Anyone can submit a program to this directory for free, and most submissions describe their own terms accurately. But "most" isn't "all," and the gap between what an email claims and what a program's own page actually states is exactly where our listings would quietly drift wrong if we stopped checking.
Over a recent batch of 25 submissions, here's what actually happened when each one got checked against its own source rather than taken at face value.
When the page says something different than the email
A finance research platform submitted its affiliate terms as 30% on the first subscription payment, $500 minimum payout. Its own program page said something else: roughly 45% of the first payment ("half after the card fee," by its own worked examples) plus 10% recurring, and a $100 minimum — not $500. The actual terms were better than claimed, which is unusual, but the point stands: we recorded what the page said, not what the email said.
An AI consulting referral program described its commission as recurring "while contracts remain active" — language that implies indefinite, lifetime revenue share. Its own page stated a 12-month cap from the customer's contract start date. Twelve months of recurring commission is still a solid program. It is not the same program the submission described.
A real estate marketplace submitted specific attribution and payout mechanics — browser local storage with no fixed expiration, a $1 minimum via Stripe Connect. Its public program page confirmed the headline commission rate (30%, rising to 40% for top affiliates) but said nothing at all about attribution or payout terms. We listed the confirmed rate and left the rest blank, rather than publish submission-only claims as though they were independently checked.
This happened often enough across the batch that it's now the default assumption going in: read the page first, treat the submission email as a lead, not a source.
What gets declined outright, regardless of how the terms read
Some submissions get turned down on category grounds before the commission rate even matters:
A real-estate "wholesaling education" program offered a $2,500 bonus for every student a referred affiliate brought in who went on to close a deal — tying affiliate payout directly to a recruited student's financial outcome. Same shape as the income-opportunity schemes we've declined before, just wearing a different industry's clothes.
A self-hosted crypto payment gateway, sold as source code to anonymous business buyers, paid out exclusively in USDT with no fiat option anywhere in the flow. Self-hosted, unlicensed merchant payment infrastructure is disproportionately built for businesses that can't get approved by a regulated processor — gambling, grey-market storefronts, worse. The commission terms were perfectly coherent. That wasn't the problem.
An astrology app, declined on sight, the same way every submission in that category has been — not because the terms were bad, but because it's not a category this directory covers, full stop.
A $12 "faceless content creator" playbook claimed 50% commission through Whop. The actual product page showed no affiliate program at all — no signup flow, no terms, nothing beyond a buy button. Declined not for what it was, but because there was nothing to verify.
The throwaway-hosting problem, twice
Two separate submissions arrived with no owned domain anywhere in them. One used randomly-generated Framer deployment subdomains for both its sales page and its "affiliate resources" page — the kind of address that can vanish or change without a trace. Another arrived in an email that the mail provider itself had already flagged as a likely phishing attempt, which made the hosting question moot. Neither got evaluated further, because an unverifiable address is a decline on durability grounds alone, independent of whether the underlying offer is genuine.
What actually got added
Roughly two-thirds of the batch passed cleanly: a Shopify customer-support AI tool, a headless CMS, a Dutch mineral-makeup brand operating since 2007, a handful of small Gumroad creators being upfront that their programs were brand new with no track record yet. Unremarkable programs, verified plainly, which is the entire point — the interesting cases are the exceptions, but most submissions aren't exceptions.
The actual value of this step isn't catching fraud, which is rare. It's catching the much more common thing: someone describing their own program slightly more generously than their own terms page does, usually without meaning to.