Blog
Playbook

A 90-day plan to launch a SaaS partner program

A week-by-week 90-day partner program plan for SaaS: setup, recruiting the first 20 partners, one content campaign, measuring, and what to drop if behind.

Oct 26, 2026 · 12 min read · RelayWonder
Prospectsown data · reviewers · communitiesRepliedApprovedActive5 of 20 drive most sales

A 90-day partner program plan for a SaaS brand fits into 13 weeks and four phases: weeks 1 to 2 set up terms, tracking and payouts; weeks 3 to 6 recruit the first 20 partners from your own customers, reviewers, communities and targeted outreach; weeks 7 to 10 activate them and run one content campaign; weeks 11 to 13 measure and adjust. Illustrative targets are 20 approved partners by week 6, 10 of them active by week 10, five pieces of partner content live, and the first paying customers attributed before day 90. The plan works because each phase has one job. If you fall behind, drop scope inside the phase, not the phase itself: a working program with 12 partners beats a perfect program that never launches.

What a 90-day partner program plan looks like

Illustrative 13-week plan. Targets are examples for a small SaaS brand, not benchmarks; set your own from your traffic and customer base.
PhaseWeeksGoalIllustrative target at phase endMain deliverable
Set up1 to 2A program a partner can join and be paid byOne end-to-end test purchase attributed and reversed correctlyTerms, tracking, payout process, program page
Recruit3 to 6The first 20 approved partners24 applications, 20 approvedOutreach to customers, reviewers, communities and targeted creators
Activate7 to 10Partners who actually publish10 of 20 active, 5 campaign pieces liveOnboarding, asset kit, one content campaign
Measure11 to 13Know what to keep and what to changeFirst paying customers attributed; a decision on terms and sourcesA one-page review and the next quarter's plan

Ninety days is long enough to see partners publish and customers arrive, and short enough to keep the program from becoming a side project nobody owns. Assign one owner for the whole plan. Partner programs that are everyone's job in week 1 are nobody's job by week 6.

clickday 30day 60day 90referral window: a signup here is creditedoutside the window
A 13-week timeline divided into set up, recruit, activate and measure, with a target marked at the end of each phase.

Weeks 1 to 2: terms, tracking and payouts

The setup phase has one test: can a stranger join, share a link, have a customer pay, and see a correct commission? Everything here serves that test.

Write the terms

  • Commission: one rate and one duration, for example 20% of each paid invoice for the first 12 months. Tiers and bonuses can wait for quarter two.
  • Referral window: 60 days is a common default; lengthen it only if your own data shows a longer path from first visit to payment.
  • Hold: 30 days before commission becomes payable, matched to your refund policy.
  • Minimum payout: $50, paid in monthly batches.
  • Attribution: last touch for links, and a written rule for when a coupon code and a link point to different partners.
  • Disclosure: partners must disclose the relationship clearly and close to the link. The FTC gives "paid link" next to an affiliate link as an adequate example [2].
  • Prohibited: self-referral, posting codes to coupon sites, bidding on your brand name in search ads.

Install and test tracking

Install the tracking script on your site, connect billing, and make sure checkout passes the referral id to Stripe. In Stripe Checkout this is client_reference_id, a string Stripe describes as a way to reconcile the session with your internal systems [1]. Then run the full test: click a test partner link, buy, confirm the commission row appears, refund, and confirm a negative row cancels it. Repeat with a partner coupon code and no click. Do not recruit anyone until both paths work.

Decide how you will pay

Choose the payout rails before the first partner asks. PayPal Payouts sends batch payments to many recipients from a business account [5]; Wise Business batch payments accept up to 1,000 payments in a single CSV or XLSX file [6]. Collect tax forms once at onboarding: US persons give a Form W-9 to provide a correct taxpayer identification number [3], and foreign individuals give a Form W-8BEN to the payer [4]. Ask your accountant which you need for your situation; the point in week 2 is that the process exists.

Finish the phase with a program page that lists the terms as plain lines, one number each, and an application form. Partners compare programs in seconds; so do the AI assistants they increasingly ask.

Weeks 3 to 6: how to recruit the first 20 partners

The first 20 partners almost never come from a launch announcement. They come from people who already know your product, people who already write about your category, and a short list of creators you contact one by one. Work the sources in order of warmth.

