Whop sells access, not files. That single distinction decides whether it is the right platform for what you are building, and it is the thing most comparisons get wrong by lining Whop up against Gumroad on fee percentages alone.
Gumroad hands a buyer a file and the transaction ends. Whop grants a buyer entry to something that continues: a Discord server, a Telegram channel, a members area, a software licence, a stream of picks or signals or lessons. When they stop paying, the access is revoked automatically. The product is the ongoing thing, and the platform's job is to keep the membership list and the payment status in sync without you touching either.
That makes Whop a fit for recurring communities and a poor fit for one-off downloads. If you are selling a $19 template pack once per customer, the access layer is overhead you do not need. If you are selling a $49-per-month room that people join and leave continuously, the access layer is most of the work, and doing it yourself means writing bots that add and remove roles, reconciling failed payments against a member list, and handling the person who charges back in month four while still sitting in your server.
Whop's published fee schedule, checked at \`docs.whop.com/fees\` on 21 August 2026, prices each function separately rather than charging one blended rate:
Several of those only apply if you switch the feature on, which is why the effective rate two sellers pay on the same revenue can differ by more than a point.
A number of third-party reviews state that Whop charges a flat 3 per cent platform commission on top of card processing, sometimes described as applying to sales gated through Discord or Telegram. Whop's own fee documentation and its network pricing page, both checked for this guide on 21 August 2026, itemise payments, payouts, fraud, optimisation and affiliate fees and show no separate flat platform commission.
Treat that gap as unresolved rather than resolved in your favour. One of those third-party posts also quotes a card rate of 2.9 per cent plus $0.30, which does not match Whop's documentation either, and a source that is wrong about the processing rate is not a reliable source for the commission. The practical instruction is the same one that applies to any platform whose own page and the internet disagree: model both scenarios, and get the answer in writing from the platform before you price anything.
The flat $0.30 is what moves those numbers. It is a fixed toll on every charge regardless of size, so it lands hardest on cheap tiers and on monthly billing, where you pay it twelve times a year instead of once. If a 3 per cent commission does turn out to apply to your account, add three points to every line above and check that the business still works.
Membership businesses fail in a specific and predictable way, and the arithmetic that predicts it takes about a minute.
A membership has two flows: people joining, and people leaving. If you add a steady number of members each month and lose a steady percentage of the ones you have, the two flows meet at a level and stop moving. That level is your ceiling:
Twenty signups a month against 10 per cent monthly churn gives 200 members. Not 200 in the first year and more later. Two hundred, permanently, until one of the two inputs changes. At the $39 tier that is about $7,530 a month net, and it is where the business sits forever if you keep doing exactly what you are doing.
Halving churn does the same work as doubling signups, and it is usually cheaper. Documented monthly churn for paid Discord communities commonly runs 5 to 15 per cent, which puts most rooms somewhere in that table.
The second thing the identity tells you is how much work goes into standing still. At 200 members and 10 per cent churn you lose twenty members a month, which is exactly your twenty signups. Every month you run hard enough to grow by twenty and end the month where you started. Founders describe this as a plateau and look for a marketing fix. It is not a marketing problem. It is the ceiling arriving on schedule.
Three consequences follow, and they should shape what you build before you build it.
Recurring pricing is a different exercise from pricing a download, because the number you pick is multiplied by how long people stay, and those two variables push against each other.
Start from what the member is actually buying, and be precise about the unit. A trading room is not selling education, it is selling the feeling of not missing a trade. A fitness membership is not selling workouts, which are free everywhere, it is selling the fact that someone notices when you stop showing up. Price the thing the member would be upset to lose.
Whop operates a marketplace where buyers browse listings by category. New sellers routinely assume this supplies demand. It does not, and building on that assumption is the most common way a launch produces nothing.
Marketplace listings distribute attention that already exists in proportion to signals the marketplace can see: member counts, review volume, revenue history. A listing with none of those sits below listings that have all of them. The marketplace rewards traction rather than creating it, which is a reasonable design and a poor plan for someone at zero.
