ClickBank is a marketplace where people who make digital products meet people who want to sell them for a commission. It has been running since 1998, which in this industry is close to geological time, and its longevity is the reason it still matters and also the reason its reputation is mixed.
Two sides exist. Affiliates pick a product from the marketplace, get a tracking link, send traffic, and earn a percentage of each sale. Sellers list a product, set the commission they will pay, and let affiliates bring buyers. Most people arrive as affiliates because it requires no product, and the marketplace is full of offers paying 50 to 75 per cent, which sounds like an unusually generous arrangement.
The generosity is real and it is also smaller than the number implies, for a reason almost no guide explains correctly.
ClickBank does not take a cut of your commission. It takes a cut before your commission is calculated, and this is the single most consequential thing to understand about the platform's economics.
The mechanism is wholesale pricing. Per ClickBank's own fee documentation, updated 13 March 2026, ClickBank buys the product from the seller at 92.5 per cent of the sale price minus $1, which means it keeps 7.5 per cent plus $1 of what the customer paid. The commission percentage the seller advertises is then applied to that wholesale figure, not to the retail price.
ClickBank's own worked example: a product sells for $100, ClickBank buys it from the seller for $91.50, and a 50 per cent commission pays the affiliate $45.75.
Run that across the price points the marketplace actually uses, at the common 75 per cent commission:
A nominal 75 per cent is really 67 to 69 per cent of what the customer pays, and the gap widens as the product gets cheaper because the flat $1 is a larger share of a small sale. Anyone modelling campaign profitability on the advertised percentage is overstating revenue by six to eight points before a single other cost.
Subscriptions are priced differently again. For rebills under $40, ClickBank buys at 90.1 per cent of retail with no dollar component. A $37 monthly rebill at 75 per cent commission pays the affiliate $25.00, or 67.6 per cent of retail. Free trials are purchased at a wholesale price of $0.50 and $1 trials at $1.00, which is why trial-heavy funnels pay affiliates almost nothing on the front end and make their money on the rebill.
None of this is hidden. It is stated plainly in the fee documentation, and it is simply skipped by most of the content teaching people to use the platform.
New affiliates routinely make their first few sales and then discover that the money is not coming. This is not a malfunction. It is a documented policy with a specific name.
The stated purpose is preventing affiliates from buying through their own links to harvest commissions. The effect on a legitimate beginner is that the first payout is gated behind a pattern rather than an amount. Three sales to three friends who all used the same card will not clear it. Neither will five sales made last week, because of the fourteen-day age requirement.
Plan for it rather than being surprised by it. The requirement is a one-time gate, and once cleared it does not reapply, but it means your realistic first payout is further away than your first commission suggests.
Wire transfers carry a $45 fee, raised from $35 in October 2024 alongside the payout fee rising from $2.50 to $5.00. Returned direct deposits and wires attract a $20 reprocessing charge and switch your account to cheque payments until you fix the banking details.
US-based affiliates earning more than $600 must have a taxpayer ID number linked to the account, which is ordinary tax administration and worth doing before it blocks a payment.
This is the trap that costs people money for doing nothing, and ClickBank states plainly that it will not warn you.
An account holding a positive balance with no earnings is dormant, and the charges escalate: $1 per pay period after 90 days without earnings, $5 per pay period after 180 days, and $50 per pay period after 365 days.
Read the last number again. At fortnightly pay periods, a year-dormant account with a small balance is being charged $50 every two weeks, which is $1,300 a year against a balance that may be a fraction of that. An affiliate who tries ClickBank, earns $80, drifts away, and comes back eighteen months later will find the balance gone and may find the account in deficit.
Two responses are sensible. If you are stopping, withdraw the balance first, which may mean lowering your payment threshold to the $50 minimum so a payment can issue at all. If you are pausing, know that the clock starts from your last earnings rather than your last login.
The standard refund window on ClickBank is sixty days, and sellers may set a custom window between thirty and ninety days. When a customer refunds, the commission you were paid is clawed back.
The consequence is that your reported earnings and your keepable earnings are different numbers for two months. A campaign that looks profitable in week one can be unprofitable by week nine, and this is the specific way that paid-traffic affiliates lose money on the platform: they scale ad spend against gross commissions, and the refunds arrive after the spending decision has already been made.
Refunds and chargebacks can also push an account negative. ClickBank's accounting policy is explicit that it may then withdraw funds from other accounts you control, seize available funds, or invoice you for the deficiency. That is a normal remedy for a platform carrying refund liability, and it means a negative balance is a real debt rather than a number that resets.