Illustrative recruiting funnel for weeks 3 to 6. Application and approval rates are assumptions to plan against, not measured benchmarks.
SourceContactedIllustrative application rateApplicationsApproved
Your own customers (email to active accounts)5002%109
Reviewers and bloggers who already covered your category4015%65
Targeted creators found by search (YouTube, newsletters, blogs)1006%65
Communities (one useful post in each of three)3 postsn/a21
Total2420
  1. Week 3: email your active customers. Short, personal, from the founder or the program owner: what the program pays, who it is for, one link to apply. Customers who already recommend you informally are your best partners.
  2. Week 4: contact reviewers and bloggers who have written about your category in the last year. Offer them early access to anything new and a personal code for their readers.
  3. Weeks 4 to 5: targeted outreach to creators whose audience matches your buyers. Check audience quality before you write: recent uploads, views relative to subscribers, real engagement. Send a personal first message and at most two follow-ups.
  4. Week 6: post in two or three communities where your buyers spend time, only where the rules allow it, and only with something useful rather than a pitch.

Approve quickly. An application that waits a week signals how the rest of the relationship will go. Aim to answer every application within two business days, and decline with a reason when the fit is wrong.

partnerBrand ABrand BBrand CBrand D
Four recruiting sources, customers, reviewers, targeted creators and communities, feeding 24 applications and 20 approved partners.

Weeks 7 to 10: activate partners and run one content campaign

Approved is not active. In most new programs a large share of approved partners never share a link unless someone gives them a reason and the material to do it. This phase turns approvals into published content.

Onboard every partner the same way

  • A welcome email on approval with their link, their personal code, the terms in five lines, and one suggested first post.
  • An asset kit: screenshots, a short product description in three lengths, a comparison table, logo files.
  • A 15-minute call with your five most promising partners. Ask what their audience asks about; that is the brief for their first piece.

Run one content campaign

One campaign, not three. Write a brief with a clear angle, list the deliverables you will accept (a YouTube tutorial, a blog post, a newsletter feature) and set a per-piece bounty. An illustrative budget: $150 per piece for five pieces, $750 in total, on top of normal commission. Partners apply, some with a quote; accept, decline or counter-offer, then check each submitted piece for a working link, the code, a disclosure and an accurate description before you confirm delivery.

The campaign does two jobs. It produces five pieces of content that keep sending traffic after day 90, and it shows you which partners deliver on time and to the brief. Those partners are your core for quarter two.

Illustrative target for week 10: 10 of 20 partners active, meaning at least one click or code redemption in the last 30 days, and five campaign pieces live.

Weeks 11 to 13: measure and adjust

Measure a short list of numbers and resist judging the program on cash alone. In the first 90 days, partner costs arrive faster than customer value, so the payout ratio will look worse than it is.

Illustrative measurement at week 13. Expected values use 5% monthly churn: (1 − 0.95¹²) ÷ 0.05 ≈ 9.19 paid invoices, $49 plan, 20% for 12 months; all inputs are assumptions.
MetricFormulaIllustrative week-13 value
Active rateActive partners ÷ approved partners10 ÷ 20 = 50%
ClicksTotal tracked clicks1,200
TrialsTrials attributed to partners60
Paying customersCustomers with a paid invoice attributed to a partner12
Referred revenue to date12 customers × 2 paid invoices × $49$1,176
Commission to date$1,176 × 20%$235.20
EPC (earnings per click)Commission ÷ clicks$235.20 ÷ 1,200 ≈ $0.20
Partner cost to dateCommission + campaign bounties$235.20 + $750 = $985.20
Expected first-year value of these 12 customers12 × $450 (9.19 invoices × $49 at 5% churn)about $5,405, against about $1,831 in commission and bounties

Read the last two rows together. To date, partners have cost $985 against $1,176 of revenue, which looks like an 84% payout ratio. Over the customers' first year, the same 12 customers are expected to pay about $5,405 and cost about $1,081 in commission plus the $750 already spent, about $1,831 in total. The first number is a cash-timing artifact; the second is the one to decide on.

Decisions to make in week 13

  • Sources: which recruiting source produced active partners, not just approvals? Put quarter two's effort there.
  • Terms: did any partner decline because of the rate, the window or the minimum payout? One complaint is noise; three is a signal.
  • Formats: which campaign deliverable produced clicks? Brief more of that.
  • Partners: which five partners produced most of the clicks? Call each of them.
EPCcommission ÷ clicksConversioncustomers ÷ clicksPayout ratiocommission ÷ revenueActive rateactive ÷ approved
Week-13 metric cards: active rate, clicks, paying customers, earnings per click, and partner cost against expected first-year value.

What to drop if you are behind

Every 90-day plan slips somewhere. Decide in advance what goes first, so a slip does not turn into a stall.

Recommended fallbacks for a slipping 90-day plan; guidance, not data.
Where you are behindDropKeep
Setup runs into week 3Custom program page design, tiers, bonusesTerms, the end-to-end test, the payout process
Fewer than 12 approved by week 6Community posts and cold creatorsCustomer email and reviewers; extend recruiting by two weeks
Partners approved but nobody publishing by week 8The full campaign; run a two-piece versionOnboarding calls with the top five
Campaign pieces lateWaiting for all fiveMeasure on the pieces that are live
No paying customers by week 13Changing the commission rateChecking tracking, then landing pages, then partner fit, in that order

The last row deserves a warning. When no customers arrive, the instinct is to raise the rate. A rate change rarely fixes a program with no conversions; broken tracking, a weak landing page or partners whose audience does not buy are far more common causes. Check them in that order before you touch the terms.