Sellers who launch successfully almost always arrive with an audience they built somewhere else. The Whop page is the checkout at the end of a funnel that runs on YouTube, TikTok, X, a newsletter, or a free Discord that has been running for a year. The platform handles payment, access and cancellation, and it does that well. Supplying the people is your job, and it is the majority of the work.
The practical test before you build anything: can you name the specific place your first fifty members will come from, and have you already posted there? If the answer is a platform you have not yet joined, you are planning to solve the hard problem after solving the easy one.
Affiliates as a distribution channel
Whop's affiliate program is the part of the platform that most directly addresses the discovery problem, and it is underused.
The mechanics, per Whop's own documentation: you enable the program, affiliates receive a unique link, referrals are tracked automatically, and commissions are paid out without you administering anything. Whop charges 1.25 per cent per transaction for affiliate processing. You can run global affiliates drawn from Whop's network, custom affiliates on individually negotiated rates, and direct revenue-share arrangements with partners.
The economics deserve attention because affiliate commissions on recurring products behave differently from one-off sales. A 30 per cent commission on a single $99 sale costs you $29.70 once. A 30 per cent lifetime commission on a $39 monthly membership costs you $11.70 every month for as long as that member stays, which on a member who stays fourteen months is $163.80. That is the correct price for a customer you would not otherwise have acquired, and the wrong price for one who would have found you anyway.
Two structures avoid the second case. Pay a high commission for the first twelve months and nothing after, which pays affiliates well for acquisition without renting them your retention. Or pay a lower lifetime rate and be explicit that it is lifetime, which motivates affiliates to send members who stay rather than members who convert. Choose the structure before you recruit anyone, because reducing a published commission rate afterwards costs you the affiliates you already have.
Content Rewards
Whop runs a second acquisition mechanism, Content Rewards, which pays creators per view for posting clips that promote a product. A campaign is funded with a budget, priced at a CPM, and capped per creator and in total.
Observed campaigns on Whop's Content Rewards page carry CPMs around $0.75 to $1 per thousand views with maximum payouts per creator, and third-party guides describe a wider range across the marketplace. Treat those as examples of what campaigns have offered rather than a published rate card, because each campaign sets its own terms.
The model is genuinely useful for a specific situation: you have a product that demonstrates well in short video, and you would rather pay for views than for clicks. It is a poor fit where the product needs explaining, because a clip cannot carry a complex proposition, and clips optimised for views select for entertainment over intent. Budget it as a top-of-funnel experiment with a measurable cap, not as a substitute for having an audience.
Disputes are an operating discipline
The $15 dispute fee is the line in the schedule that quietly decides whether some businesses work.
A chargeback on a $39 subscription costs you the $39 and the $15, a swing of $54 against a product whose net contribution is $37.65. One dispute wipes out the margin on 1.43 other members. Three disputes a month against 200 members is a 1.5 per cent dispute rate and a $162 monthly cost, which is more than four members' worth of revenue evaporating into administration.
Payment processors also watch dispute rates as a risk signal independent of the fees, and a rate that climbs can put holds on funds or end an account. The early dispute alert at $29 exists to let you refund before a dispute is formally filed, which is worth using when the alternative is a dispute fee plus a rate that is being monitored.
Most disputes in access businesses come from three causes, and all three are preventable. Members forget what the charge is, because the billing descriptor does not match the brand they joined. Members cannot find the cancel button and use their bank as the cancel button. Members feel the product did not deliver what the sales page promised. The fixes are unglamorous: a descriptor that matches the product name, a cancellation flow that takes two clicks and does not fight the user, and a sales page that describes what is actually inside.
Refunding a complaint immediately costs less than disputing it. A $39 refund is cheaper than a $39 loss plus $15, and it does not touch the rate that gets your account reviewed.
Deciding what to sell
The platform question is downstream of the product question, and the product question has a short version: what do you know or have access to that a specific group of people would pay monthly to stay close to?
Recurring access works when the value regenerates. Live market commentary regenerates daily. A community of practitioners regenerates whenever a member posts a problem. A service that produces something each month regenerates by delivery. An archive does not regenerate at all, which is why courses sell better as one-off purchases than as memberships.
