SaaS Affiliate Marketing: How to Start and Scale an Affiliate Program

Most SaaS affiliate programs do not fail because the commission was wrong or the software was bad. They fail because nobody treated the channel like a channel. A program gets launched, a page goes up, twelve affiliates sign up, two of them send a click, and six months later somebody asks whether affiliate marketing works for B2B SaaS at all.

It does. But it works the way paid or outbound works: you build the machine, you recruit deliberately, and you keep optimizing. This guide covers both halves of that job. Part one is how to launch a program properly. Part two is how to grow one into a channel that shows up in your revenue reporting. Part three is what to do if you already have a program somewhere else and it is underperforming.

It is written for three people: the founder or marketer launching a first program, the team migrating off a tool that stopped fitting, and the operator whose program technically exists but has flatlined.

What SaaS affiliate marketing actually is

SaaS affiliate marketing is a performance channel where independent partners promote your software and earn a commission on the revenue they bring in. In B2B SaaS the mechanics differ from ecommerce affiliate marketing in three ways that change every decision downstream.

• Commissions are usually recurring, not one-time. An affiliate who refers a customer paying $200 a month can earn on that subscription for months or for its lifetime. That is what makes the channel attractive to serious affiliates, and it is why a recurring commission beats a bigger one-time payout in almost every case.
• Sales cycles are long and multi-touch. Someone reads a comparison post, leaves, comes back through a newsletter three weeks later, books a demo, and buys the following month. Your attribution window and cookie duration have to survive that.
• Your affiliates are businesses, not hobbyists. The people who move B2B SaaS revenue are newsletter operators, agencies, consultants, review sites, and community owners. They have their own P&L and they will do the arithmetic on your program before they touch it.

If you want the definitions in isolation, the affiliate marketing glossary covers the terminology. This guide assumes you want to run one.

Before you start: three things that decide whether this works

These are the three misconceptions that come up on almost every call with a SaaS team about to launch. Getting them wrong is the single biggest predictor of a dead program.

1. Affiliates are not a magic bullet

The most common expectation is that you join a network, affiliates discover you, and they sell for you. That is not how it goes. You need product-market fit, a funnel that converts, and a website that does its job before affiliates can do anything. If you cannot convert your own traffic, affiliates cannot convert it for you. Experienced affiliates can tell within a minute of landing on your site whether it will convert, and they allocate their audience accordingly.

2. It is not a passive channel

Listing on a marketplace and waiting is not a strategy. Affiliate marketing needs the same active management as any other acquisition channel: recruit deliberately, optimize what converts, iterate on the offer. The programs that grow are the ones where somebody owns the number.

3. It will not get you your first customers

Pre-PMF companies sometimes try to use affiliates to find their first paying customers. It cannot work, because affiliates only earn when people buy. If nobody is paying yet, there is nothing to earn and no reason for a partner to invest effort. If you are at that stage, start with an in-app referral program and let your existing users validate the motion first.

There is a signal that tells you when you are ready, and it is not a revenue number. It is when people start asking you. The moment "do you have an affiliate program?" starts arriving unprompted from agencies, consultants and people who write about your category, the demand already exists and you are simply failing to capture it. That is the trigger.

If you are genuinely too early, do the manual version instead. Find five or ten agencies or consultants who already believe in the product and work with them by hand. A fair number of them will not even want a commission at that stage. They want a product that makes their clients happy and a direct line to someone who can fix things. Give them that now and they are your first real affiliates the day you do launch a program.

What a program that is failing actually looks like

There are two shapes, and the second is far more common than the first.

The first is impatience. One company set up its program, used AI recruitment to contact around 400 companies inside a single day, and told us three weeks later that affiliate marketing does not work. The expectation had been imported from paid search, where you bid on a keyword and the demand is already sitting there waiting. Affiliate marketing does not behave like that. You are selling someone else on selling for you, and that is a longer run.

The second shape is the one that quietly kills most programs. A company adds a link in the footer, builds a landing page, lists the program somewhere, and then waits. A few applications arrive. The numbers are small, nothing compounds, and six months later the conclusion is that the channel does not work for them.

The test is simple. If you were running paid ads, you would check the numbers every week and optimise them. If you were running cold outbound, you would test until something landed. So why set up an affiliate program and then watch it passively, expecting it to perform exactly as you configured it on day one? If you are passive about this, it is going to be very hard to make work.

What a realistic timeline looks like

Set expectations with your team before you launch, because the gap between expectation and reality is what kills programs in month four.

• First paid referral: typically 3 to 6 months. That is the normal range, not a warning sign.
• Slower end: 5 to 6 months, typical when a company relies purely on inbound interest from a network and does no active recruitment, especially with niche software.
• Faster end: 0 to 30K MRR in 3 months. That is the best case we have seen, and it came from aggressive recruitment combined with letting affiliates bid on the brand name in paid search.

The spread between those outcomes is almost entirely explained by recruitment effort. Programs that wait perform like the slow end. Programs that go and get partners perform like the fast end. You can see both patterns in the customer stories: TimelinesAI built the channel from zero, and Potion turned its own users into affiliates.