The practical discipline is to hold a reserve against refunds rather than treating each payout as profit. What proportion depends entirely on the offer, and the only way to know is to run traffic and watch your own refund rate across a full sixty-day window before scaling anything.
The ClickBank marketplace surfaces a metric called gravity, and it is widely misread as a measure of quality or of how much a product earns. It is neither. Gravity is a weighted count of distinct affiliates who have made at least one sale of that product recently, with more recent sales weighted higher.
High gravity tells you the offer converts for many different affiliates, so the funnel works and the tracking pays out. It also tells you the offer is crowded, that the obvious traffic angles are being run by people with more budget and more history, and that ad platforms have seen the landing page many times.
Low gravity tells you either that the offer is new, or that affiliates tried it and stopped. Those are very different situations and gravity cannot distinguish them. Zero gravity on a product that has been listed for two years is a warning rather than an opportunity.
The numbers that matter more sit next to it. Average earnings per sale tells you what a conversion is actually worth after the wholesale calculation. The presence of a rebill tells you whether the offer pays once or continuously. Initial versus average sale value tells you how much of the earnings come from upsells the customer sees after the first purchase, which matters because a funnel earning most of its money from a $297 upsell behaves very differently from one earning it all on a $47 front end.
Before promoting anything, buy the product. This sounds expensive and it is the cheapest research available. You will see the upsell sequence, the email follow-up, the refund experience and the quality of the thing you are about to recommend to people who trust you.
ClickBank's open marketplace model means product quality varies enormously, and the categories that pay the highest commissions are the ones where that variance is most dangerous.
Supplements, weight loss, blood-sugar and brain-health offers, survival products, and make-money-online courses dominate the high-commission end of the marketplace. Those categories pay 75 per cent because the products carry high margins and the claims do the selling. Some are legitimate. Some make health claims that would not survive regulatory scrutiny, and a few of the make-money offers are teaching people to promote ClickBank offers, which is a closed loop rather than a business.
This matters to you commercially, not only ethically. Promoting a low-quality offer produces refunds that claw back your commissions, chargebacks that damage the account, and complaints that attach to your name rather than the seller's, because you are the one whose recommendation the buyer acted on. An audience you spend two years building can be spent in a fortnight.
The filter is simple and it costs you access to some of the highest-paying offers in the marketplace. Would you tell a specific person you know to buy this? If not, the commission is not the relevant number.
Where the traffic comes from
ClickBank supplies the products, the tracking and the payments. It does not supply buyers, and that is the entire job.
Content and search. Building a site or channel around a specific problem, ranking for the questions people ask while trying to solve it, and recommending a product inside a genuinely useful answer. This is slow, compounds, survives platform changes better than the alternatives, and is the approach most compatible with a real audience. Expect six to twelve months before it produces meaningful revenue.
Email. The most durable asset in affiliate marketing, because it is the only channel you own. The pattern that works is to give something genuinely useful in exchange for the address, deliver value repeatedly, and promote occasionally. The pattern that fails is treating the list as a broadcast channel for offers, which trains people to stop opening.
Paid traffic. The fastest and by a wide margin the most expensive way to learn. It converts a marketing problem into an arithmetic problem, which is clarifying, and the arithmetic is unforgiving. Most affiliate advertising accounts on Meta and Google are also operating against policies that restrict exactly the claims these offers rely on, and account bans are common enough to be a planning assumption rather than an accident.
Short-form video. Effective for consumer products with visible demonstrations, poor for anything requiring explanation. It also puts your face and your recommendation in front of a large audience quickly, which amplifies the catalogue problem in both directions.
Native and display. Where much of the high-volume supplement traffic runs. It requires meaningful budget to test and expertise in creative and compliance, and it is not where a beginner should start.
The paid-traffic arithmetic
If you do run ads, the model needs to hold at the level of a single customer, and the wholesale calculation belongs in it from the start.
Take a $97 offer at 75 per cent commission. Your real commission is $66.54, not $72.75. If your landing page converts cold traffic at 2 per cent, one hundred clicks produce two sales and $133.08 in gross commission, so you can pay up to $1.33 per click to break even before refunds.
Now apply a refund rate. At 15 per cent, which is unremarkable for this category, your two sales become 1.7 keepable sales and $113.12, so your break-even click price falls to $1.13. At 25 per cent it falls to $0.99. Refund rates in some supplement and make-money categories run higher than that, which is why the offer's refund rate is a more important number than its commission percentage and why you should never scale before you have observed one across a full refund window.
Then subtract the cost of running the operation: the $5 payout fee, tracking software, landing page hosting, and the campaigns that lost money while you found the one that did not.
