? If not, a lower-capital tactic on this site is the kinder choice for now.
Getting Started Today
- Create an Amazon Seller Account ($39.99/month Professional plan)
- Subscribe to a research tool (Jungle Scout or Helium 10, $40-100/month)
- Complete your product research (2-4 weeks)
- Source and order samples (2-3 weeks)
- Place your first order once samples pass inspection
Amazon FBA is slow money. It is a real business that asks for real investment of both money and time. If you are willing to put in the work, it offers a proven path to a scalable e-commerce business with real equity value.
The best time to start was 5 years ago. The second best time is now.
What Amazon actually charges you
Now you know how the selling side works. What it costs to run decides whether the income figures at the top of this page apply to you.
This page used to describe product research and advertising before setting out the fees, which was the wrong order. Fees decide whether a product is viable before any of the rest matters.
The two things you always pay
A selling plan. Amazon offers an Individual plan at $0.99 per item sold and a Professional plan at $39.99 a month. The arithmetic is simple: above roughly forty units a month the Professional plan is cheaper, and it also gives you bulk listing, advertising eligibility and the other tools a real operation needs.
A referral fee on every sale. This is the commission Amazon takes for the sale itself, and you pay it whether you fulfil the order yourself or through FBA.
Referral fees are set per category, and the minimum is what catches people
Most FBA content quotes 15%. That is one common rate among many, and Amazon's published table varies a lot by category.
| Category | Referral fee | Minimum |
|---|
| Amazon Device Accessories | 45% | $0.30 |
| Appliances, full-size | 8% | $0.30 |
| Automotive and Powersports | 12% | $0.30 |
| Base Equipment Power Tools | 12% | $0.30 |
| Appliances, compact | 15% up to $300, then 8% above | $0.30 |
| Baby Products | 8% at $10.00 or less, 15% above | $0.30 |
Two structural features matter more to you than any single rate.
You pay the percentage or the minimum, whichever is greater. A $0.30 minimum on a $2 item is 15%; on a $1 item it is 30%. This is one of several reasons cheap products fail on this platform, and you will never see it if you read only the headline percentage.
Some categories tier the rate. Appliances and baby products both change rate at a price threshold, so your effective percentage moves with your price point.
Check your specific category before you model anything. A plan built on 15% that turns out to sit in a 45% category is a different business altogether, with a different answer. Do you know which row your product falls into? Look it up before you read another word about sourcing.
FBA fees sit on top
Choosing FBA adds fulfilment and storage charges, and these follow the physical size and weight of your product, whatever its price.
Fulfilment fees are charged per unit and scale with size and weight tier. That is the practical reason behind the advice to prefer small, light products: a heavier item in a larger size band can eat the entire margin on a mid-priced product.
Storage fees are charged on the volume you hold, and they rise sharply for inventory that has been sitting a long time. Overstocking costs you twice: your capital is tied up in goods, and you pay a rising charge to keep them.
Amazon publishes a Revenue Calculator where you enter a product's dimensions, weight, category and price and compare FBA against fulfilling it yourself. Use it before you commit to a product. Amazon notes its results are estimates and actual costs may vary, which is fair warning to leave margin in your own model.
The stack, in the order it hits you
For a single unit sold through FBA, the deductions run roughly like this: the referral fee on the sale price, then the FBA fulfilment fee for that size and weight tier, then a share of monthly storage, then the advertising cost per unit sold, then your landed product cost including freight and duty, then returns and the units you write off.
Work that stack for a real product before you order inventory. The usual failure is modelling only the referral fee and the product cost, arriving at a comfortable margin, and discovering the rest of the stack after committing your capital to a thousand units.
Why the price floor exists
Put the pieces together and they explain a rule you will hear repeated without any reason given.
Below roughly $20, the fixed parts of the stack become a large share of the sale. The referral minimum, the per-unit fulfilment fee and the advertising cost per conversion stay the same whatever you charge, so they eat a growing share of a low price. Above roughly $70, impulse buying falls away and conversion drops, which raises your advertising cost per sale.
That is where the commonly cited $20 to $70 band comes from. The fee structure produces it, which is why it holds across categories.
The capital you will need, stated plainly
FBA often gets presented alongside models that need almost nothing to start. It needs real money, and being clear about the number protects you from the most common way this goes wrong: running out of money mid-cycle.
Where your money goes before any comes back
Samples. Several suppliers, several units each, shipped individually by air. You cannot skip this, and it is the cheapest money you will spend, because it is what stops you ordering a thousand units of something unsellable.
