The six stages between a partner you have never met and a customer who renews, what counts at each one, where the number comes from and where it usually breaks.
A SaaS affiliate funnel has six stages: partner recruited, partner activated, click, trial, paid and retained. Write down the one event that counts at each stage before you launch, and take the last three from Stripe, not from the affiliate dashboard. The biggest drop is usually between recruited and activated: in Rewardful's data on 2,847 programs, only 7.6% of affiliates make a single referral. Read the funnel per partner and by monthly cohort, because a SaaS customer can take weeks to pay. When a number drops, fix that stage before you add more partners.
Most guides to affiliate funnels are written for the affiliate: pick a niche, build a page, send traffic. If you own the product, your funnel starts earlier. Before anyone clicks, a stranger with an audience has to become a partner, and that partner has to publish something. After the click, a visitor has to start a trial, pay and keep paying.
So the funnel has two partner stages and four customer stages. Each one needs a single event that counts, a place where that event is recorded, and one rate that tells you whether the stage works.
| Stage | The event that counts | Where it is recorded | Rate to watch |
|---|---|---|---|
| 1. Partner recruited | Terms accepted, tracked link and promo code issued | Tracking platform, outreach list | Contacted to signed, per partner type |
| 2. Partner activated | First live content with the link, or first tracked click | Tracking platform | Signed partners with traffic within 30 days |
| 3. Click | A visit or code redemption credited to a partner | Tracking platform, Stripe for codes | Clicks per active partner, share via code |
| 4. Trial | Account or trial created with the partner attached | Your app, Stripe subscription in trial | Click to trial |
| 5. Paid | First paid invoice, not refunded | Stripe | Trial to paid, refund rate |
| 6. Retained | Still paying after the first renewal | Stripe | Share of referred customers still active |
If stages 1 and 2 leak, nothing reaches stage 3. If stages 4 to 6 are not tracked, you cannot tell which partners bring customers and which bring curious visitors.
Rewardful's State of SaaS Affiliate Programs 2026 covers 2,847 active programs. Several of its figures line up with the stages above:
The same report says that, on average, about 10% of affiliates generate a program's affiliate revenue. Two things follow. The first drop is the steepest, so most programs lose more at stage 2 than anywhere else. And because so few partners carry the revenue, a funnel read as one total hides the partners that matter.
Treat these as a sanity check, not a target. They come from one platform's customers and its own definition of a referral. Your own history is the better benchmark after the first quarter.
What counts: a partner who has accepted your terms and holds a working link and code. An application is not a recruit, and neither is a reply saying "sounds interesting".
What to measure: contacted, replied and signed, split by partner type: comparison sites, newsletters, YouTube reviewers, LinkedIn creators, agencies and communities. The split shows which lists are worth building again.
Where it leaks: lists that are too short, applications from coupon and deal sites, and approvals that take a week. A few dozen messages rarely produce a working program; we plan for 300 to 500 contacts by day 30. Where to find each partner type is in the 8 partner types for SaaS.
What counts: the partner's first live piece of content with your link, or their first tracked click. Pick one and use it for every partner.
What to measure: the share of partners signed in a given month who are active within 30 days, and the days from signing to first click.
Where it leaks: this is the stage the Rewardful numbers point at. Partners sign, get a link and hear nothing more. The causes and the first 60 days of follow-up are covered in why most affiliates never refer, so here the point is only to count it. If this rate is low, recruiting more partners adds more silent ones.
What counts: a visit credited to a partner through their link, or a checkout where their promo code was used.
What to measure: clicks per active partner, and how many of each partner's customers came through the code rather than the link.
Where it leaks: the link breaks or the credit gets lost. People copy a link without its tracking parameter, privacy browsers block the cookie, and viewers watch a YouTube or TikTok video on a phone and type your URL later on a laptop. A code catches what the link misses. In Stripe a promotion code is a customer-facing code on top of a coupon, and many codes can share one coupon, so every partner can have their own. A partner whose customers arrive mostly by code is working even if the click count looks thin. That pattern is common with creator content.
First check: click every partner's link yourself, on a phone and on a desktop, and confirm the visit shows up under the right name.
What counts: an account or trial created with the partner still attached. In Stripe, a subscription on a free trial with a zero price sits in the trialing status until the trial ends.
What to measure: click to trial per partner. Inside the trial, also track the moment your product proves itself, such as the first report built or the first teammate invited. A partner whose trials never reach it is sending the wrong audience.
Where it leaks: two places. The referral can drop off between the landing page and the signup form, often when signup runs on a different subdomain or a separate app. And the landing page can ignore what the partner showed. If a reviewer demoed one feature, land their visitors on that feature, not on a generic homepage.
First check: sign up through a partner link in test mode and confirm the Stripe customer carries that partner.