What affiliate revenue actually looks like once it works

We went through 10,596 affiliate-months across three real programs to see how affiliate revenue is actually distributed. The shape is not what most people expect, and knowing it in advance will stop you killing a program that is working normally.

Three findings.

In any given month, the top 20% of active affiliates produced 80 to 87% of that program's affiliate revenue. Over 90% of all affiliate-months produced no paid referral at all. And between 30 and 88% of revenue in a month came from affiliates who sent zero clicks that month, which is to say from referrals they made in the past that are still paying.

What that means for you.

First, most of your affiliates being quiet is not a symptom of a broken program. It is the normal shape of this channel. If you look at month four and see a long list of names with nothing next to them, you are looking at exactly what a working program looks like at month four.

Second, a small number of partners will carry the whole thing. The return on finding and keeping one more genuinely good affiliate is far higher than the return on adding another fifty signups. That should change where you spend your time, and it should change what your tiers reward.

Third, this channel compounds in a way paid does not. Revenue keeps arriving from work that was done months ago, which is why break even lands at 9 to 12 months rather than at month two, and why judging the program at month three tells you close to nothing.

Part 1: How to start an affiliate program

Ten steps, in the order you should actually do them. The sequence matters more than it looks: most of the programs that stall did steps 9 and 10 before steps 3 and 5.

Step 1: Decide whether you need affiliates, referrals, or both

These are different motions and they attract different people. A referral program rewards your existing customers for introducing people they know. An affiliate program pays third parties with an audience to promote you at scale.

Customers refer occasionally and in small numbers, but they convert well because the trust is already there. Affiliates send volume and need commercial incentives. Most mature SaaS companies run both, and they are not in conflict. The help center covers the differences, whether to run them together, and which to start with.

Referral and affiliate are two rungs on a longer ladder, and it helps to know which one you are actually building. The gradient is the point: the further down you go, the fewer partners you get, the better each one converts, and the more your own sales and onboarding processes have to bend to accommodate them.

• Referral Your own users, rewarded in product rather than cash. Dropbox giving both sides 10GB is the cleanest example there is. Highest Almost nothing, it runs in-app
• Affiliate Content sites, newsletters, YouTubers, comparison pages. Paid a share of revenue. High A few hours a week once set up
• Agency or consultant Implements and advises. Owns the client relationship. Low Sales handoff, account rules, sometimes a discount instead of a commission
• Integration or tech partner Builds with you and lists you. Lowest Roadmap time, joint support, real process change

Step 2: Work out who your affiliates actually are

This is where most programs go wrong, by recruiting individuals instead of businesses with access to your ideal customer profile in bulk. Skip anyone whose pitch is that they will share your link with their network. Look for four archetypes:

• Newsletter operators in your category. They have a repeat audience that already trusts them, and their economics work on recurring commissions.
• Agencies and consultants who implement software for clients. They recommend tools as part of their job and they carry unusual authority. Treat them as affiliates, not as prospects.
• Sites and blogs already ranking for the queries your buyers search. They have the traffic and the intent, they just need a reason to feature you.
• Community owners with a Slack group, forum, or private membership in your niche.

There is a practical shortcut here: find out who is already promoting your competitors. Those people have proven they can sell software like yours to an audience like yours. Our guide to finding affiliates for your SaaS covers the sourcing methods in detail.

Agencies and consultants are affiliates, but they behave nothing like content affiliates, and a program built only for content affiliates will lose them. They own the client relationship, and that has two consequences you will discover the expensive way if you do not plan for it. Some will not want you contacting the end client directly at all, which changes your onboarding, your product emails and your sales handoff. And some would rather their client received a discount than take a commission themselves, because a kickback they have to disclose complicates the advice they are being paid to give. Build both options in from the start, a commission or a client-side discount, and let the partner choose.

Money is also not the strongest lever with this group. A two way lead flow, where you send them implementation work as well as taking referrals from them, is the most effective retention mechanic available to you, because a partner who receives leads from you cannot leave cheaply. Co-marketing does similar work: a joint webinar, a newsletter mention in both directions, a case study with their logo on it.

This is also the honest reason content affiliates sit at the bottom of your commission table rather than an arbitrary one. They never touch the client. The agency does the implementation, the support and the renewal conversation, so it earns a different rate. If you want a sense of the range of partner types operating in B2B right now, we broke down 16 B2B affiliate marketing examples by how each one actually works.

Step 3: Set commissions from the affiliate's side of the table

Founders usually set commission by starting from their own margin and working out what they can afford to give away. That produces a rate nobody wants. Start from the other side instead.

Run this calculation:

Monthly ARPA x commission % x number of months paid = what one referred customer earns an affiliate. Multiply by 10 customers.

Now look at that number and ask honestly: would you recommend your own product to your audience for that amount of money? If the answer is no, your rate is too low, and no amount of recruiting will fix it. The affiliate commission calculator does this arithmetic for you, and the EPC calculator shows what your program looks like in the earnings-per-click terms affiliates actually compare on.

