> ## Content Index
> Fetch the complete content index at: https://www.datadab.com/blog/llms.txt
> Use this file to discover other available public pages before exploring further.

# Should You Launch a B2B SaaS Affiliate Programme? What the 32 That Do Actually Pay
- URL: https://www.datadab.com/blog/b2b-saas-affiliate-programme-terms/
- Published: 2026-10-22T04:30:00.000Z
- Updated: 2026-10-22T06:53:13.000Z
- Description: Most B2B software companies run no public affiliate programme. If you are considering one, the terms that decide your payout are the ones nobody prints.
- Author: Amit Ashwini
- Tags: Partnership

*Most B2B software companies do not run one. The ones that do offer something narrower than the headline rate suggests, and the terms that matter are the ones nobody prints.*

Every few weeks a founder asks us the same question. Should we launch an affiliate programme, and what should we offer?

Usually they have seen a competitor announce one, or a partner tool has pitched them a dashboard. The question sounds simple. The answer depends on what the market actually does, and most founders are guessing.

So we read it. We went through every public affiliate and referral page across 479 YC-backed B2B software companies in October, for the [affiliate and referral benchmarks study](https://www.datadab.com/research/b2b-saas-affiliate-benchmarks), and recorded only the terms each company states.

Here is what the market looks like, and where a founder who launches one will be compared.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-01-hero.png)

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-02-rarity.png)

## Most B2B SaaS companies do not run one

Only 32 of the 479 companies state affiliate or referral terms on their own site. That is 6.7%, with a 95% confidence interval of 4.8 to 9.3%.

The rest are not hiding a programme. They simply do not have one.

Plenty of pages look like programmes at first glance and are not.

We found 22 reseller or agency tracks, which are channel partnerships with different economics. We found 20 job-referral pages, which are about hiring. And we found 13 articles about referral programmes, which tell you how they work without offering one.

Strip those out and the category is thin. A founder who assumes everyone has an affiliate page is wrong about most of the market. Roughly one company in fifteen states one.

That thinness is the first thing to take seriously. A rare offer can be a competitive advantage. A rare offer built badly is just an expensive way to announce that you have not thought it through.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-03-hard-to-find.png)

## Most programmes are hard to find

Even the companies that run one make it difficult to locate. Seventeen of the 32 programmes sit under a terms or redirect path, such as /terms/redirect. The rest are scattered across /partners, /affiliates, /partner-program, /referrals and a handful of other addresses.

No single convention dominates. Affiliates hide under partners, legal and solutions pages as often as under an obvious /affiliates address.

That has a consequence for both sides. An affiliate who cannot find your terms cannot plan a campaign around them, and a founder who buries the offer in a legal page is making recruitment harder than it needs to be.

If you launch one, give it a path a stranger can guess and link it from the main navigation.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-04-band-not-number.png)

## The headline rate is a band, not a number

The median headline commission in the sample is 25%. The interquartile range runs from 20% to 30%.

Most programmes do not publish one rate. They publish a band that moves with volume, plan or tier: 5 to 20%, 20 to 30%, 25 to 50%.

Webflow's band, 10 to 50%, is the widest in the sample, a forty-point spread. Jasper's, 25 to 30%, is the narrowest.

A minority publish a single rate. Close offers 30% lifetime, Podium 30%, Flowglad 50% and Bik.ai 70%.

Read a band as a signal rather than a price. A company that publishes 10 to 50% is telling you its top rate is reserved for someone it expects to bring a lot of business.

A first referral is unlikely to get the top of the range, and the page rarely says what gets you there.

The single-rate programmes are easier to model. The banded ones are easier to recruit from, because the top of the band sounds generous.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-05-recurring.png)

## Recurring beats one-off, and the duration decides it

Among the programmes that state how long a commission lasts, recurring terms outnumber one-off terms 11 to 4.

Lifetime and first-year terms dominate. Flat one-off bounties exist, and some are large. But they are the exception, and they suit products with short sales cycles and low ticket prices.

For a subscription product the arithmetic is simple. A 20% recurring commission on a customer who stays three years pays out more than a 50% one-off on the same customer.

The one-off looks bigger on the page. The recurring one is usually what the affiliate actually earns.

This is where founders most often misjudge. They compare headline rates, ignore duration, and then wonder why an affiliate with a real audience signs up to a competitor instead.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-06-cookie.png)

## Get the cookie in writing

A cookie window decides whether an affiliate gets paid for a sale that happened a few weeks after the click. It is the single most important term in the programme, and it is the one most teams leave blank.

Only three of the 32 programmes print a number. All three say 90 days: Webflow, Deel and Storylane. Everyone else leaves attribution unstated.

That means the 30-day and 60-day conventions you may have heard cannot be tested here, because almost nobody publishes them.

