affiliate marketing: how it works and when it pays
affiliate marketing sounds easy for one reason: you don't need your own product. that's true. it's also the catch.
think about what a shop actually has to do: make the product, ship it, handle support - and find buyers. the first three are the easy parts to scale. finding buyers is the expensive part. an affiliate program is a shop paying someone else to do exactly that part. you get paid only when you deliver a buyer.
affiliate marketing gives you the product for free. finding the buyer - the hard part - stays your job.
tl;draffiliate income is a fee for bringing customers. whether it pays depends on four things - how the commission is shaped, how long the tracking lasts, whether people already want the thing, and whether you own your traffic - plus one line of math this page writes out, because most guides never do.
what affiliate marketing is
a shop (the merchant) gives you a special link. when someone clicks it, the shop remembers that visitor came from you - for a set time window, anywhere from 24 hours to 90 days depending on the program. if that person buys inside the window, you get a cut: a percentage of the sale, a flat fee, or a monthly share of a subscription.
you bring the readers and the trust. the shop brings the product, the shipping, and the checkout. on the map of ways to make money online, this belongs to building an audience: the audience is the asset, and affiliate links are one way it pays you.
the one line of math
your monthly affiliate income is:
visitors x how many click x how many buy x the price x your cut - refunds
you only control the first two numbers. the buying happens on the shop's page, at the shop's price, in the shop's checkout. rough industry figures: 0.5-1% of readers click a typical affiliate link, 3-8% click on a page that reviews the product directly, and of the people who click through, 1-3% buy in a normal shop while 4-8% sign up for a well-matched software tool (wecantrack, partnero, 2026 - all of these swing a lot by niche).
run the math and the dream gets specific. an example month, 10,000 readers in each row (made-up numbers, but using the rates above):
| the page | the math | the month's income |
|---|---|---|
| general amazon list post | 10,000 x 5% click x 3% buy x $50 price x 3% cut | ~$22 |
| focused review page | 10,000 x 8% click x 3% buy x $120 price x 4.5% cut | ~$130 |
| software with monthly commission | 10,000 x 5% click x 4% sign up x $29/month x 30% share | ~$174 - and it repeats next month, plus the next batch |
same readers in every row. the difference is the offer, not your effort. that's the main thing to know before writing a single post: traffic multiplies the offer. big traffic times a weak offer is still almost nothing.
thing one: the shape of the commission
three shapes, checked on the programs' own pages in august 2026:
- one-time percentage. amazon pays 4.5% on books and kitchen, 3% on toys, pets, and furniture, 10% on luxury beauty, 0% on gift cards (amazon's fee table). send a buyer for a $1,000 sofa, get $30, start over.
- flat fee. shopify pays up to $150 for a plan signup. hosting programs pay $50-$500 a sale. big fees usually mean slow, careful buyers - fewer of them close.
- monthly share. kinsta pays up to $500 up front plus 10% of the subscription for as long as the customer stays. many software and email tools pay a monthly share in the 20-50% range. this is the only shape where last month's work still pays you this month.
one-time commissions reset you to zero every month. monthly shares stack up. over a year, that difference matters more than traffic does.
thing two: how long the tracking lasts
the tracking window (the "cookie") decides whether you get paid at all. amazon's is 24 hours, softened by a cart rule: whatever the visitor puts in their cart within those 24 hours still counts if they order before the cart expires. software programs usually run 30-90 days (shopify 30, kinsta 60, some programs 90).
match the window to how people buy the thing. impulse buys fit a 24-hour window. a $2,000 camera that people research for three weeks does not - the reader you convinced on monday buys on friday through someone else's link, or no link at all. promoting slow purchases through short windows means doing the work and losing the fee.
thing three: it's easier to catch demand than to create it
two very different jobs hide under one name:
- catching demand: writing for people who are already about to buy - "best email tool for creators", "kinsta vs wp engine" - and helping them decide. small audience, high percentage of buyers.