How generous to be depends on your stage:

• Early stage: give more. You have no brand awareness, so the affiliate has to do more convincing, and you need leverage against established competitors.
• Established: you can pay less. Brand recognition means affiliates will recommend you anyway.
• Niche ICP: pay more, or pay for longer. Fewer affiliates have access to your audience, so you have to make it worth their specialization.

The three mistakes that show up again and again: no tiering so top performers have nothing to climb toward, setting rates from your margin instead of the affiliate's perspective, and early-stage companies paying less than established ones, which removes any reason to take a chance on an unknown brand. The help center has a full breakdown of what a good commission rate looks like in B2B SaaS.

Step 4: Choose your commission model and cookie window

Four models, and the choice signals what kind of partner you want:

• Recurring: a percentage of the subscription for a fixed number of months. The standard for B2B SaaS and the easiest to recruit against.
• Lifetime: recurring with no end date. A real recruiting weapon when you have little else to offer, and the single hardest term to walk back once affiliates are signed. See what to default to below before you choose it.
• One-time: a single payment per customer. Simple, but it competes badly against recurring offers.
• Hybrid: an upfront payment plus a recurring tail. Useful when affiliates need cash flow sooner than your payback period allows.

On cookie windows: B2B software gets researched slowly and often by committee. A 30-day window will silently lose you conversions that an affiliate genuinely earned, and affiliates know it. Ninety days is a reasonable default for considered purchases. The help center explains how recurring versus one-time commissions work in practice and why recurring commissions motivate affiliates.

What to default to, and when lifetime is the right call

Most B2B SaaS programs land somewhere between 20 and 50% of the subscription revenue. Where you sit in that band is mostly a function of how well known you are. Early on you are asking someone to spend their audience's attention on a product nobody has heard of, so you pay a higher percentage and you pay it for longer. Once you are established, 20 to 30% for 12 to 24 months is the normal shape, and it is the one I would default to.

Lifetime recurring commission is the exception, not the default. It is a genuine recruiting weapon when you have nothing else to offer, and plenty of programs have launched on it deliberately. But be clear about what you are signing. You are committing a fixed share of that customer's revenue for as long as the customer stays, on an account you will keep serving, supporting, onboarding and upgrading for years after the affiliate's work is finished. That sits in your CAC permanently, and it is the one term you cannot quietly walk back once affiliates have signed up on it.

If you do launch on lifetime, design the exit at the same time you design the launch. The pattern that works is this: start high, then as the company grows, lower the rate and shorten the duration for new affiliates, and move your original affiliates into a private tier where they keep exactly the deal they signed. Nobody loses anything they were promised, your commission cost stops scaling one for one with revenue, and you keep the partners who took the early risk. Trying to do this later, without a private tier, is how programs end up in a public falling out with their best affiliates.

Set the cookie window against your actual sales cycle

Now pick the cookie window, with your real sales cycle in front of you. This is the connection almost every new program misses. If your median time from first touch to paid subscription is three months and your cookie window is 30 or 60 days, you have built a program where the affiliate does the work, the buyer takes the normal amount of time to decide, and the attribution expires before the money arrives.

The affiliate will not complain about this, because they cannot see it. They will just conclude that your product does not convert and quietly stop sending traffic. You will conclude the same thing about them.

So: look up how long your deals actually take, from first touch to first payment, and set the window longer than that. For anything involving a trial plus an internal evaluation, 90 days is a sensible floor. If you genuinely cannot extend it, accept that you will keep having the pay per lead conversation with every serious affiliate you talk to, because a short window is usually the real reason people ask for lead based payouts in the first place.

Work out what the program actually costs you, before you set the number.

Add up twelve months of everything: the time someone spends running it, valued honestly at what that person costs you, the platform fee, any prepayments or bonuses, materials, and the commission you expect to give away and for how long. Divide that total by the number of paid referrals you realistically expect in the same twelve months. That is your program CAC, and it is the only number that tells you whether the commission you just picked is affordable. Judge it against LTV, not against what competitors advertise on their program pages.

Two anchors so your estimate is not built on nothing. Break even on this channel usually lands at 9 to 12 months, and the first paid customer typically arrives in month 5 or 6. A twelve month view is the shortest honest one. You can sanity check your own numbers with our affiliate revenue timeline calculator.

Step 5: Get tracking right before you recruit anyone

I was an affiliate before I built anything for affiliate programs. Through my own site I promoted around 80 SaaS tools over a few years. One of those programs I sent more than 15,000 clicks to. Not one attributed sale ever came back.

I do not think anyone was cheating me. I think their tracking was broken, and nobody on their side ever had a reason to go and look, because from where they sat the program was simply producing nothing. That is what broken tracking does. It does not generate a complaint. It generates silence, and then it generates an affiliate who quietly decides your product does not convert and never comes back.

So before you recruit a single person, run this check.

