Your recurring affiliate income stops growing at a number you can work out this afternoon. Take the new paying agencies you refer each month and divide by your monthly churn rate. Four new agencies a month at 5% churn gives you 80 paying agencies, and that's where you level off. More months of work won't move it. Only the inputs move it. This article shows you how to forecast recurring affiliate income with that formula, so the ceiling is a figure you've calculated, not something you discover the hard way two years from now.
The formula gives you the end point. A cohort sheet gives you the road to it, month by month, and that's where it gets useful. You'll see roughly when your commission flattens. You'll also see why cutting churn by a point or two usually lifts the ceiling for less effort than chasing extra signups. The two levers move the ceiling by exactly the same arithmetic. Churn wins on effort, and because a lower rate slows the decay of every agency already paying you, not just the new ones.
We'll build it together using SendTo's own affiliate terms and carry one example through every step: 4 new paying agencies a month, 5% monthly churn. Each step gives you the action, what you should see on screen, and the mistake that throws forecasts off at that point.

By the end you will have a 36-month forecast and a ceiling number you can steer
What you'll end with is a single-tab spreadsheet. It shows paying agencies and monthly commission for each of the next 36 months. It shows your ceiling, the number of paying agencies your current pace settles at. And it shows the month your income reaches about 90% of that ceiling, which is the honest answer to "when will my commissions stop growing?"
You need three things before you open a blank sheet. First, your last 6 to 12 months of referral data, meaning how many referred agencies started paying each month and how many cancelled. Your affiliate dashboard has both. Second, the commission rate. On SendTo that's 20% of the partner subscription each referred agency pays. Third, the partner fee itself, which you'll take from the pricing page and type into one cell. We won't quote a figure here, because prices change and your sheet should change with them.
Step 1: Count only what you are actually paid on
Start with a plain list. One row per referred agency that has paid at least one month, with two columns: the month it started paying and the month it cancelled, if it has. Leave the end month blank for agencies still paying. That's your raw material, and nothing else goes in.
Here's why the list is that narrow. SendTo pays you 20% of the platform fee each referred agency pays us, every month that agency keeps paying it. You never get a share of what that agency's own clients pay the agency. So an agency that resells SendTo to 40 retainer clients earns you the same commission as one with 3 clients. Your forecast counts agencies paying the partner subscription, full stop.
What you should see is a short column of start months and a shorter column of end months. Ten to thirty rows is normal for an affiliate in their first year. If it looks thin, that's fine. It's real.
The 90-day window decides which month a referral lands in
Attribution on SendTo lasts 90 days from the first click, and the first touch wins. So an agency owner who clicks your link on 10 January and signs up on 2 April is still yours. In your sheet, though, that agency belongs to the month it started paying, not the month it clicked. This is why a busy content month often shows up in your numbers one to three months later. Don't push referrals back to the click month to make a chart look tidier.
Step 2: Work out your real monthly churn rate, not the one you hope for
Churn for a month is the number of agencies that cancelled that month, divided by the number paying at the start of it. If 40 agencies were paying on 1 March and 2 cancelled during March, March churn is 2 divided by 40, or 5%. Work this out for every month you have, then average at least six of them.
What you should see is a figure that jumps about. One month at 0%, the next at 8%, then 3%. With small numbers, a single cancellation swings the rate a lot. Take the average and move on. If you want a fuller picture of why referred agencies leave, our piece on understanding affiliate churn covers the retention side.
Split churn into early-life and settled if the first 90 days look heavy
Agencies tend to cancel more in their first three months than later. They're still setting up, still deciding whether to resell, still finding out if their clients take to it. After that, the ones that stay are usually running it for paying clients and leave far less often. One blended rate hides this, and it makes your long-run forecast too gloomy and your short-run forecast too rosy.
To split it, tag each agency-month as either early (the agency's first three paying months) or settled (month four onwards). Then divide cancellations by agency-months in each group. Say over nine months you logged 30 early agency-months with 3 cancellations, and 120 settled agency-months with 4. Early-life churn is 3 divided by 30, or 10%. Settled churn is 4 divided by 120, about 3.3%. Blended, it's 7 divided by 150, roughly 4.7%, which rounds to our 5% example.
The common mistake here is trusting three months of data from eight agencies. That's a sample where one owner's bad week sets your whole forecast. Use six months at least, and treat anything under twenty paying agencies as a range, not a number.
Step 3: Build the cohort grid so each month's referrals decay on their own
Each month's new agencies are a cohort, and each cohort shrinks on its own clock. Your grid has start months down the side, one row per cohort, and calendar months across the top, 36 columns. Put the cohort size, the number of agencies that started paying that month, in a column next to each row label. For future months, use your average new paying agencies, which is 4 in our example.
