What it actually costs to start an Amazon affiliate site.
Most answers to this are either “practically nothing!” or a padded list of hosting affiliate links. Here are the real numbers for year one, including the two largest costs — which almost nobody lists, because neither of them is something you can be an affiliate for.
The short version
- The cash floor is about $80 for year one. A domain and basic hosting. Everything else can genuinely wait.
- A realistic year one is $300–$600 if you are serious, and most of that is not software.
- The two biggest costs are products and time, and neither appears on any list you will find — because nobody earns commission recommending them.
- Spending more does not speed anything up. A $2,000 setup ranks no faster than an $80 one.
- Budget for month twelve, not month one. The failure mode is not overspending — it is running out of patience while paying $8 a month.
Search this question and you get two kinds of answer. One says it costs almost nothing, which is technically true and practically misleading. The other is a list of twelve tools, each linked with an affiliate tag, totalling several hundred dollars a month for a site with no visitors.
Here is the version with the awkward items included.
The actual costs, year one
| Item | Realistic cost | Notes |
|---|---|---|
| Domain | $10–15/year | Genuinely necessary. Ignore anyone selling you a premium domain — nobody types your URL. |
| Hosting | $5–30/month | Shared hosting is fine until you have real traffic. The introductory rate roughly doubles on renewal — budget for the second year, not the first. |
| Theme | $0–60 once | A free, fast theme is a better choice than a paid multipurpose one. This is the easiest place to waste money. |
| Affiliate plugin | $0–120/year | Free options exist. Paid ones save time rather than making the site rank. |
| $0 | Free tiers cover the first thousand subscribers comfortably. | |
| AI writing | $3–30/month | Only if you use it. On your own API key a 2,000-word draft costs a few cents; tools that resell words charge $25–99/month for the same output. |
| Keyword tools | $0–100/month | Skip entirely in year one. Free sources cover you until you have data of your own. |
| Products to review | $100–1,000+ | The largest cost, and the one no list mentions. See below. |
| Your time | 200–400 hours | The other largest cost. Also never listed. |
The two costs nobody lists
Products
Content built on first-hand experience is what survived the 2024–2026 search updates, and first-hand experience requires owning things. You do not need one of everything — but you need enough that your reviews contain something a spec sheet cannot supply.
A workable minimum is one product per category you cover, which for a properly narrow niche means two or three purchases in year one. In tools or kitchen equipment that might be $300. In a niche where the products cost $600 each, it is a real constraint and worth knowing before you commit — which is one of the factors in choosing a niche.
The honest alternative, if you cannot buy anything, is to be explicit that your reviews are research-based rather than hands-on. That is a legitimate approach with a real method behind it — see reviewing a product you do not own — but it is a weaker position, and pretending otherwise is not an option.
Time
A serious first year is somewhere around 200 to 400 hours: setup, research, and 40–60 articles written properly. At even a modest notional hourly rate that dwarfs every cash cost on the table above.
This matters practically rather than philosophically. It means the question “how much does this cost” is really “can I sustain 5–8 hours a week for a year”, and it means saving $40 on hosting to spend eight hours configuring something cheaper is a bad trade.
Three budgets
What not to buy in year one
- A keyword research suite at $99/month. $1,200 a year for a site with no traffic. Your own Search Console, forum questions and “people also ask” boxes cover you until you have data worth analysing — see judging keyword difficulty.
- A premium multipurpose theme with a page builder. Heavier, slower, and page weight is a real cost. → why page weight matters
- Logo design. Nobody has ever declined to buy a drill because a blog’s logo was made in Canva.
- A company or LLC on day one. Wait until there is income to justify the paperwork, unless your jurisdiction requires otherwise.
- Backlink packages. At best wasted; at worst actively damaging.
- A course. Everything in a $500 affiliate course is published free, including in this guide.
- Managed WordPress hosting at $30+/month before you have traffic. Move when the traffic justifies it, which is a nice problem to have.
The rule that saves the most money. Buy nothing until you have felt the specific pain it solves. Every tool on that list is genuinely useful later, to someone with traffic and data. Bought in month one they are a way of feeling productive without publishing anything, and publishing is the only thing that matters in month one.
When it pays for itself
At the sensible budget — call it $400 in year one — break-even needs roughly $35 a month in commission by month twelve.
Using the funnel model from why affiliate sites do not earn — 10% of readers clicking, 7% of those buying, $10 average commission — that is about 500 visitors a month. Achievable in year one with 40–60 well-targeted articles, and not achievable with twelve.
The realistic timeline, stated plainly. Expect no meaningful revenue for six months. Expect first commissions somewhere in months three to six. Expect break-even on cash around month nine to twelve, and on time considerably later than that. If those numbers make the project unattractive, better to know now than in month seven — and if a site is not earning at month twelve, the diagnosis is usually in the funnel rather than the budget.
The order to spend in
Spread across the year rather than paid up front.
Month 1 — about $80
Domain, hosting, free theme, free plugin. Then write. Nothing else is a prerequisite for anything.
Months 2–4 — about $150
Your first product to review, and an AI key if you use one. Both directly improve what you publish.
Months 5–8 — about $120
A paid plugin once manual work is genuinely costing you hours, and a second product.
Months 9–12 — as earned
Reinvest commission. This is when tools start being worth their price, because you finally have data for them to analyse.
Once there is traffic, the next question is which monetisation earns most on it, and the answer differs by page — that is Associates versus display ads.
Part of The Amazon Affiliate Site Playbook — the whole job in six stages, from choosing a niche to keeping a mature site earning.
Spend the budget on products, not on tools
Ama Affiliate Pro runs AI on your own API key — no credits, no word limits, a few cents per article at provider rates — so the writing costs stay near zero and the money goes into the products your reviews actually need.
Start a 14-day trialCommon questions
How much does it cost to start an Amazon affiliate site?
About $80 for year one at the absolute minimum — a domain and shared hosting. A realistic serious budget is $300–$600, and most of that is products to review rather than software. Spending more does not make the site rank faster.
Can I start an Amazon affiliate site for free?
Almost. You still need a domain and hosting, which is around $80 for a year. Free platforms are a poor substitute: many prohibit affiliate links, you do not control the site, and you cannot move it later. Buy the domain.
Do I need to buy the products I review?
Not all of them, but at least one per category is close to essential now — first-hand experience is what distinguishes content that survived the recent search updates. If you genuinely cannot, be explicit that your reviews are research-based and describe the method rather than implying hands-on testing.
What is the biggest cost people forget?
Two. Products to review, and your own time — 200 to 400 hours in year one. Neither appears on the cost lists you will find, because nobody earns an affiliate commission recommending them.
Do I need paid keyword research tools?
Not in year one. A $99/month subscription is $1,200 for a site with no traffic to analyse. Your own Search Console, forum questions, “people also ask” boxes and Amazon’s search suggestions cover you until you have data of your own.
How long until it pays for itself?
At a $400 budget, break-even is roughly $35 a month by month twelve — around 500 monthly visitors on typical conversion assumptions. Expect first commissions in months three to six and cash break-even around months nine to twelve. Break-even on your time takes considerably longer.
Is expensive hosting worth it at the start?
No. Shared hosting handles a new site comfortably, and a site with 200 monthly visitors gains nothing from a $30/month plan. Move when traffic justifies it. Do budget for the renewal, though — introductory rates commonly double in year two.
Should I set up a company before starting?
Usually not on day one, unless your jurisdiction requires it. Wait until there is income to justify the paperwork and the accounting. Do keep records of what you spend from the beginning, because those costs are typically deductible once you are trading.