• Does signup happen on the same domain the affiliate links to? If your app lives on a subdomain or a separate domain, the referral parameter has to survive the jump.
• Where does checkout happen? If payment is hosted by Stripe or another billing vendor rather than on your own site, confirm the referral ID is passed through to the payment record and comes back on the webhook.
• Does the referral survive a real session? Cookie restrictions in Safari and other browsers can shorten or drop cross-domain cookies well before your stated window. Test it, do not assume it.
• Run a real signup yourself, end to end, through an affiliate link, and confirm it lands attributed in the dashboard. Then do it again on mobile.

If you fail this check, stop. Fix it before you do anything else in this article. Recruiting affiliates into broken tracking does not cost you nothing, it costs you the affiliates.

Settle your attribution rules in writing at the same time, while nothing is at stake. First click or last click. What happens when a paid ad and an affiliate both touch the same account. Which affiliate wins when two of them touch it. How long a click counts. If you want to think properly about multi-touch attribution before you write those rules down, our podcast episode with Steffen Hedebrandt of Dreamdata is a good place to start [VERIFY-PODCAST-URL].

Do this before you talk to a single affiliate. Nothing destroys a program faster than partners who believe they sent conversions you did not record. Trust, once lost, does not come back, and affiliates talk to each other.

What needs to work:

• The tracking script fires on your site and survives your signup flow. See the full installation guide and installing the tracking script.
• Payment data flows in, so commissions calculate on real revenue rather than estimated deal sizes. Connecting Stripe covers the standard path; anything non-Stripe runs through the API.
• Self-hosted or app-based signups are handled with the signup snippet.
• You have decided how to treat assisted conversions and what your attribution rules are, before anyone disputes one.

It is also worth choosing tracking that does not depend on third-party cookies. Cookieless tracking using your own user IDs is more accurate, and it is a meaningful selling point when your buyers have a privacy or security review.

Step 6: Build a program page that convinces professionals

Your affiliate program page is a sales page aimed at people who evaluate programs for a living. It needs the commission rate, the model, the cookie window, the payout threshold and method, who the product is for, and what proof exists that it converts. Vague pages get skipped, because an affiliate who cannot calculate their expected earnings will go and read a program page that lets them.

What an experienced affiliate checks before they read your commission

Before the rate matters, professionals are reading your company for signs that it is real. This is the same check we run on a prospect's site during a first call, and it takes about fifteen seconds.

• Above the fold. Customer logos, review scores, any social proof at all. Companies without paid customers have none of it, and the section usually looks unfinished because there is nothing to put there yet.
• The menu bar. A thin nav says there is not much behind it yet.
• The footer. A full footer with social profiles, a privacy policy, terms and a DPA is the clearest signal that a company has been operating for a while. An empty one is just as clear in the other direction.

None of this is about design taste. An affiliate is deciding whether the traffic they send you will convert, and a site that looks early is a site that probably will not convert their audience yet.

Step 7: Onboard affiliates like they are your first customer

The guiding question for the whole program is: how can I help my affiliates make money? Onboarding is where you answer it. Treat the first affiliates the way you treated your first customers, manually and attentively, and then automate what you learn.

Different partners need different things. A newsletter operator needs copy angles and a unique link. An agency needs a technical explanation and a case study. A review site needs product access and screenshots. Set up automated onboarding emails so every new signup gets something useful in the first 48 hours instead of silence.

Give every approved affiliate a free account on your product, or a heavy discount if a free one is not possible. Affiliates who actually use the thing write from experience, and it shows in the copy and in the conversion rate. The ones working from a feature list write something that reads like a feature list.

Step 8: Sort out payouts before the first commission is due

Decide your payout threshold, your schedule, and your hold period for refunds. Then automate it. Manual payouts are the hidden tax on affiliate programs: they consume finance time every month and a single late payment tells your best partner to prioritize someone else. See how automated payouts work and how to pay out affiliates.

Set the payout threshold from the affiliate's side

Thresholds of $200, and sometimes $500, are still common. They are a mistake. Even $100, which sounds modest from where you are sitting, usually means an affiliate has to generate five to ten commissions before seeing any money, and for someone who joined last month that is a very long wait for proof that your program pays at all.

Set it so an affiliate gets paid after two or three commissions. The first payout is the moment the relationship becomes real, and pulling it forward buys you more activity than almost anything else you can do at that stage. The mistake is setting the threshold around your own convenience without thinking about the person waiting on it.

The same logic applies to how you pay. The moment you say you only pay out one way, in one currency, you start losing good affiliates, agencies and consultancies included.

Write the clawback rule before you launch, not after the first refund

Decide in advance, publish it in the terms, and then apply it identically to everyone. What happens when a customer refunds inside your refund window. What happens on a chargeback. What happens when a customer downgrades, which normally means the commission follows the new subscription value from the next payment onward. And whether commission stops or continues if the customer churns partway through the commission period.