If you launch, you are choosing a cookie window and then choosing whether to tell people. Choose the number, print it and hold to it. An affiliate who cannot see the window cannot plan around it, and one who discovers later that it was 30 days will not forgive you.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-07-two-motions.png)

## Affiliates and referrals are different motions

The study separates two kinds of programme that people often merge.

Affiliates, usually creators, reviewers and newsletter writers, get percentage bands. Referrals, usually customers and partners sending a warm introduction, get a fixed bounty or an account credit. The two pay on different logic, and an offer designed for one will not recruit the other.

If you want a creator programme, the percentage and the cookie are the offer. If you want a customer referral programme, the bounty and the credit are the offer, and the reward has to be worth more than the friction of making the introduction.

Most early programmes try to do both on one page. That usually means the founder has not decided who they are recruiting.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-08-gates.png)

## Gates correlate with generosity

Seven of the programmes screen applicants before anyone is admitted. Close, Webflow, Deel, Noloco and Assembly all gate, and Deel runs its gate through a third-party partner platform. Tella and AssemblyAI apply approval at their discretion.

The more striking pattern is that every programme in the sample paying above 25% screens its applicants. Open-door programmes cluster at 10 to 20%. The gate and the rate arrive together, and the study reads that as an explicit trade: a higher rate buys selectivity.

We read it as quality control. A company paying a large recurring share wants partners who will send buyers who stay, not anyone who signs up for a link. An application is a cheap filter on exactly that.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-09-name-motion.png)

## Name the motion before you name the page

The word "partner" causes more confusion than any other term in this category, and the sample shows why.

Twenty-two of the "partner" pages we found are reseller, agency or technology tracks. They have no public commission. They are built around co-selling and certification, which is a different motion from affiliate marketing, where someone is paid for each referral they send.

If you run a reseller track and call it an affiliate programme, creators will read the page, find no rate, and move on.

If you run an affiliate programme and call it a partner programme, you will attract agencies looking for a co-sell arrangement and not the audience you wanted.

Name the motion on the page before you name the programme. The offer follows from the name, and the wrong name is the most expensive way to attract the wrong people.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-10-launch-with.png)

## What to launch with

If you have to pick one configuration from this sample, the modal stated offer is 20 to 30% recurring, a 90-day cookie and an application gate.

That is a defensible starting point, not a recommendation. Test it against three questions before you copy it.

Can you afford the recurring share for the full term of the average customer? A 25% commission on a customer whose margin you have not modelled is a loss you will discover in month nine.

Will you honour the cookie window when a sale is disputed? If you cannot, print a shorter number.

Who is your first affiliate? If you cannot name one, you are building a page for an audience you do not have. The gate will protect you from the wrong partners, but not from having no partners at all.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-11-who-for.png)

## Who this is for

\* *Founders deciding whether to launch.*\* Most of the market does not run one. A narrow, well-written offer beats a broad, vague one, and the rare-offer advantage only holds if the terms are good enough to repeat.

\* *Creators and affiliates reading a programme page.*\* Ask for the cookie window in writing before you promote anything. Only three programmes in the sample print one.

\* *Partnership and marketing leads comparing offers.*\* Compare duration and attribution before headline rate. A 50% one-off with no stated window can pay less than 20% recurring with 90 days.

\* *Partnership teams deciding what to call the offer.*\* Do not confuse a reseller track with an affiliate programme. The two need different terms, different pages and different people to run them, and the 22 partner-named pages show what happens when the label and the motion disagree.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-12-confidence.png)

## How confident we are

The study records what companies publish. It does not measure what those companies actually pay, how many affiliates they have, or how often the terms are negotiated away.

Stated terms can understate reality. A company may offer a custom rate to a large partner and say nothing about it on the public page. We counted only what was printed, and we say so wherever a number depends on that choice.

The sample is 479 venture-backed B2B software companies, weighted toward recent YC batches. Thirty-two programmes is a small number, and the confidence interval on the 6.7% headline reflects that.

Terms were read from first-party domains, with robots rules checked before each page was read. No company was contacted, and no terms were requested that were not already public.

![](https://www.datadab.com/blog/content/images/2026/10/b2b-saas-affiliate-programme-terms-13-wrap-up.png)

## Wrap-up

Most B2B software companies do not run an affiliate programme, and the ones that do rarely publish what they actually pay. That gap is where a well-built programme can stand out.

Build around duration and attribution, not the headline rate. Print the cookie window. Put the programme somewhere a stranger can find it. Gate the applications. And decide, before you launch, whether you are recruiting creators or customers, because the two offers are different.

Want to see whether your own pages state the terms a partner would need?

Start with the page an affiliate reads before deciding whether to promote you, then [scan it for the terms that are missing](https://www.datadab.com/tools/decision-friction-scanner).

Terms are as printed on 32 public pages in October 2026\. Stated terms can understate negotiated deals, and the study measures what is published, not what is paid.

[Scan your partner page](https://www.datadab.com/tools/decision-friction-scanner)