Each cell holds the agencies from that cohort still paying in that calendar month. The pattern is cohort size times one minus churn, raised to the number of months since the cohort started. If the calendar month comes before the start month, the cell stays blank. Keep churn in its own input cell and point every formula at it with an absolute reference.
Single rate:
=IF(col_month < start_month, "", cohort_size * (1 - churn) ^ (col_month - start_month))
Two-stage (early rate for the first 3 months, settled rate after):
=IF(col_month < start_month, "",
IF(col_month - start_month <= 3,
cohort_size * (1 - early) ^ (col_month - start_month),
cohort_size * (1 - early) ^ 3 * (1 - settled) ^ (col_month - start_month - 3)))
Number your months 1 to 36 in a helper row above the dates and a helper column beside the start months, so the subtraction is simple. Then add a total row under the grid that sums each column. That total is your paying agencies for that calendar month.
What you should see is a triangle of shrinking numbers. In our example, the first cohort reads 4, then 3.8, then 3.61, and by month 12 it's down to about 2.16. The total row climbs fast at first. After 12 months it's at about 36.8 paying agencies, and by month 36 about 67.4. Fractions are fine. They're expected values, not people.
Step 4: Turn paying agencies into commission with one input cell
One cell turns agencies into money. Put the partner fee from the pricing page in a cell labelled "partner fee", and the commission rate, 20%, in the cell beside it. Under the total row, add a commission row where each cell is paying agencies times partner fee times commission rate, with both inputs as absolute references.
Keeping the fee in one cell is the whole point. When the fee changes, you update one number and all 36 months recalculate. Hard-code the fee into each formula and you'll miss a cell, and the forecast quietly goes wrong in a way you won't spot for months.
To keep the worked example free of any price, type 1 into the fee cell. The sheet then shows commission in fee units, meaning multiples of one agency's partner fee. Each paying agency earns you 0.2 fee units a month. At month 12, our 36.8 agencies earn about 7.4 fee units. At month 36, 67.4 agencies earn about 13.5. Swap in the real fee and those become your figures.
What you should see is a commission row that climbs steeply in year one, keeps rising more slowly in year two, and starts to flatten in year three. That flattening is the ceiling coming into view, and Step 5 puts an exact number on it.
One more thing belongs on this row, which is timing. SendTo pays commission monthly, by PayPal, after the month it was earned. If you're planning cash rather than earnings, add a second row that shifts each month's commission one column to the right. It's the same total, just arriving a month later, and it stops you budgeting against money that hasn't landed yet.
Step 5: Find the ceiling and the month your income levels off
The whole forecast comes down to one division. Your ceiling is new paying agencies per month divided by monthly churn. At 4 new agencies a month and 5% churn, that's 4 divided by 0.05, or 80 paying agencies. Put it in a cell above the grid and label it "ceiling". When you forecast recurring affiliate income, this is the number your total row is always heading towards.
The time it takes to get there depends only on churn. You reach about 63% of the ceiling after roughly 1 divided by churn months, and about 90% after roughly 2.3 divided by churn months. At 5% churn that's about 20 months and about 46 months. Add a cell for each one so they update when your inputs change.
What you should see is the grid agreeing with the formula. At month 20, the total row reads about 51.3 agencies, which is 64% of 80. Extend the grid to 48 columns and month 46 reads about 72.4, which is 90%. In fee units, the ceiling is 80 times 0.2, or 16 fee units a month. If the grid and the formula disagree by more than a rounding error, an absolute reference has slipped somewhere.
Step 6: Run the two levers side by side and see why churn usually wins
Copy your tab twice, so you have three. Leave the first as the baseline at 4 new agencies and 5% churn. In the second, change new agencies to 5 and keep churn at 5%. In the third, keep 4 new agencies and drop churn to 4%. Change nothing else.
| Scenario | Ceiling (agencies) | Paying at month 36 | Commission at month 36 (fee units) |
|---|---|---|---|
| Baseline: 4 a month, 5% churn | 80 | 67.4 | 13.5 |
| More signups: 5 a month, 5% churn | 100 | 84.2 | 16.8 |
| Lower churn: 4 a month, 4% churn | 100 | 77.0 | 15.4 |
Both changes lift the ceiling to exactly 100. By the formula they're equal, so we won't pretend otherwise. Within 36 months the extra signups actually pull ahead, because lower churn also slows how fast you climb. At 4% churn, 90% of the ceiling takes about 58 months instead of 46.
Churn still wins on two counts. The first is effort. Five agencies a month instead of four means 25% more paying referrals every month, permanently, so 25% more content, traffic or conversations. Cutting churn by a point mostly means helping the agencies you've already referred get through their first 90 days. A setup checklist, a walkthrough of the client portal, a short email on day 30. That's a few hours of work, once.