Validate before building. The sequence that wastes the least time is to find where the audience already gathers, spend a few weeks being useful there for free, and watch what people repeatedly ask for. Then offer the paid version to that specific group with a specific promise and a date. If twenty people will not pay for it when you describe it, building it will not change their answer.
The categories that dominate creator-commerce platforms are trading, betting, fitness, e-commerce coaching and creator education, and that concentration is not a recommendation. Those categories are crowded, they attract the highest volume of competing offers, and several of them attract regulatory attention. A narrower category with a clear buyer is a better place to start than a large one where you are the newest of four hundred rooms.
Everything the platform automates is also a dependency, and the dependency is real. Your members' payment relationships, access rights and cancellation flows live inside a private company's product. Terms change, categories get restricted, and accounts get reviewed.
The mitigation is portability, and it costs almost nothing if you set it up at the start. Own the email addresses of your members outside the platform. Keep the community itself on Discord or Telegram, which you control, and use the platform for payment and access rather than as the home of the relationship. Maintain a landing page on a domain you own so that traffic you have built points somewhere you keep.
None of that is a prediction about this platform specifically. It is the same discipline that applies to building on any intermediary, and the sellers who have been through one platform migration all describe the same lesson, which is that the list is the asset.
A realistic timeline
The honest version, assuming you are starting with a small existing audience rather than none:
- Months 1 to 2: the product exists, the page is live, and the first ten to thirty members arrive, most of them people who already knew you. Revenue is a few hundred dollars a month. The main output is learning what members actually use.
- Months 3 to 6: the acquisition channel either works or it does not, and this is when you find out. Churn becomes measurable for the first time because there is finally a base to lose. Many memberships end here, and the ones that end usually do so because the value did not regenerate.
- Months 6 to 12: if signups and churn have stabilised, the ceiling identity now tells you where the business is heading. Growing past it means a second acquisition channel, a higher price, better retention, or annual billing. This is the point where affiliates start to matter, because they add signups without adding your hours.
- Year 2 and beyond: the businesses that persist are the ones where retention improved rather than the ones where marketing got louder.
On the odds, the figure worth internalising is that well under 1 per cent of products on creator-commerce platforms earn $10,000 a month or more. The distribution is extremely concentrated, a small number of large sellers account for a large share of platform volume, and the median product earns very little. Whop is a private company and does not publish audited seller data, so treat any specific revenue figure attached to a named seller, including the ones circulating in guides and videos, as unverifiable marketing rather than a benchmark.
Common mistakes
Pricing at $9 to test the water. The fixed $0.30 makes the fee economics poor, and cheap tiers attract high-churn members. Test the offer, not the price floor.
Launching without an audience. The marketplace distributes attention to listings that already have traction. Build the audience first, in the place your buyers already are.
Leaving affiliates switched off. It is the one growth mechanism on the platform that costs nothing until it produces a sale.
Selling an archive as a membership. Once a member has consumed everything, the renewal has nothing behind it. Archives should be one-off purchases.
Treating churn as a retention-email problem. If the product stops being useful, the sequence will not save it. Fix what the member is paying for.
No email list. The members exist inside the platform until you take their addresses out of it. Do that from the first member.
A cancellation flow that fights the user. Every obstacle you add converts a cancellation into a dispute, and a dispute costs $15 plus the revenue plus your standing with the processor.
Assuming the fee is 2.7 per cent. It is 2.7 per cent plus $0.30, plus whatever optional features you enabled, plus possibly a platform commission you have not confirmed either way. Model the number you actually receive.
Where this is heading
Access commerce keeps consolidating onto platforms. Running payments, role management, dunning and tax compliance separately is more work than the margin saved, and the direction of travel for independent sellers has been toward platforms that bundle them. That trend favours the category rather than any one company in it.
Fee structures will keep unbundling. Whop's schedule already prices orchestration, billing, tax and fraud as separate lines rather than one blended rate. Unbundled pricing is honest and it is also harder to compare, so expect platform comparisons to get less reliable, not more, and expect the headline rate to become a less useful number over time.