The practical version is a holding period: commissions become payable once the refund window has closed, not on the day the payment lands. Affiliates accept that easily when it is written down in advance. What destroys trust is handling the first refund case by case, in an email, six weeks after they were told the money was theirs.

Sort out invoicing and tax before the first payout run

This is where first payout runs actually go wrong, and it is worth twenty minutes with your accountant before you need it.

Decide whether you self-bill, meaning you generate the invoice on the affiliate's behalf and they approve it, or whether each affiliate invoices you. Self-billing scales, individual invoices do not. Collect tax details at approval rather than at payout: a W-9 for US persons, a W-8BEN or W-8BEN-E for non-US affiliates where a US entity is paying. For affiliates in the EU, know whether they are VAT registered and where, because the treatment differs between business and private affiliates and between inside and outside the EU. Set a minimum payout threshold, say plainly what happens to balances below it, choose your payment rails with cross-border fees in mind, and publish a fixed payout date.

Affiliates tolerate slow far better than they tolerate vague.

Step 9: Recruit your first ten affiliates by hand

Nobody discovers a new program on their own. Your first partners come from outbound, in roughly this order of difficulty:

1. Your own customers. They already use the product and can speak to it credibly. An in-app referral program surfaces the willing ones automatically.
2. Your competitors' affiliates. Proven ability to sell into your market. They are findable through the content that ranks for your category.
3. The four archetypes from step 2, approached individually with a specific reason why their audience fits.
4. A marketplace or network, which is how you get discovered rather than doing all the finding yourself.

Send a proposal, not an invitation. Name the audience overlap, state the commission, and show what a realistic month looks like in money.

The moment one of those first ten produces a result, write it up. Not as a customer case study, as an affiliate case study: what they published, how long it took, what it has paid them since. Partners with real audiences will not gamble their time on a program with no track record, and this is the only asset that answers that objection. Ten hand-picked affiliates become fifty because the first two gave you something to show the other forty eight.

Step 10: Decide what you will measure

Pick your metrics now, so that month three is a diagnosis rather than an argument. Track leading indicators separately from lagging ones. Leading: affiliates recruited, affiliates activated, clicks generated. Lagging: trials, paying customers, revenue, and commission cost as a share of that revenue.

The most useful early number is not revenue, it is the activation rate: the share of signed-up affiliates who have sent at least one click. It tells you whether the problem is recruitment or enablement, and those need completely different fixes.

Measure revenue, not signups

Clicks matter and signups matter, but neither one tells you whether an affiliate is worth having. An affiliate can send you a pile of referrals that are all personal email addresses and never become paying customers. Judge partners on revenue generated, or on earnings per click if you want to compare partners working at different scales.

When the channel gets reported upward to a CEO or a board, one number belongs on the slide: revenue from the channel, and whether it is going up or down. Everything else is context for the people running it.

Part 2: How to scale an affiliate program

A launched program and a scaled program are different problems. Launching is about getting the machine built. Scaling is about turning a handful of active partners into a channel with a number attached to it.

If your program has stalled, check these four things in this order

Most stalled programs are diagnosed as a recruiting problem and are not one. Work through these in order and stop at the first one that fails.

1. Is tracking working right now? Not was it working at launch. Run a real signup through a real affiliate link today and confirm it lands attributed. Product releases, domain changes and checkout migrations break attribution silently.
2. How many of your affiliates have ever sent a single click? Compare that to how many signed up. If the gap is large, you have an activation problem, not a recruiting problem, and adding more names will make the ratio worse rather than better.
3. Of the affiliates who did send clicks, how many reached a paid referral, and when did you last speak to any of them? These are your top 20%. If nobody has contacted them in three months, start there this week.
4. Only now, recruit. If tracking works, a healthy share of affiliates are sending clicks, and your best ones are being looked after, then volume is genuinely your constraint and the recruiting section below is where you should be.

Set a target as a share of new revenue, then work backwards

Do not set a target like more affiliates. Set it as a share of new MRR, then work out what has to be true. If you want the channel to contribute a meaningful slice of new revenue, the arithmetic is:

Target new MRR from affiliates / average MRR per referred customer = customers needed per month. Divide by the average customers an active affiliate sends per month = active affiliates needed.

Run that once and the strategy usually becomes obvious. Most teams discover they do not need hundreds of affiliates, they need a couple of dozen genuinely active ones, which is a completely different plan from what they were about to execute. The affiliate revenue timeline models how that ramp plays out over the first year.

Fix activation before you recruit anyone else

In almost every program, a small minority of affiliates produce nearly all the revenue, and a large majority never send a single click. The instinct is to recruit more. That is usually wrong, because you are about to add more people to the group that does nothing.

Work the existing list first:

• Segment by behavior: never sent a click, sent clicks but no conversions, converting. Each needs a different intervention.
• For the silent group, the blocker is usually that they do not know what to publish. Give them the angle, not a logo pack.
• For the clicking-but-not-converting group, the problem is fit or landing page. Send them to a page built for their audience rather than your homepage.
• For converting affiliates, ask what would let them do more, then remove that obstacle. This is the highest-return conversation in the whole program.