When will partners in a new program get paid?

Be honest about this in week 1, because partners will ask in week 8. With a 30-day hold and a $50 minimum payout, a partner who refers one customer on a $49 plan at 20% earns $9.80 a month and needs six paid invoices, $58.80, to cross the minimum. Their first payout comes after that sixth invoice clears the hold. A partner who refers six customers in week 8 earns 6 × $9.80 = $58.80 on the first invoices, crosses $50 at once, and is paid in the first monthly batch after the hold.

Say this on the program page in plain words. Partners who know the schedule wait; partners who discover it feel misled. A bounty from a content campaign helps here: it gives active partners a payment they can see within the 90 days.

How RelayWonder supports a 90-day launch

RelayWonder's program defaults match the setup phase: a 60-day referral window, a 30-day hold, a $50 minimum payout and last-touch attribution, with commission calculated per Stripe invoice and refunds recorded as negative ledger rows. For recruiting, Hunter searches the web, YouTube and Reddit for creators in your category, checks YouTube audience quality, scores fits A, B or C with a reason, and sends outreach from your own mailbox with follow-ups that stop when someone replies. For activation, content campaigns handle the brief, applications, counter-offers, content checks and bounties. Payouts are exported as PayPal or Wise CSV files and paid by you from your own account.

What to do in week 1

  1. Name one owner for the 90 days.
  2. Write the terms: rate, duration, window, hold, minimum payout, attribution rule, disclosure, prohibited practices.
  3. Install tracking and connect billing.
  4. Run the end-to-end test with a link and with a code, including a refund.
  5. Choose PayPal or Wise for payouts and set up tax form collection.
  6. Draft the customer email you will send in week 3.

FAQ

How long does it take to launch a SaaS partner program?

A workable program can launch in about two weeks of setup and reach its first paying customers within 90 days. A 13-week plan splits into setup (weeks 1 to 2), recruiting (3 to 6), activation (7 to 10) and measurement (11 to 13). Illustrative targets are 20 approved partners by week 6 and 10 active by week 10.

Where do the first 20 partners usually come from?

From people who already know your product or category: your own customers, reviewers and bloggers who have covered the category, and creators you contact one by one. In an illustrative plan, emailing 500 customers at a 2% application rate yields 10 applications. Communities add a few more if you post something useful.

What should be in a new partner program's terms?

One rate and one duration (for example 20% for 12 months), a referral window such as 60 days, a hold such as 30 days, a minimum payout such as $50, an attribution rule, a disclosure requirement and a list of prohibited practices. Leave tiers and bonuses for later.

What should I measure at the end of the first 90 days?

Active rate, clicks, trials, paying customers, earnings per click and partner cost. Judge cost against the expected first-year value of the referred customers, not cash to date: in an illustrative case, 12 customers cost about $985 so far but are expected to pay about $5,405 in their first year.

What should I drop if the 90-day plan falls behind?

Drop scope inside the phase, not the phase. Skip tiers and custom design during setup, cut cold outreach before customer email during recruiting, and shrink the content campaign to two pieces if partners are slow. Keep the end-to-end tracking test at all costs.

When will partners in a new program receive their first payout?

It depends on the hold and minimum. With a 30-day hold and $50 minimum, a partner earning $9.80 a month from one customer needs six paid invoices to cross $50, then waits for the hold. A partner with six customers ($58.80 on the first invoices) can be paid in the first monthly batch after the hold.

Sources

  1. Stripe API reference: Create a Checkout Session · client_reference_id for reconciling a session with internal systems. Checked October 2026.
  2. FTC: FTC's Endorsement Guides: What People Are Asking · Clear and conspicuous disclosure of affiliate relationships; "paid link" as an adequate example. Checked October 2026.
  3. IRS: About Form W-9 · Form W-9 used to provide a correct taxpayer identification number. Checked October 2026.
  4. IRS: About Form W-8 BEN · Given by foreign individuals to the withholding agent or payer. Checked October 2026.
  5. PayPal Developer: Payouts · Batch payments to many recipients. Checked October 2026.
  6. Wise Help Centre: Guide to batch payments · Up to 1,000 payments in a single CSV or XLSX file. Checked October 2026.
Topicspartner program 90-day planlaunch a SaaS partner programaffiliate program launch checklistrecruit first 20 partnerspartner activationpartner program metricsaffiliate program timeline