Regulatory attention follows the money into the largest categories. Trading signals, betting picks and financial coaching are among the highest-revenue categories on creator-commerce platforms and among the most exposed to financial-promotion rules. Sellers in those categories should expect verification requirements and disclosure obligations to tighten rather than relax.
Retention becomes the differentiator as acquisition costs rise. When every seller has access to the same clipping marketplaces and the same affiliate networks, acquisition converges and the ceiling identity does the sorting. The businesses that compound are the ones with churn in the low single digits, which is a product property.
Portability becomes a selling point. Sellers who have been through one migration ask different questions of a platform than first-timers do, and the ability to export members, keep a domain and move payment rails without losing the relationship will matter more as the cohort of experienced sellers grows.
Choosing between Whop and the alternatives
The comparison is usually run on headline fee percentages, which is the least useful axis. What separates these platforms is what happens after the payment succeeds.
Whop revokes access automatically when a payment fails or a member cancels, and syncs that state into Discord and Telegram. That is the entire value proposition. If your product is a room, it removes the job you would otherwise be doing by hand every day.
Gumroad delivers a file and closes the transaction. For one-off digital products it is simpler, and simpler is correct when there is nothing to revoke. Its fee is a flat percentage with no per-transaction cent charge, which makes it better than Whop for genuinely cheap one-off items and irrelevant for recurring access.
Teachable, Kajabi and the course platforms own the content-delivery experience, with lesson players, progress tracking and completion certificates. If the product is structured curriculum consumed once, that infrastructure earns its cost. If the product is a live room, you are paying for a video player you will not use.
Stripe or Lemon Squeezy plus your own build gives the lowest processing cost and total control, and costs you the build. The access layer is the part people underestimate: granting and revoking roles, handling failed payments without ejecting a member who will pay tomorrow, reconciling a cancelled subscription against a Discord that says otherwise. That is weeks of work and permanent maintenance.
The decision reduces to one question. Does your product need access revoked when payment stops? If yes, a platform that does it automatically is worth its margin. If no, you are paying for machinery you do not need, and Gumroad or Stripe will cost you less.
One point about the merchant-of-record distinction, because it is often glossed over. Some platforms act as merchant of record and take on sales-tax and VAT liability across jurisdictions. Whop prices tax calculation and remittance as an optional 2 per cent per transaction where tax is collected. That is a real cost and also a real service, and whether it is worth two points depends entirely on how many jurisdictions your buyers sit in. A seller with buyers in one country should price that differently from one selling into the EU and the UK.
Setting it up: the decisions that matter
Most setup steps are mechanical. Four are not, and getting them wrong is expensive to reverse.
Your billing descriptor. This is the text a member sees on their bank statement, and it is the single largest driver of preventable disputes. It must match the name they think they joined. If they joined "Alpha Signals" and the statement says the name of your holding company, a proportion of members will not recognise it and will call their bank. Set this before your first sale.
Your payout method. The schedule makes this consequential. Next-day ACH costs $2.50 per payout regardless of size. Instant bank deposit costs 4 per cent plus $1.00, and crypto or Venmo costs 5 per cent plus $1.00. On a $5,000 monthly payout, next-day ACH costs $2.50 and instant costs $201. Paying 4 per cent for two days of speed is a decision worth making deliberately rather than by accepting a default. Weekly instant payouts on a small balance are the version of this that quietly destroys margin.
Which optional features you enable. Orchestration at 0.8 per cent claims to lift revenue by routing payments through multiple providers, billing automation costs 0.5 per cent, tax handling 2 per cent. Each is defensible and each is a percentage of everything. Enable them one at a time with a before-and-after number, because enabling all of them at launch means you never learn which one paid for itself.
Affiliate terms. Published commission rates are hard to reduce later without losing the affiliates who joined at the old rate. Decide between first-year and lifetime commissions before you recruit anyone.
Beyond those, the mechanical steps are quick. Create the account, connect the Discord or Telegram the product lives in, configure the tiers, write the store page, and enable affiliates. The store page matters more than its length suggests: it is the only thing standing between a visitor and a decision, and the version that converts describes what is inside, who it is for, and who should not buy it.