The fastest way to wake up a program that already has affiliates

A program with 300 signed affiliates and 12 active ones is the most common shape we see. Once you have confirmed the funnel itself converts, there are three moves, in this order.

Run a campaign at all of them at once. Offer a menu of actions with a bonus attached: a blog post, a listicle, a newsletter mention, a video. You are spending money upfront, but the arithmetic works. If only 5% of 300 affiliates write a listicle, that is fifteen new articles where you can be the top entry. You have bought search and AI-answer visibility as well as fifteen newly active partners.

Study the twelve who already work. Everyone stares at the inactive list. The active ones are the ones with something worth copying, and what they do can be published back to the rest of the program as proof, even as a single line: this partner did that, and earned this much from it.

Make your active partners scalable. Consultants and agencies usually promote one to one, inviting their own clients and nothing more, with nothing public anywhere. Their conversion rates are already good, which is exactly why this is the quick win. Ask for a newsletter mention, a listing on their site, or an article you supply. They get leads out of it and you get backlinks and exposure.

One tactic we deliberately avoid: leaderboards. They publish personal information about your partners, which is why Reditus does not offer them.

Recruit at volume once activation works

Now scale the top of the funnel. Four channels, roughly in order of effort:

• Marketplace listing. Being discoverable by affiliates already looking for programs in your category. This is inbound rather than outbound, and it compounds.
• A network with existing affiliates. Reditus has a network of 26,000+ B2B SaaS affiliates, which is a different proposition from tracking software that leaves recruitment entirely to you.
• AI-assisted search against a database of potential partners, filtered to people with real access to your ICP.
• Targeted outbound, which never stops being effective for the top 20 partners you actually want.

If you want to see what a well-built program page looks like from the affiliate's side, browse the affiliate program directory. It is also the fastest way to benchmark your terms against everyone else competing for the same partners.

Find the affiliates who already promote your competitors

Someone already publishing affiliate content about your competitor knows the model, knows the category, and already ranks for the searches you want. They are the highest-yield list you can build, and building it is a repeatable process rather than a hunch.

1. Put your two or three closest competitors into a backlink tool such as Semrush or Ahrefs.
2. Filter their referring domains to sponsored and nofollow links, and to pages that look like reviews, comparisons, alternatives and best-of lists.
3. Open the links and read their structure. Affiliate links carry a tracking parameter or route through a redirect domain, and the pattern usually tells you which affiliate platform the competitor runs on, which in turn tells you what that publisher is used to.
4. Export the matching referring domains and dedupe them against your existing affiliate list.
5. Enrich for a named contact with something like Hunter.io or Snov.io. You want the person who writes the posts or owns the site, not a generic inbox.
6. Pitch on the only thing they care about: what they would earn on your program versus the one they are already in, and what you will give them that the competitor does not. Product access, data they cannot get elsewhere, or an angle nobody has written yet.

This is the manual method and it sits alongside a marketplace listing rather than instead of it. Do it because your own network runs out somewhere around month three or four. That is the point at which recruiting has to stop being who you happen to know and start being a list you work through.

Use tiers and campaigns to move the middle

Flat commission rates give your best affiliates no reason to push harder. Tiered commissions fix that by raising the rate as volume grows, which converts your good partners into committed ones. You can also change an affiliate's tier individually when you negotiate.

Campaigns are the other lever, and they are underused. Instead of paying only on conversion, pay a fixed amount for a specific action: a review post, a newsletter mention, a comparison article. This works particularly well for partners with real audiences who will not gamble their time on a program with no track record. Setting up campaigns takes a few minutes, and one-off commissions handle the negotiated exceptions.

Tiers that unlock on revenue alone reward the people who were going to perform anyway. Go back to the distribution numbers: if the top 20% of active affiliates in a month produce most of the revenue, then a ladder with revenue-only thresholds offers the other 80% nothing they can act on this week. It is a scoreboard, not a mechanism.

So build at least half your unlocks around behaviour. Do X, get Y, where X is something any affiliate can do in an afternoon. Leave a review on G2 or Capterra. Publish a post that mentions the product. Complete the partner training. Include you in a newsletter send. Each one moves a dormant affiliate one step closer to their first click, and the review unlocks quietly solve a second problem for you at the same time.

One campaign worth running from day one, because the first 30 days decide whether an affiliate ever becomes active at all: make your first sale within 30 days of joining and your commission rate goes up five points. Announce it in the welcome email and repeat it in the dashboard.

Help your affiliates win the places buyers actually look

This is the highest-leverage thing you can do and almost nobody does it. Your affiliates are producing comparison posts, listicles, and reviews. Those pages are also what AI assistants read when someone asks which tool to use. When you help an affiliate produce genuinely good content, you get the referral traffic and a durable citation in the answers your buyers are getting elsewhere.

Practically: give affiliates real data, honest comparisons, and specifics they cannot get from your homepage. Our research on AI citations covers how third-party content shapes what assistants recommend, and the AI citation finder shows where your category is being cited today.