The honest summary is that paid affiliate traffic is a business with thin margins, meaningful platform risk and a two-month lag between the number you see and the number you keep. It is not a beginner's path, and the content teaching it as one is usually itself a ClickBank offer.
Compliance is not optional
Two obligations apply regardless of what an offer's affiliate resource page suggests.
Disclosure. In the United States, the Federal Trade Commission requires that a material connection between you and a product be disclosed clearly and conspicuously, close to the recommendation, in language an ordinary reader understands. A disclosure at the bottom of a page, hidden behind a link, or written in legal language does not satisfy this. The UK, EU, Canada and Australia have equivalent requirements. This applies to a video description, a social post and an email, not only a website.
Claims. You are responsible for what you say, including when you are repeating the seller's copy. Health claims about supplements and earnings claims about business products are the two areas that attract enforcement, and affiliates have been the subject of action rather than only the sellers. Swipe copy provided by a seller is not a legal shield, and some of it is written to convert rather than to comply.
The safe practice is to describe what the product is, what it did for you if you used it, and what it will not do. That converts worse than a claim and it is the version you can defend.
The seller side
If you have a digital product, ClickBank offers something a payment processor does not: an existing population of affiliates who will promote it for a percentage.
The costs are documented. A one-time activation fee of $49.95 falls due when your first product is approved, and is not charged if approval is refused. There is no monthly fee. ClickBank buys each sale at 92.5 per cent of retail minus $1, and your affiliate's commission comes out of what remains, so a $100 product at 50 per cent commission leaves you $45.75. Handling refunds for up to ninety days and carrying the chargeback liability is part of the arrangement.
The reason to accept that structure is distribution. The reason not to is that a strong affiliate program needs an offer that already converts, because affiliates send traffic to what works and ignore what does not. Listing a product that has never sold and waiting for affiliates to discover it is the seller-side equivalent of waiting for a marketplace to supply demand.
Whether it beats selling directly comes down to one question: do you have traffic, or do you need it? A seller with an audience keeps more by selling direct. A seller with a converting funnel and no audience is buying distribution, and 7.5 per cent plus $1 plus a commission is a reasonable price for it.
A realistic timeline
Assuming the content route, which is the one most likely to still exist in two years:
- Months 1 to 3: you pick a niche, buy and evaluate several offers, build the asset, and publish consistently. Revenue is close to zero. The Customer Distribution Requirement means your first sales may not be payable yet.
- Months 3 to 6: the first sales arrive, the first payout clears, and your first refunds arrive to teach you the difference between reported and keepable earnings. A few hundred dollars a month is a good outcome here.
- Months 6 to 12: if the traffic source is working, this is where compounding starts. Content published in month two is now ranking, and the same work produces more.
- Year 2: the affiliates who are still doing this have an email list and a traffic source they control. The ones who do not have neither are usually gone.
On the odds, affiliate marketing has an extremely concentrated income distribution. A small number of operators with established traffic assets take a large share of the payouts, and the median participant earns very little. ClickBank publishes no audited affiliate income distribution, so treat any specific earnings figure attached to a named affiliate, particularly one appearing in a course sales page, as marketing rather than data.
Common mistakes
Modelling on the advertised commission. A 75 per cent commission pays about 67 to 69 per cent of retail after the wholesale calculation. Six to eight points is the difference between a profitable campaign and a losing one.
Scaling before a refund window closes. Sixty days is the standard, and a campaign is not proven until it has survived one.
Choosing offers by gravity alone. It counts affiliate activity, not quality or earnings, and it cannot tell a new offer apart from an abandoned one.
Promoting products you have not bought. You cannot judge the upsell sequence, the refund experience or the quality without going through it as a customer.
Ignoring the dormancy ladder. Withdraw your balance before you stop, because $50 per pay period after a year is a genuine cost and there is no notification.
Leaving the payment threshold at $100. The $5 payout fee is 5 per cent at that level and 1 per cent at $500.
Treating swipe copy as compliant. It was written to convert. Responsibility for claims sits with the person who published them.
Building on rented traffic only. A social account or an ad account can be gone in a day. The email list is the part you keep.
Where this is heading
Search behaviour is changing what content affiliates can rank for. AI answers absorb the simple informational queries that used to bring traffic to review pages, which compresses the value of thin comparison content and raises the value of material that demonstrates first-hand use. Affiliates whose content is a rewritten sales page have the most to lose.
Regulatory attention on health and earnings claims continues to tighten. These are the two categories where ClickBank's highest commissions sit, and enforcement has repeatedly extended to promoters rather than stopping at sellers. Expect disclosure and substantiation requirements to be enforced more consistently, not less.