The first inventory order. Suppliers set minimum order quantities, and the minimum is usually higher than a cautious person wants. This is your largest single line.
Freight, duty and customs. Sea freight is cheaper per unit and takes weeks; air is fast and expensive. Duty depends on the product classification and the origin, and it is charged on landed value, which is more than you paid the factory. Trade policy has been volatile enough in recent years that you should check any rate quoted in a course from two years ago.
Photography and listing assets. You can do these yourself at near zero cost, or spend a few hundred dollars to have them done properly.
Advertising during launch. A new listing has no ranking and no reviews, so your early sales are bought. The earlier guidance on this page of 25 to 30% of revenue during launch is a real cost, and you spend it before the product is profitable.
A buffer for the reorder. People forget this one most often, and it kills otherwise working businesses. If your product sells, you must order the next batch before the first has finished paying out. Sell through your entire stock, wait for Amazon's payment cycle, then order, and you are out of stock for the weeks that matter most, losing the ranking you just paid to build.
The realistic shape
Your plan needs enough capital for the first order, the launch advertising, and the second order placed before the first has been fully recovered. Trying it on the first order alone is the standard failure, and it hits at the moment of success, which is what makes it so crushing.
Here is the honest read. This is a business with real inventory risk and a working capital cycle, closer to retail than to the zero-cost models elsewhere on this site. That is fine, as long as you go in with capital you can afford to have locked up for months, and treat your first product as tuition. If this money vanished into boxes for five months, would your rent notice? Your answer tells you how big a first order you can place.
Account risk can end the business
Everything above assumes your account keeps existing. That assumption deserves a hard look, because suspension is common and its consequences are severe.
Your funds are held on suspension. A suspended account cannot sell, and the balance is usually withheld while the matter is reviewed. Your inventory stays in Amazon's warehouses, running up storage fees, unsellable.
Intellectual property complaints are the most frequent trigger. A rights holder reports your listing, Amazon removes it and records a mark against your account, and the burden of resolving it falls entirely on you. This can happen through no fault of your own if a supplier ships you goods that infringe.
Authenticity and safety complaints escalate quickly, especially in categories touching health, cosmetics and anything for children.
Performance metrics matter more than sellers expect. Late shipment rate, order defect rate and cancellation rate all carry thresholds, and crossing them puts your account at risk however good the product is.
Multiple accounts are prohibited, and detection extends to shared addresses, payment methods and devices. Opening a second account after a suspension is the fastest route to permanent removal.
Reducing your exposure
Keep documentation for everything you sell. Supplier invoices showing the chain of custody are what Amazon asks for when authenticity is questioned, and a supplier who cannot provide a proper invoice cannot help you when it matters.
Check trademarks before choosing a product, using the free USPTO search described in this site's print-on-demand guide. Brand-registered categories and products resembling protected designs are problems you can avoid.
Stay out of gated or safety-sensitive categories until you understand them. The approval processes exist because the complaint rates are high.
Read Amazon's policy updates. They change, they apply retrospectively, and Amazon will not accept that you missed a change announced three months ago.
Keep a second channel alive. A website, a wholesale relationship or another marketplace. Sellers whose entire revenue runs through one Amazon account find out about that concentration risk at the worst possible moment.
You will see the same pattern in every platform business on this site: the account belongs to Amazon, the rules change without your input, and your protection is never depending on a single one. If your account were frozen tomorrow, where would your next sale come from? Have an answer before you need one.
Reading the success numbers honestly
FBA produces more inflated claims than almost any category, and the reasons are built into how the business works.
Revenue gets quoted, profit almost never does. A seller reporting a six-figure month is reporting money that passed through the account before referral fees, fulfilment, storage, advertising, product cost, freight, duty and returns. On typical margins that headline can mean a modest fraction in profit, and occasionally a loss.
Amazon's own figures describe revenue too. Marketplace statistics quote sales through the platform, which tells you nothing about seller earnings.
Survivorship is severe. The sellers making content about FBA are the ones it worked for. The ones holding unsold inventory are not making videos about it, which distorts the visible success rate beyond recognition.
Course sellers dominate the conversation. A large share of prominent FBA voices earn more from teaching the method than from practising it, and the numbers in their marketing often come from the teaching business.
What to measure instead
Profit per unit after the full fee stack, calculated with Amazon's Revenue Calculator and your real landed cost, before you order.