Keep quality control on from day one

Scale attracts the wrong kind of partner. Decide early where you stand on brand bidding, coupon and deal sites, and cookie stuffing, and write it into your program terms. Automatic fraud detection handles the obvious cases; the judgment calls are yours. Reviewing applications rather than auto-accepting everyone costs a little growth and saves a lot of cleanup.

Vet at the application form, because it is the cheapest place to do it. Four questions get you most of the way:

• Are you already using our product?
• Do you already have clients or readers who need this?
• What is your monthly traffic, or the size of your list?
• Why are you applying to this program specifically?

The fourth one does most of the work. Anyone who answers it in a sentence that could apply to any program in any category is not going to publish anything.

The trade-off is real, so mind it: every extra field costs you applicants, including good ones who are busy. Four questions and a URL is about the limit before serious people close the tab. If you need harder gates, a traffic minimum or a requirement that they already work in your category, state them on the program page rather than burying them in a form nobody finishes.

Who should actually own this

In the programs that work, someone owns it, and that someone is not the founder. It is the clearest pattern across our customer base.

A founder already has too much on their plate, and an affiliate program is indirect marketing: you are not talking to customers, you are managing the people who talk to customers. Founder attention belongs on the direct channels. The owner wants to be a growth manager or an affiliate marketing manager, and the larger the company gets, the more it should be a dedicated affiliate or partner marketer.

Not everyone can afford that, and plenty of programs are too small to justify a full role. So the rule is not hire someone immediately. It is stop having the founder run it as soon as that becomes possible.

The reason the role matters is that this channel compounds. Early on it takes real work and returns very little, which is precisely the point at which a busy founder quietly abandons it. Someone who already runs several acquisition channels recognises that shape, and can keep the other channels producing while this one builds underneath them.

The objections you will actually hear, and how we answer them

We went back through 55 sales calls and counted what people push back on. What follows is ranked by how often each one came up, because the order tells you what to prepare for first. If you are building the internal case for a program, this is the section to steal.

One thing is missing from the list on purpose. The joint most-raised item across those 55 calls, in 9 of them, was tracking blocked because signup or checkout sits off the company's own domain. That is not an objection, it is a qualification gate, and it is handled in Step 5 above. If you fail that check, nothing else here matters yet.

"Are there even affiliates in my category, or in my country?" (8 of 55)

Almost every B2B category has them. They rarely call themselves affiliates. They are consultants, agencies, newsletter writers, YouTubers, freelancers and comparison sites who already recommend tools for a living. Test it before you build anything: search "best [your category] tools" and "[your closest competitor] alternatives" and look at who ranks. Then run the competitor backlink method described above. If those searches return real independent publishers, your affiliates exist. On country: language matters far more than borders. A Dutch program sells fine to a German audience if someone is writing in German.

"Who actually runs this day to day, and who answers affiliate questions?" (8 of 55)

Once it is set up, a few hours a week rather than a hire. The work is approving applications, a monthly payout run, and talking to the small number of affiliates who are actually producing. Affiliate questions split three ways, and it is worth knowing which is which before you buy anything: platform questions (my link is not working, where is my payout, how do I read this dashboard) should be answered by your affiliate platform's own support, not by you. Product questions go to your normal support. Commercial questions (can I run this campaign, can I get a better rate) are yours, and those are the ones you want. Two things to check on any tool before you sign: whether it charges an extra percentage on payouts on top of the subscription, and whether it supports your affiliates directly or routes every question back to you.

"Can we pay per lead or per demo instead of a share of revenue?" (7 of 55)

Usually asked by sales-led companies with a long cycle, and usually a symptom rather than a preference. The affiliate is worried they will do the work and not get paid, which is normally a cookie window problem. Fix the window first and this often goes away. If you still want a lead component, understand what it changes: paying per lead moves the risk from the affiliate to you, and it changes who applies. You will attract volume, and you will pay for form fills you have to qualify anyway. If you do it, cap the monthly spend, pay on a demo that was held and matched your ICP rather than on a booking, and keep a recurring revenue share on top so the incentive still points at customers rather than at calendars. (FAQ)

"Affiliates will bid on our brand name and we will pay commission on traffic we already had." (6 of 55)

A real risk, and the answer is a written clause rather than a norm. Put it in the program terms before you open: no bidding on your brand name or brand plus modifier keywords, no use of your name in a display URL, no landing pages that imitate your site. Then check it monthly, because unenforced terms are just a suggestion.

Worth knowing that the opposite choice is also legitimate. Potion went from 0 to 30K MRR in three months, and it worked because they allowed everything, including affiliates running paid ads on the brand. That is a deliberate strategy with a real upside. Make the choice on purpose, in writing, rather than by leaving it unsaid.

"What breaks when we migrate an existing program?" ([X] of 55, confirm from call data)

Three things, and only three. Historical attribution, meaning which affiliate owns which existing customer and what they are still owed. The affiliate's existing links, which are already published in articles you do not control. And payout continuity, because an affiliate who misses one payment cycle during a migration assumes the worst. Part 3 below covers how to handle each without a gap.