Recurring offers become more valuable relative to one-off products. A rebill at 90.1 per cent wholesale with no dollar deduction is better arithmetic for the affiliate than a single sale, and the offers competing for good affiliates increasingly reflect that.
The platform's reputation stays bifurcated. ClickBank's technical infrastructure, tracking and payment reliability are mature after more than two decades. Its open catalogue means quality will keep varying, and the affiliates who do well are the ones who treat curation as their actual job rather than treating the marketplace as a shelf to point at.
Owned audiences keep winning. Every change above favours affiliates with an email list and a durable traffic source, and penalises those renting attention from a platform that can change its mind.
The niche decision determines most of the outcome, and it gets made badly because people start from commission size rather than from whether they can reach the buyer.
Start from the audience you can actually reach. There is some group whose problems you understand well enough to write something useful about without research: a job you have held, a condition you have managed, a hobby you have spent years on, software you use daily. That understanding is the only durable advantage available to a beginner, because everything else in this business can be bought.
Then check that the group buys things. Enthusiasm and spending are different. A niche full of people who solve problems for free is a poor place to earn commissions regardless of how large it is.
Then check the offer supply. A niche with one ClickBank product is a niche where you have no alternative when that product's funnel changes or its refund rate climbs. Two or three credible offers gives you room to move without rebuilding your content.
Then apply the filter that actually protects you. Buy the leading offer, go through the entire funnel including the upsells, and ask whether you would tell a specific person you know to buy it. If the answer is no, this is the wrong niche for you no matter what it pays, because you will either promote something you do not believe in or you will quietly stop working on it.
The categories worth being especially careful about are the ones that dominate the high-commission end. Supplements and health offers carry claim risk and high refund rates. Make-money-online products frequently teach the promotion of make-money-online products, which is a closed loop rather than a market. Survival and preparedness products convert well on fear, which is effective and not something everyone wants to build a business on. None of these are automatically disqualifying, and all of them deserve a harder look than a gardening or software offer paying half as much.
Building the asset
Content that earns affiliate commissions does one job: it helps someone who is already trying to solve a problem make a decision they were going to make anyway.
That rules out most of what gets published. A rewritten sales page adds nothing the reader could not get from the sales page. A list of ten products with identical descriptions helps nobody choose. Both were viable when search engines rewarded volume and both are declining assets now that answers to simple questions get generated rather than clicked.
What survives is material that could only have been written by someone who used the thing. Specific results, over a stated period, with the parts that did not work included. Comparisons where you have used both. Answers to the awkward questions the sales page avoids: what it costs after the upsells, how the refund actually goes, who it is genuinely wrong for.
Three formats do most of the work:
The problem article. Someone searches a symptom, a task or an error. You explain the problem properly, cover the options including the free ones, and recommend where a paid product is genuinely the better answer. Most of the article has nothing to sell in it, which is why it gets read and linked.
The honest review. You bought it, you used it, and you say what happened. The credibility comes from the criticism. A review with no downside reads as advertising, converts worse, and produces the refunds that claw back your commissions.
The comparison. Two or three real options with a clear recommendation for different situations. This captures people at the end of the decision, which is why it converts best and why it is also the most competitive.
Publish on something you own. A domain and a site you control survive a platform's change of policy. A channel or profile is a distribution route rather than an asset, and it should point at the thing you own.
Tracking, or you are guessing
ClickBank affiliate links support a tracking parameter, and using it separates affiliates who improve from affiliates who repeat themselves.
Set a distinct tracking ID for each placement: the specific article, the specific email, the specific video. Without it, twenty sales in a month tell you that something worked. With it, they tell you which two pieces of content produced eighteen of them, which is the difference between doing more of what works and doing more of everything.
Four numbers are worth watching, and only one of them is revenue.
Clicks per placement, which tells you what is getting attention. Conversion rate per placement, which tells you where attention turns into buyers, and which frequently disagrees with the click numbers in useful ways: a small article with high intent often outperforms a popular one. Refund rate per offer, measured over a full sixty-day window, which is the number that decides whether the revenue was real. Earnings per click, which is the figure that lets you compare a cheap offer that converts well against an expensive one that rarely does, and the only one that can be compared directly against a paid traffic cost.
Record them monthly. The pattern that emerges within three or four months is usually that a small minority of your content produces almost all of the income, and knowing which minority is the whole game.
Working out whether it is worth it for you
Run the calculation before committing months to this.