Return on the capital you tied up, over the full cycle from paying the supplier to receiving the payout. A healthy percentage over a five-month cycle is a very different proposition from the same percentage over six weeks.
Cash conversion. Money in inventory is money you cannot use. Of two products with identical margins, the one that turns over twice as fast is the better one.
Advertising cost of sale, tracked as it changes. Launch spending is expected to lose money; if it has not fallen clearly once organic rank settles, the product does not work at that price.
Who Should Skip This
If the capital would hurt to lose, do not start. Inventory can be unsellable, and the way this fails is your money sitting in boxes instead of in an account.
If you want to start earning within weeks, this is the wrong model. Research, sampling, manufacturing, freight and launch put the first meaningful revenue months away.
If you will not do the fee arithmetic, the rest of your effort is wasted. Product selection decides the outcome, and you make that decision with a calculator, leaving your gut feeling about what people want out of it.
If you cannot tolerate depending on one platform, know that an account suspension can end the business overnight with your capital sitting in a warehouse.
What FBA genuinely offers, and what keeps drawing people to it, is real: access to the largest product-search audience in the world, fulfilment and customer service at a scale you could never build yourself, and a business you can sell as an asset. It is a real business with real capital requirements, and it rewards you if you treat it as one.
Where This Goes Next
Three judgements about direction, offered as reasoning more than forecast.
Fee pressure keeps rising. Amazon has added fee categories steadily, and each addition narrows the band of products that work. If you model a product on today's fees with no headroom, you are planning for a margin that will turn out smaller than you expect.
Standing out matters more as sourcing gets easier. Generic private label, where a dozen sellers rebrand an identical factory product, competes only on price and advertising spend. The sellers who last have a genuine product improvement, a brand people ask for by name, or a category where expertise keeps others out.
Your presence off Amazon becomes the lasting asset. A brand with its own customers, its own list and its own storefront is worth substantially more than an FBA listing, sells for a higher multiple, and survives an account problem. The sensible way to read the whole model: Amazon is an excellent distribution channel and a poor place to keep the entire value of your business.
Think about what that means for timing. The listings that hold page one next year are being built and reviewed right now. Every month you wait, those top spots fill with sellers who got there first, and the reviews they collect become a wall in front of you.
Sourcing decides everything downstream
Product research decides whether a product can work. Sourcing decides whether yours does, and it is where inexperienced sellers lose the most money.
Order samples, and order more than one round
Order from at least three suppliers before you choose. You are testing three things at once, and the product is only one of them.
Product quality, obviously, judged against your own use. Photographs prove nothing.
Communication. A supplier who is slow or evasive over a sample order will be slower and more evasive over a problem with a thousand units. This is the best predictor you have, and it costs nothing to watch.
Willingness to accommodate. Ask for a small change: different packaging, a logo, a modified component. How they respond tells you whether you have found a manufacturing partner or someone who just takes orders.
Then order a second sample from your chosen supplier, made to your specification. The first sample proves they have a good product. The second proves they can make yours.
Inspection before shipping
Arrange a third-party inspection at the factory before the goods leave. It costs a fraction of the order, and it is your only chance to catch a problem while the supplier still has your money and your goods in the same place.
Once a container is at sea, a defect is your problem. Once it is in Amazon's warehouse, a defect is your problem plus removal fees.
The terms worth agreeing in writing
What happens to defective units, and who pays for replacement and freight.
The exact specification, including materials, dimensions, packaging and labelling. "The same as the sample" gives you nothing to hold anyone to.
Payment structure. A deposit with the balance paid against inspection is standard and protects you. Paying in full up front gives away your only leverage.
Compliance documentation for any product with safety, electrical or chemical requirements. You will need it if a marketplace or a customs authority asks, and asking for it afterwards rarely works.
The mistake that ends first attempts
Ordering a large first batch to reach a better unit price.
The saving is real, and it is the wrong trade. Your first order is an experiment, and an experiment should be cheap to get wrong. Order the smallest quantity a supplier will accept, accept the worse unit economics, and treat the difference as the price of finding out whether anyone wants your product before you commit serious capital to it.
Sellers who negotiate hard on unit price for their first order often end up with several thousand units of something that does not sell, which costs them far more than the margin they protected. What is the smallest first order your supplier will accept? Ask that before you ask for a discount.
So what does waiting cost? Another season of watching someone else's brand fill the page you could have studied tonight. Pull up that first page of results, note the prices and review counts, and pick one listing you know you could improve. That is the whole first step.