The four numbers to track
Most sellers track revenue, which is the number that tells you least about where the business is going. Four others do the work.
Monthly churn rate. Members lost in a month divided by members at the start of the month. This is the input to the ceiling identity and the most important number in the business. Measure it monthly from the first month you have a base, and separate trial-conversion losses from established-member losses, because they have different causes and different fixes.
Signups per month by source. The other input to the ceiling. Tracked by source, it tells you which channel to invest in. Tracked in aggregate, it tells you almost nothing, because a month of twenty signups from a channel you can repeat is a different business from twenty that came from one viral post.
Effective fee rate. Total fees divided by gross revenue, calculated from your actual statement rather than from the headline rate. This catches optional features you forgot you enabled, international card loading you did not anticipate, and the payout method draining more than you assumed.
Dispute rate. Disputes divided by transactions. Watch it as a rate rather than a count, because it is the rate that processors monitor and the rate that predicts an account review.
Those four, tracked monthly in a spreadsheet, will tell you what the business is going to do before it does it. Revenue tells you what it already did.
Tax and compliance
Two obligations catch new sellers, and both are ordinary rather than exotic.
The first is income tax on what you earn, which is your responsibility regardless of what the platform does. Payments arriving in a personal account are still taxable income, and a business that reaches a few thousand a month generates a tax bill large enough that discovering it in April is a genuine problem. Set aside a percentage from the first payout.
The second is sales tax and VAT on what you sell, which depends on where your buyers are rather than where you are. Digital services sold into the EU and the UK carry VAT obligations that begin at low thresholds, and US sales-tax treatment of digital goods varies by state. This is what Whop's optional 2 per cent tax handling addresses. If you sell into multiple jurisdictions, two points to have the calculation and remittance handled is usually cheaper than the alternative. If your buyers are concentrated in one place, handle it yourself and keep the two points.
Neither of these is a reason to delay launching. Both are a reason to keep clean records from the first sale, because reconstructing a year of transactions afterwards costs more than recording them as they happen.
Working out whether it is worth it for you
Run the numbers on your own situation before committing months to this. The exercise takes ten minutes and answers the question honestly.
Start with the audience you have today, not the one you plan to build. Count the people who would plausibly hear about a launch: newsletter subscribers, an existing free community, followers who actually see your posts. Conversion from a warm audience to a paid membership commonly lands in the low single digits, so take 2 per cent as a working assumption. A thousand engaged followers suggests roughly twenty founding members, which at $39 is about $753 a month net. That is the honest starting point, and it is smaller than most people expect.
Then apply the ceiling. If you can repeat that twenty a month and your churn settles at 10 per cent, the business tops out near 200 members and about $7,530 a month net. If churn settles at 15 per cent, it tops out near 133 members and about $5,010. Decide now whether that number justifies the hours, because it is the realistic outcome of the plan working rather than the plan failing.
Then ask the harder question: what would have to be true to beat it? Usually one of four things. A repeatable channel that produces more than twenty signups a month. A product people stay in for a year rather than seven months. A higher price justified by a narrower and better-served audience. Or affiliates who add signups without adding your hours. Any one of those changes the ceiling materially. None of them are marketing tactics, and all of them are decisions about what you are building.
If the answer to all four is no, the honest conclusion is that this is a few-thousand-a-month business, which is a perfectly good outcome to pursue deliberately and a disappointing one to arrive at by accident eight months in.
Who should skip this
Anyone starting from no audience at all should build the audience first and come back. The platform solves payment, access and cancellation. It does not solve attention, and attention is the constraint.
Anyone selling a one-off download should use a simpler tool. The access machinery is the reason to pay Whop's margin, and if you have nothing to revoke you are paying for it and not using it.
Anyone who cannot commit to showing up should skip memberships entirely. The product in a community business is partly your presence, and a room where the founder went quiet churns out faster than one that never launched. This is the most common reason these businesses end, and it is knowable in advance.
Anyone in a regulated area who is not prepared to handle the compliance should choose a different category. Financial signals, betting picks and health coaching all carry obligations that do not disappear because the sale happened through a platform, and the platform is not the party that regulators approach.