Pick a realistic offer in your niche and compute the actual commission after the wholesale deduction, not the advertised percentage. On a $97 offer at 75 per cent, that is $66.54. Then subtract a realistic refund rate. At 15 per cent, each sale is worth about $56.56 in money you keep.
Now decide how many sales a month you would need for this to be worth the hours. Thirty sales a month is roughly $1,700 kept, which is a meaningful side income. Thirty sales a month at a 2 per cent conversion rate requires 1,500 visitors a month with buying intent, which is a real content asset that takes six to twelve months to build, or a paid traffic operation with the margins described above.
That is the honest shape of it. The arithmetic works, the numbers are not fantastical, and the timeline is measured in quarters rather than weeks. Anyone presenting it as faster than that is usually selling the course.
Who should skip this
Anyone who needs income within three months should choose something with a shorter path. The content route takes two to three quarters to produce meaningful revenue, and the paid route requires capital you are prepared to lose while learning.
Anyone unwilling to buy the products they promote should not do this. Recommending something you have not used is how affiliates end up attached to offers that generate refunds and complaints, and the reputational cost lands on the recommender.
Anyone uncomfortable with the catalogue should look at other affiliate networks. The highest commissions on ClickBank sit in categories with genuine claim and quality risk, and a person who will not promote those is competing for a smaller and more crowded slice of the marketplace. That is a perfectly reasonable choice and it should be made deliberately rather than discovered later.
ClickBank compared with the alternatives
The network you choose determines the commission, the product quality and how much of the risk you carry, and the three main options differ on all three.
Amazon Associates pays low single-digit commission rates that vary by category, and converts extremely well because the buyer already has an account, a saved card and a delivery address. The cookie window is short, commonly twenty-four hours, so you are paid for purchases made almost immediately after the click. The trade is volume against margin: you earn a few per cent of a physical product, and you earn it on almost everything that person buys in the window, including things you never mentioned. Product quality risk is low because Amazon carries the customer relationship. This is the sensible default for anyone recommending physical goods.
ClickBank pays 50 to 75 per cent nominal, which after the wholesale deduction is roughly 45 to 69 per cent of retail, on digital products with high margins. Conversion is harder because the buyer is meeting an unfamiliar brand at a sales page. Refund exposure is high and lands on you through clawbacks. Product quality varies enormously and the curation burden is yours. The reason to accept all of that is that a single sale can be worth more than a hundred Amazon commissions.
Networks like ShareASale, Impact and CJ sit between the two, hosting brands with established reputations paying middling commissions, often on software and services with recurring components. Approval is per-merchant rather than automatic, which is friction at the start and a quality filter in practice.
The choice is usually decided by what your audience buys rather than by which network pays best. A woodworking audience buys tools, which is Amazon. A small-business audience buys software, which is the mid-tier networks. An audience with a specific problem that a course or a supplement addresses is where ClickBank's economics become compelling, and where its risks concentrate.
Running more than one is normal and sensible. The failure mode is choosing the network first and then hunting for an audience whose purchases fit it, which is the reverse of how this works.
Recurring offers deserve their own analysis
Most affiliates evaluate an offer on what a single sale pays. Subscription offers should be evaluated on what a customer pays over their life, and the two rankings frequently disagree.
The wholesale arithmetic already favours rebills. A rebill under $40 is purchased at 90.1 per cent of retail with no dollar deduction, against 92.5 per cent minus $1 on a one-off sale, so a $37 rebill at 75 per cent commission pays $25.00 while a $37 one-off pays $23.98. The gap is small on the first payment and it repeats.
The real difference is what happens afterwards. A $47 one-off product paying 75 per cent gives you $31.86 once. A $37 monthly subscription paying 75 per cent gives you $25.00 a month, so it passes the one-off product in month two and keeps going. If the average subscriber stays five months, the subscription is worth $125 against the one-off's $31.86, from the same click.
That makes retention on the seller's side into your revenue, which is a genuinely unusual position to be in. You are dependent on a product you do not control continuing to be worth paying for, and you have no visibility into its churn beyond what you can infer.
Two checks help. Buy the subscription and stay in it for a few months, which tells you whether the value regenerates or whether month three is a repeat of month one. And weigh the average sale value against the initial sale value in the marketplace listing: a large gap indicates the offer earns most of its money after the first payment, which is where your commission is going to come from too.
The trap is a subscription with an aggressive front-end discount or trial. Free trials are purchased at a wholesale price of $0.50 and dollar trials at $1.00, so your commission on the trial itself rounds to nothing. Everything you earn depends on conversion to the paid subscription and on retention after it, both of which are entirely in the seller's hands. Those offers can work well and they should be treated as a bet on someone else's operations rather than as a sale you have made.