"Won't affiliates just take credit for customers who would have bought anyway?" ([X] of 55, confirm from call data)

Partly a tracking question and partly a terms question. The brand bidding clause above removes the most common version of it. Beyond that, settle your attribution rules in advance (Step 5), and when a specific account looks wrong, check whether the affiliate touch came before or after your own first touch. Do that a handful of times in the first quarter and you will know quickly whether the concern is real in your program or theoretical.

The objection that barely came up: cost.

In 55 sales calls, the size of the commission itself was not among the most-raised concerns, which is itself a finding worth carrying into your internal pitch. If you do get the finance question, the sequencing answers it: you collect the customer's payment before you pay the commission out of it, so the channel funds itself from revenue it created rather than from a budget line you had to win.

Two of the sixteen objections in the dataset were about our own product rather than about affiliate programs generally: which plan the marketplace sits on, and which billing systems we connect to directly. Short answers, and then back to the topic. The marketplace is on a higher plan. Stripe is the direct payment integration, and anything else connects through the API. Everything else on that front lives on the pricing page and in the help centre.

Who should own the program internally comes up often enough that it has its own answer in the FAQ below.

Part 3: Migrating an existing affiliate program

If you already have a program somewhere else and it has stalled, the diagnosis usually comes down to one of three things.

• You have tracking software but no affiliates. Most tools track well and recruit not at all, so growth stays capped by how many partners you can find yourself.
• Tracking is unreliable. Affiliates who suspect they are not being credited quietly stop promoting, and you see it as a slow decline rather than a complaint.
• The admin has become the job. Manual payouts, spreadsheet reconciliation, and support requests eat the time you should spend recruiting.

Migration is less disruptive than most teams expect. The mechanics are covered in migrate your affiliate program in 5 steps, and there are email templates for notifying your affiliates, which is the part people underestimate. Tell partners before you move, not after, and confirm their commission terms carry over unchanged.

Two things to protect: historical commission obligations to existing affiliates, and link continuity so the content partners already published keeps working. Get both in writing before you start.

Teams that have done this: Expandi left FirstPromoter, involve.me grew 6x after switching from Rewardful, Insightful tripled affiliate revenue after the same switch, and Warmup Inbox tripled after moving off in-house tooling. If your tracking already works and the gap is purely recruitment, adding a network alongside an in-house program is a valid path too.

Will your affiliates survive the move?

This is the question nobody asks out loud, and the honest answer is that activity usually goes up rather than down.

Affiliates come across with their existing links, so the articles and videos already sending you traffic keep sending it and keep tracking. Beyond that, three things usually improve from the affiliate's point of view: payouts in the old program were often late, payment options widen, and instead of logging into a portal showing one program they see every program they work with in one place. Affiliates who are already in the network skip signup entirely, because the partnership is simply created for them.

We have not had a customer lose their affiliates to a migration. That is the mechanism working rather than a guarantee, and anyone promising you certainty on this is selling.

How you avoid paying the same commission twice

The overlap between two systems is where double payments happen. The cutover is the Stripe connection, and it is made on the first of a month on purpose, so the boundary is unambiguous. From that date forward, commissions generate in the new system. Commissions from earlier months are carried over by hand, and anything already paid in the old tool is marked as paid so both sides reconcile. Edge cases get a manual check rather than an assumption.

When not to migrate yet

Sometimes the answer is to wait, and we do tell people that. If a company is very early, nobody knows the brand yet, and its own users and network are not already recommending it, then changing tools will not fix anything. Why would a network of affiliates start recommending a product that its own customers do not recommend?

Migration itself is straightforward. Timing is the real question, and it is usually about when the program was started rather than which tool it runs on. If that describes you, the Growth plan is the right place to sit: set the program up without a marketplace listing, let people join, find out whether it works, keep the cost low, and move into the network once there is a foundation underneath it.

Where to start this week

If you are launching: do steps 1 through 5 before you contact anyone. Getting commission and tracking right first is what separates programs that compound from programs that get quietly archived.

If you already have a program: do not recruit anyone new until you have segmented your existing affiliates by activation and talked to the ones who convert. The growth is nearly always sitting in the list you already have.

If you are migrating: start with why the current program stalled. If it is recruitment rather than tracking, changing tools alone will not fix it, and you should plan the recruitment motion at the same time.

About the author. Joran Hofman was an affiliate before he built anything for affiliate programs, promoting around 80 SaaS tools through his own site. He founded Reditus after that experience, and the numbers in this article come from three real programs and from 55 sales calls with SaaS companies deciding whether to launch one.

*Last updated: [publish date]. The distribution data covers 10,596 affiliate-months across three programs. The objection frequencies come from 55 recorded sales calls.*

If you would rather work from a checklist than a narrative, our step by step guide to setting up an affiliate program covers the same build in a shorter, more procedural form.