I read the affiliate page of 252 developer and SaaS tools and recorded five terms for each one: commission rate, recurring or one-time, cookie window, minimum payout, and payout method. Every field carries the URL it came from and the date it was read. 483 of those 1,260 fields — 38% — are not published by the vendor at all. That number is the whole finding, but the shape of it is more useful than the size: Term Share of the 252 that never state it Minimum payout 65% Cookie window 52% Payout method 45% Recurring or one-time 23% Commission rate 7% The term almost everyone publishes is the commission rate. The three most often missing are the three that decide whether the money reaches you at all. Why the rate is the least useful number A percentage tells you nothing until you know how many times you get paid it. On a $100/month product: a $150 one-time bounty pays you $150, ever 60% for 12 months pays about $720 in year one, then stops 30% for the customer's lifetime pays $360 a year, indefinitely Same customer, three different outcomes, and the lowest percentage wins by a distance. Of the 252 programs, 143 pay recurring commission, 51 pay once, and 58 don't say which. The cookie window decides whether research-heavy traffic is worth anything Someone clicks your link, the company drops a cookie, and if they buy after it expires you get nothing. For anything people think about before buying — which is most developer tooling — this is the term that decides whether a post you wrote is an asset or a donation. The real range: Jasper at 14 days against TubeBuddy at 365, and two programs state that attribution never expires. Another 131 publish no window at all. The minimum payout is where small affiliates actually lose Most programs hold your earnings until they cross a threshold. Earn $80 in a year against a $250 minimum and you have earned nothing you can spend. Thresholds here run from $5 to $450, seven state no minimum, and 164 don't say. Payout method is the one to read first if you're not in the US Some programs pay by PayPal only. Others use Wise, Payoneer, direct bank transfer or EFT. Guides written for a US audience treat this as a footnote. If you're somewhere PayPal is awkward or expensive, it is the difference between money arriving and money technically existing. 113 of the 252 don't say how they pay. One counting decision worth explaining 152 of the 252 name no affiliate network. I deliberately left that out of the transparency count. A company that names no network may be running the program itself rather than concealing anything, and "in-house" and "undisclosed" are different facts. Inferring one from the other would mark every company that built its own system as secretive. So the network is published where a vendor states it, and counted nowhere. This is the kind of decision that makes or breaks a dataset, and it is usually invisible in a roundup post. "Not stated" is a finding, not a gap Where a vendor publishes nothing, the record says not stated rather than filling it with an estimate. Knowing that a company won't publish its cookie window is worth knowing before you build content around it. That rule is also why this took reading 252 pages instead of aggregating three listicles. A roundup whose author earns commission from the programs it ranks has an incentive to sort by rate rather than by accuracy. Nothing here earns a commission from any program listed. The data affiliateprogramterms.com — the full directory. 50 programs are free to read in full, each field with its source URL and check date. github.com/dave8172/affiliate-program-terms — the free tier as a generated README, if you'd rather read it in a repo. Every count in it is computed from the dataset rather than typed, so it can't drift from the records it describes. If you find a figure that's gone stale, open an issue with the vendor URL. Terms change without notice, which is the reason every field here carries a date.