The second is that a lower rate works on agencies you've already referred. A signup boost only adds new cohorts. A churn cut slows the decay of every row in your grid at the same moment. If you already have 60 agencies paying, one point less churn keeps 0.6 more of them every month, starting this month, with no new referrals needed. If Step 2 showed heavy early churn, test the cut on the early-life rate first. Dropping it from 10% to 7% is often the cheapest point you'll find.
Step 7: Add seasonality so the forecast doesn't surprise you in summer
New referrals don't arrive evenly across the year. Interest in email marketing peaks in January, November and December, and it's lowest in June. Add a row of twelve monthly multipliers above your cohort-size column, then set each future cohort to your average new agencies times that month's multiplier.
A reasonable starting shape is January 1.3, February 1.1, March 1.0, April 0.95, May 0.85, June 0.7, July 0.8, August 0.85, September 1.0, October 1.05, November 1.25 and December 1.15. These add up to 12, so your yearly total stays the same. Swap them for your own figures once you have a full year of data.
It's April 2026, so be clear about what's ahead. April to June is the soft quarter. The 90-day attribution window also pushes some of the dip into July and August, because clicks from a quiet June turn into paying agencies later.
What you should see is a dip in the growth curve around mid-year and a stronger climb from November into January. The ceiling stays exactly where it was, because seasonality moves the timing of referrals, not the average. For ways to hold commission steady through the dip, see our guide to retaining earnings during seasonal market shifts.
Work backwards from a target like $100 a day to see the referrals it needs
People who ask how to make $100 a day from affiliate marketing are asking an arithmetic question, and the sheet answers it. Nothing here promises you'll reach a target. It shows what a target takes. You need three steps.
- Multiply your daily target by 30 to get a monthly target.
- Divide the monthly target by commission per agency, which is your fee cell times 20%. That's the number of paying agencies you need.
- Multiply that number by your churn rate. That's the number of new paying agencies a month you need to hold it.
Work it in fee units and the price never matters. Say your monthly target equals 20 fee units. You'd need 100 paying agencies, because each one earns 0.2 fee units. At 5% churn, that's 5 new paying agencies a month, or 4 a month at 4% churn. Those are the two scenarios you already built in Step 6. Put your real fee in the cell and your real target in a new cell, and the sheet tells you which one you need.
The 80/20 rule shows up in your cohort grid
The 80/20 rule says a minority of referrals produce most of your income. With recurring commission, those are the well-fitted agencies that stay for years, typically the ones already running retainers for paying clients. Sort your Step 1 list by months paid and you'll usually find a small group at the top. Find out where those referrals came from, then make more of that content.
Success rates and staying faceless
There's no reliable single success rate for affiliate marketing, and we won't make one up. The figures you can measure are your own. What share of clicks become paying agencies, and what share of those are still paying after month three? Those two decide your ceiling.
You can also do it faceless. The sheet doesn't care whether agency owners know your face. Written reviews, tutorials and an email list all count the same. Being faceless doesn't excuse you from disclosure, though, so read our guide to writing affiliate disclosures that keep reader trust.
Check the forecast against reality every quarter and fix it when it drifts
A forecast only stays useful if you check it. Each quarter, open your affiliate dashboard and compare actual paying agencies with the total row for that month. If they're within about 10%, leave the inputs alone. With small numbers, a gap that size is just noise.
If actuals run low, work through the causes in this order, because the first ones are the most common and the quickest to check.
- Attribution timing. Clicks from the last 90 days may not have turned into paying agencies yet, and first touch decides who gets credit.
- Early-life churn. Compare first-90-day cancellations with your modelled rate, and switch to two-stage churn if they're running heavier.
- Seasonal dip. Check whether the gap lines up with the soft months in your multiplier row.
- Fee change. Make sure your fee cell matches the current pricing page.
If actuals run high, don't raise the ceiling straight away. Wait for three months in a row to confirm the higher pace, then update new agencies or churn, whichever one actually changed. One strong month is usually a single post or launch, not a new run rate.
For a full year-end review on top of the quarterly checks, our guide on how to review your 2025 affiliate results and set 2026 targets uses the same inputs.
Your next step is to fill in the inputs and steer the ceiling
You now have a ceiling, a month for reaching 90% of it, and two levers that move it by the same arithmetic. Start with churn. Measure it, split it into early-life and settled, and work on the first 90 days, because a lower rate keeps more of every agency already paying you.
Affiliates earn 20% of the partner subscription every referred agency pays SendTo, every month it pays, with 90-day first-touch attribution and monthly PayPal payouts.
Join the SendTo affiliate programme