Keep the Winners, Sell the Losers: What Amazon and the Site Marketplaces Say About That Plan

A content writer with twelve years of client work posted his method on a forum last November and has kept a monthly ledger since. He builds several Amazon Associates sites at once, outsources 60 to 100 AI-assisted buyer guides per site, lets Google index them, updates by Search Console data, and does one thing most guides never mention: he sells the sites that do not perform and keeps the ones that do. His reported figure went from $46.62 in month one to over $1,100 by month five, part of it labelled a bonus.

Those are his numbers, and this page does not repeat them as facts. What it does is read the pages that decide whether the plan works at all: Amazon's own rules for a new account, and the three marketplaces a small affiliate site can actually be sold on. The plan has a shape the forum thread never draws, and the shape is in the terms.

One account, one 180-day clock

Joining Associates is free; the button reads Join Now for Free. But a new account is not approved when it is opened. Amazon evaluates the application once three qualifying sales have been referred, and can withdraw it if 180 days pass without enough of them. For a portfolio this cuts both ways. Five sites under one account share the clock, so the first site to convert keeps the account alive for the other four. Five sites under five accounts are five separate 180-day deadlines, and the slow ones die.

The account is also the one thing in the plan you cannot sell. The operating agreement says it cannot be assigned or transferred, by operation of law or otherwise, without Amazon's prior written approval. Hold that thought until the last section.

What a site costs before it earns

The domain and hosting are the same on every site you build, which is why the count matters. A .com at Porkbun is $11.08 to register and the same $11.08 a year to renew, WHOIS privacy and SSL included. Namecheap's Stellar plan is $5.88 a month and holds three domains. Five sites is five domains and two hosting plans, before a word is written.

The words are the variable cost. The forum method buys them in bulk as AI-assisted drafts. The token prices set the floor: Claude's Haiku 4.5 is $1 per million input tokens and $5 per million output tokens; OpenAI's gpt-4o-mini is $0.15 and $0.60. What a finished, edited guide costs at those rates is worked through on what an AI-written article costs. The editor's hour is the part no pricing page prints, and Amazon's participation rules, not this page, decide what counts as content you may put a link in.

Two rules that make a site look like a loser when it is not

Commission is earned on qualifying items placed in the customer's cart within 24 hours of arriving through your link. A buyer guide that people read on Monday and act on Friday earns nothing, and the dashboard shows a site that "does not convert" when what it shows is a site whose readers take longer than a day. Before you sell a site for not converting, look at what it ranks for and how far from a purchase those searches sit.

The second rule is the calendar. Commission earned in a month is paid about 60 days after that month ends, by direct deposit, gift certificate or check, with a $10 floor on the first two and $100 on the check. The money from a site's best month arrives two months after you decided whether to keep it. Judge sites on the earnings report, never on the bank balance, and give a new site two full payment cycles before it goes on the block.

Which marketplace will take a small site at all

This is where the plan meets the market, and the market has floors.

Empire Flippers requires a business to earn at least $2,000 a month in net profit, averaged over the last 12 months, with 12 months of revenue history and analytics tracking for at least 3 months before applying. A site you are selling because it did not work will not clear that bar, so Empire Flippers is not a door for the losers. It is a door for the winners, if you ever decide to sell one: the listing is free, the valuation is 12 months of net profit times a multiple its tool describes as typically 1.7 to 5 or higher, and the commission is a flat $10,000 on anything up to $66,666.66, then 15%. Listings carry a minimum 2 month exclusivity, and migration runs 2 to 8 weeks before the seller is paid.

Flippa's self-service track publishes no income floor. The Entry listing in the sub-$10K bracket costs $29 for 60 days, and a 10% success fee applies across all three self-service brackets. For businesses sold in 2023 the median time to close was 15 days under $50K. Flippa supplies Letter of Intent and Asset Purchase Agreement templates for the handover, and the rest of the transfer is between you and the buyer.

Motion Invest publishes no minimum income, traffic or age either; listing is by approval after a 10-step due diligence review. It costs $0 to list and the platform is paid only from the sale price, at 20% under $20,000, then 15%, 10%, 7%, down to 5% over $500,000. The clock it publishes is the shortest on this page: an average of 12 days from listing to sale, due diligence 3 to 7 days, the transfer 3 to 7 days, and a wire 3 to 6 business days after the transfer is confirmed. It can also terminate or suspend an account immediately, without notice, for any other reason, so read the terms before the listing goes up.

What the exit costs

Put one sale price through the three fee structures and the shape is obvious. On a $15,000 site, Flippa keeps $29 plus 10%, so $1,529. Motion Invest keeps 20%, so $3,000. Empire Flippers would keep its flat $10,000, which is exactly why a site that size cannot list there in the first place. The cheapest exit for a small site is the one with the listing fee, and the fastest published clock belongs to the one that takes the most.

The income stream that pauses at the handover

Now the thought from the first section. The buyer of an Amazon affiliate site does not buy the Amazon account. The operating agreement forbids assigning or transferring it without Amazon's prior written approval, so the buyer opens their own, replaces every link on the site with their own tag, and starts their own 180-day clock with their own three qualifying sales. If the site was earning commission the day it sold, that commission stops at the link swap and restarts only when Amazon approves the new account.

Motion Invest's transfer team swaps affiliate links as part of migration, which is the mechanical half. The contractual half, that the site's income has to be re-earned by a new account, is the thing to write into the listing before a buyer discovers it after paying. It is also the reason a marketplace values a site on 12 months of net profit rather than on the account: the account was never for sale.

One more thing this page could not read. The forum ledger labels part of its monthly figure a bonus. Amazon's help topic on bonus commissions returned no content to a logged-out reader, so whatever that program pays, and to whom, is not on a public page we could open. It is not on this page either.

What to do today

Open one Associates account and put every site you build under it. Register domains at the renewal price, not the promo price, and count hosting per three sites. Give every site the 24-hour rule and two 60-day payment cycles before you judge it. Then match the verdict to the door: a site under $2,000 a month in net profit goes to Motion Invest at 20% or to Flippa at $29 plus 10%, and a site above it is the one you probably keep. Write the Associates clause into the listing yourself, before the buyer reads it in the agreement.

The method is not "build sites". It is "build sites under one clock, measure them on Amazon's calendar, and know which door will take the ones you let go".