What counts: the first paid invoice that is not refunded. Stripe sends an invoice.paid event when an invoice is paid, and its documentation suggests checking that the subscription is active before you treat the customer as paying.
What to measure: trial to paid per partner, revenue per referred customer, and refunds inside your refund window.
Where it leaks: trials that end with no payment method, an audience that liked the video but never had the problem, and codes that escape to coupon sites and pick up customers who were going to buy anyway. Stripe sends a trial_will_end event before a trial ends, three days ahead by default, which is a good moment for a reminder email. If one partner's trial-to-paid rate sits far below your average, the issue is usually fit, not the product. Leaked codes are covered in our 12-check audit of a dormant program.
What counts: a referred customer still paying after the first renewal. For monthly plans, a check at month three and month six works. For annual plans, the first renewal is the only real test.
What to measure: the share of each partner's customers still active, next to the same share for all your customers. Stripe sends a customer.subscription.deleted event when a subscription ends, so the data is already there.
Where it leaks: partners who promise features you do not have, and customers who came in on a deep first-months discount and leave in the month it ends. Many affiliate reports stop at the first payment, which hides both problems. Retention is also where commission terms meet your margin; SaaS commission benchmarks walk through that math.
A program can show plenty of trials and still lose money if most come from two partners whose users never pay. Put one row per partner and one column per stage. Then read each row by its weakest column.
| What you see | Stage that leaks | First thing to check |
|---|---|---|
| Many signed partners, almost no clicks | 2. Activated | Onboarding and follow-up in the first 30 days |
| Live content, few clicks, many code redemptions | 3. Click (credit, not traffic) | Nothing is broken; credit the code |
| Clicks but few trials | 4. Trial | Referral lost at signup, or a landing page that ignores the content |
| Trials but few paid | 5. Paid | Audience fit, trial length, payment method at trial end |
| Paid but high churn | 6. Retained | What the partner promised, and discount-driven customers |
| Paid customers the tracker never saw | 3 or 4 | Missing codes, links tested on one device only |
Timing matters as much as the rows. A SaaS customer clicks, starts a trial, compares tools and pays weeks later. If you divide October's paid customers by October's clicks, you mix people at different points in the funnel. Group clicks by the month they happened and read paid conversions for that group 60 to 90 days later. Rewardful's 0.8% figure is also measured over 90 days.
Once you have a few cohorts, the decisions are simple. Partners strong at every stage get more attention and better terms. Partners with clicks but no paying customers get one conversation about audience fit. Partners who never activate come off the list. How that weekly work is organised, including payouts and the commission ledger, is in our guide to affiliate program management.
It is much easier to define the stages on day one than to rebuild them from messy data later.
If you are starting from nothing, our 14-day launch plan for a Stripe SaaS puts these steps in order.
About Enclaverse. We run creator growth for B2B and prosumer SaaS: affiliates and micro-influencers on one tracking setup, plus creator sponsorships on YouTube, TikTok, Instagram and LinkedIn. Every partner gets a tracked link and a promo code in the client's own Stripe, commissions exist only on charges that show up there after the refund window, and the weekly report can be checked line by line against Stripe. Program Launch is for companies with no program yet, Creator Engine adds creators to the affiliate track, and Program Rescue restarts a program on PartnerStack, Impact or Rewardful that nobody runs, with 15+ partners onboarded and 10+ live in 90 days or the fee refunded. See the Affiliate & Creator Program, influencer partnerships or the fully managed option, or book a call.
Six stages: partner recruited, partner activated, click, trial or signup, paid, and retained. The first two happen on the partner side, the last four on the customer side. Each stage needs one event that counts, written down before the program starts.
Rewardful's 2026 report on 2,847 SaaS programs puts the average referral-to-sale rate at 0.8% over 90 days, or 8 paying customers per 1,000 referred visitors. Use it as a sanity check, then compare each partner against your own history.
Usually the step from recruited to activated. In the same Rewardful data only 7.6% of affiliates make at least one referral. Adding partners does not fix that; onboarding and follow-up do.
Long enough for a trial and a buying decision to finish. Group clicks by the month they happened and read paid conversions for that group 60 to 90 days later. A quiet first week says very little.
No. A tracking platform gives you partners, clicks and referrals. Stripe gives you trials, paid invoices and cancellations. One weekly report that joins the two by partner covers all six stages.
Sources checked on 14 November 2026: Rewardful, State of SaaS Affiliate Programs 2026, Stripe Documentation, using webhooks with subscriptions, Stripe Documentation, trials, Stripe Documentation, coupons and promotion codes.
Thirty minutes on your partners, your Stripe data and the stage that loses the most.
BOOK A STRATEGY CALL →gabriele.soprano@enclaverse.com · +45 50 38 44 15