Recommended resources
The freely available resources this lesson will direct you to.
Hello! In the previous lesson, you saw why a click does not guarantee a commission: a purchase must be tracked, eligible, and credited under the program’s rules. Now you’ll put numbers on that journey. You’ll estimate earnings from content views, affiliate-link clicks, purchases, commission, and refunds—so you can assess a promotion without mistaking audience size for income.
Give each rate the right starting point
Imagine you publish a product recommendation. Some people see it, some click your affiliate link, and a smaller number make a qualifying purchase. Each rate describes one particular part of that journey:
- Reach for this calculation: the number of times the content was shown. Platforms often call these impressions. One person may account for more than one impression.
- Click-through rate (CTR): affiliate-link clicks divided by content impressions.
- Conversion rate: qualifying purchases divided by affiliate-link clicks.
The denominators matter. A conversion rate means three purchases per 100 clicks, not per 100 people who saw the content.
Read Shopify’s 20 Affiliate Marketing Metrics Worth Tracking in 2026 for the definitions and formulas. The purpose here is to distinguish what CTR measures from what conversion rate measures; you do not need its suggested performance benchmarks to make an estimate.
20 Affiliate Marketing Metrics Worth Tracking in 2026
Shopify defines the two rates that connect content views to purchases.
In “2. Affiliate click-through rate (CTR),” read the CTR explanation and the formula immediately beneath it. Then, in “4. Affiliate conversion rate,” read the conversion explanation and its following formula. For this lesson, treat a conversion as an eligible purchase; a program might define another action instead.
Suppose your content receives impressions and your affiliate-link CTR is . Convert the percentage to a decimal before multiplying: . You would estimate:
If of those clicks lead to qualifying purchases, the next estimate is:
You cannot receive of an individual order. The is an average expectation for planning, not a prediction that this particular post will produce exactly that many sales. Do not round it to seven yet: keeping the unrounded figure gives a more accurate earnings estimate.
A link hub adds a practical wrinkle. If viewers must first visit a page and then click an affiliate link, use the number who actually click through to the merchant as the input to an offer conversion rate measured from merchant-bound clicks. Do not mix a rate measured from affiliate clicks with a count of link-hub visitors.
Turn purchases into commission
A commission may be a percentage of the eligible purchase price or a fixed amount per approved sale. If an item costs USD and the program pays of that price, the commission per qualifying, non-refunded sale is:
Using the estimated purchases gives an estimated gross commission of:
Gross commission is not necessarily what you keep. Under a program that reverses commission on fully refunded orders, an estimated refund rate means retaining, on average, of gross commission:
The compact version works when the same price, commission, conversion rate, and refund assumption apply across the purchases:
Here, CTR, conversion rate, and refund rate must be entered as decimals. If a program pays a fixed USD per approved sale, you can use that amount directly without calculating a percentage of the product price.
The Free Affiliate Commission Calculator from systeme.io presents the per-click and refund-adjustment ideas together. Read only the focused passages below: the page also promotes its own program, and its category-wide performance figures are not dependable forecasts for a new audience.
Free Affiliate Commission Calculator (EPC + Monthly Earnings) | systeme.io
systeme.io shows how conversion and commission combine into earnings per click, and how refunds reduce gross commission.
Under “The formulas,” begin with the per-click explanation; include the formula under “EPC (earnings per click).” Then go to “Net after refunds,” read its formula and the refund passage. Focus on the calculation, not the page’s suggested category refund rates: actual reversal rules and rates depend on the program and its customers.
This calculation assumes the purchases are tracked and qualify for credit under the program rules you studied previously. It also assumes the stated commission applies to the entire USD price. In a real program, check what portion of an order is commissionable and how refunded orders are handled.
Compare offers by what a click is worth
A higher commission percentage does not necessarily produce higher earnings. The image below compares two hypothetical offers for every 100 visitors who reach an affiliate offer.

For the left-hand offer, a USD product with commission pays USD per sale. At a conversion rate, its estimated earnings per click (EPC) before refunds are:
For the right-hand offer, a USD product with commission pays only USD per sale. But at a conversion rate, its estimated EPC is:
Thus, under the image’s assumptions, the second offer earns more per click despite its lower commission rate. This comparison does not establish that the second product will convert better with your audience; its conversion rates are assumptions to investigate.
Shopify also defines EPC using results you have actually recorded. Read its “7. Affiliate earnings per click (EPC)” subsection, keeping that distinction in mind.
20 Affiliate Marketing Metrics Worth Tracking in 2026
Return to Shopify for the reported-results version of EPC: earnings divided by clicks.
In “7. Affiliate earnings per click (EPC),” read the EPC definition and the formula directly below it. Compare that reported-results formula with the estimate we calculated from assumed conversion and commission.
There are therefore two useful ways to arrive at EPC. Estimated EPC uses assumptions: conversion rate multiplied by commission per sale. Observed EPC uses recorded earnings divided by recorded clicks. If you account for reversed commission, specify that you are using after-refund earnings.
In our -impression example, estimated gross EPC is USD. After the assumed refund rate, estimated EPC becomes USD. Multiplying that by the expected clicks reproduces the USD estimate.
Use estimates without treating them as promises
A useful estimate exposes its assumptions. The USD figure depends on a CTR, a purchase conversion rate, a USD commission per sale, and a refund rate. If the same clicks convert at only , the after-refund estimate falls to:
That difference is why improving reach alone cannot guarantee earnings. Audience intent, product fit, purchasing access, and the merchant’s checkout experience can all affect whether clicks become eligible sales.
For a global audience, one overall conversion estimate can also hide important differences. A product might be available in some countries but not others, or its price and eligibility may vary by market. When you have reliable data, estimate each audience segment using its own eligible clicks, conversion rate, commission, and refunds; then combine the amounts in a common currency. Until then, label the numbers as assumptions rather than expected income you can count on.
Finally, commission after refunds is not necessarily take-home profit. It does not account for possible payment fees, currency conversion, taxes, or promotion costs. Those belong in a later profitability calculation. For now, the goal is to forecast commission clearly and check each input against what the program actually reports.
Key takeaways
To estimate affiliate commission, start with content impressions, apply CTR to estimate affiliate-link clicks, apply conversion rate to estimate qualifying purchases, then multiply by commission per sale. If refunded orders lose commission, reduce the gross estimate by the expected refund share. Keep fractional sales during the calculation because they represent an average expectation, not a literal order count.
EPC helps compare what offers may earn from the same number of clicks, but an attractive estimate is only as reliable as its assumptions. Next, you’ll compare physical products, digital products, software, and services—categories whose prices, commission arrangements, and customer needs can differ substantially.