- creating demand: making people want something they weren't looking for. that's the influencer job. it needs reach and volume, and far fewer of those viewers ever buy.
beginners get sold the second job with the first job's numbers. if you don't have reach, catching demand is the only version open to you. that's why review and comparison pages that rank in search are the classic route - and why a small audience that trusts you beats a big one that doesn't.
thing four: rented traffic can vanish
the warning story is housefresh, an independent review site that really tested air purifiers in a lab. google's march 2024 update cut its search traffic from about 4,000 visitors a day to about 200 - a 91% drop. real reviews, real testing, one algorithm change.
you can't remove that risk. you can shrink it: own your asset - a domain, a site, and an email list that reaches your readers without asking a platform's permission. google also says openly what it buries: "thin" affiliate pages that copy the shop's product description and add nothing. the safe ground is being genuinely useful to the reader. everything else is one update away from zero.
what affiliates actually earn
the industry is big and real: us companies spent $12.4 billion on affiliate programs in 2025 (emarketer). but the money splits the way it always splits online:
- 57.6% of affiliate marketers earn under $10,000 a year, while about 11% earn over $100,000 (influencer marketing hub, 2023).
- a survey of 2,270 affiliates (reported by ahrefs) found an average of $8,038 a month - but that's an average, pulled up hard by the top. in the same data, first-year affiliates averaged $636 a month.
both numbers are true. the average is the ad. the split is the business. affiliate money goes to people who already have an audience, because the fee is paid for exactly that.
when affiliate marketing is the wrong choice
- you need money soon. the math needs traffic first, and traffic takes months. sell your time first, and add affiliate links to whatever audience you build along the way.
- you have no audience and no search plan. a link nobody clicks pays nobody.
- the offer's math is against you. low price x low cut x 24-hour window barely pays at any traffic level - see the $22 row above.
- you'd buy ads to earn commissions as a beginner. that's a knife fight against pros with better data. content first; ads once the math is proven.
- you'd have to stretch the truth to sell it. fake enthusiasm is exactly what readers - and google - punish. one made-up detail costs the trust the whole model runs on.
when it's the right choice: you already publish, your topic has "best x" and "x vs y" searches you could rank for, and you can pick offers with the shape on your side - monthly shares, higher prices, longer windows, products you can actually stand behind.
how to judge an offer in six questions
what's the commission shape (monthly beats flat beats small one-time)? how long is the window, and does it match how people buy this thing? what does the thing cost? how often do people refund or cancel? has the program cut its rates before (amazon has, more than once)? and last: would you recommend this product with no link in it? a no on any of the first five is a math problem. a no on the last one is a trust problem, and trust is the only lasting advantage this business has.
disclosure is the law, and it also works
the ftc requires you to tell readers that you earn from your links, somewhere they can't miss it. treat that as a feature, not a chore: a reader who knows how the site earns and still takes your recommendation is the reader who buys and comes back. q1rk carries a few affiliate links itself, plain and disclosed - how this site earns - so the math on this page is the math this site lives with. written from inside, early, not from a yacht.
faq
what is affiliate marketing in plain terms?
a shop pays you a fee for buyers you send through a tracked link. you bring the audience and the trust; the shop does everything else. income = your traffic x clicks x buys x price x your cut.
how much do affiliate marketers make?
most make very little: 57.6% earn under $10,000 a year, while about 11% clear $100,000 (influencer marketing hub, 2023). the "$8,000 a month average" you see quoted is pulled up by the top earners; first-year affiliates in the same data averaged $636 a month.
can you do affiliate marketing without a website or followers?
technically yes - links work in videos, newsletters, and communities. practically no: the fee is paid for having an audience, so some audience has to exist. a website with search traffic is still the most durable version - how to build a website that makes money.
is affiliate marketing worth it in 2026?
as a way to earn from an audience you're building anyway: yes - us companies pay out over $12 billion a year. as a shortcut that skips building the audience: no. you get the product for free; finding the buyer stays your job.
how long until the first commission?
with an existing audience: days. from zero: months - the wait is the audience-building, not the links. the timeline: how long it takes to make money blogging.
go deeper: how to make money online for the full map, own the asset for why your traffic should be yours, and digital marketing for solopreneurs for the search-plus-email loop that catches demand. more in the notes.