Affiliate
10Beasts
The full story of Convertica's conversion work on a well-known affiliate site, which started after a talk at an SEO conference.
Affiliate CRO
Affiliate marketing conversion rates: how to measure and raise them
Affiliate marketing conversion rates decide how much commission each visitor is worth. Convertica has optimized affiliate and review sites for years, and now offers a free CRO audit of your page, then a one page rebuild or a full site rebuild, where Convertica's team builds every fix the audit found.
Turn visitors and AI agents into customers. More than 2,000 clients since 2017.
Short answer
Affiliate marketing conversion rates measure how often affiliate traffic turns into a commission. A publisher has two rates: the share of page visitors who click through to a merchant, and the share of those clicks that become orders on the merchant's site. Multiplied together, with commission added, they give earnings per visitor, which you compare against your own baseline.
- A publisher controls only half the funnel, so judge changes on sales and commission, not clicks.
- Most lost commission comes from readers who cannot quickly see which product suits them, or cannot find the link once they decide.
- Split test with separate page copies and separate affiliate tracking IDs, as in Convertica's triple setup.
- A good affiliate conversion rate is one that beats your own baseline for the same kind of page and traffic.
- Write picks, criteria and first-hand testing as text, so AI answers can cite your review pages.
What are affiliate marketing conversion rates?
Affiliate marketing conversion rates measure how often affiliate traffic turns into a commission. For a publisher there are really two rates: the share of page visitors who click through to a merchant, and the share of those clicks that become orders on the merchant's site. Multiplied together, and with commission added, they give earnings per visitor, the number that matters most. The first of those two rates is a click-through rate, and how to increase click through rate covers the levers that move it on your own pages.
This page is written for affiliate site owners and publishers: review sites, comparison sites and content sites that earn when a reader buys through a link. Convertica's past clients include Authority Hacker and Geniuslink, and it has published affiliate case studies and its own method for split testing affiliate pages. If you are a brand running an affiliate program, the conversion that matters to you happens on your product pages and checkout: see eCommerce conversion optimization.
How to calculate affiliate conversion rates
Calculate affiliate conversion rates at three points: page visitors to merchant clicks (click-through rate), merchant clicks to orders (merchant conversion rate), and commission per page visitor (earnings per visitor). The first is mostly yours to change, the second mostly the merchant's, and the third tells you whether a change made money.
| Metric | Formula | Where the data lives | Who controls it |
|---|---|---|---|
| Click-through rate to merchant | Affiliate link clicks ÷ page visitors × 100 | Your analytics (outbound click events) | Mostly you: content, tables, buttons |
| Merchant conversion rate | Orders ÷ affiliate clicks × 100 | The affiliate network or program dashboard | Mostly the merchant, but which products you send people to matters |
| Earnings per click (EPC) | Commission ÷ affiliate clicks | Network dashboard | Shared |
| Earnings per visitor (EPV) | Commission ÷ page visitors | Analytics plus network data, matched by page | The result of everything above |
Worked example (made-up numbers)
A review page gets 10,000 visitors in a month. 1,200 of them click an affiliate link, a click-through rate of 12%. The merchant reports 48 orders from those clicks, a merchant conversion rate of 4%. So 0.48% of page visitors became a sale. Divide the month's commission from that page by 10,000 and you have its earnings per visitor, the figure to compare before and after any change. To turn those rates into commission, the affiliate commission calculator starts from the same 1,200 clicks at 4% and adds an order value and a commission rate.
Track clicks per page with a unique tracking ID or sub-ID where the program allows it, so network sales can be matched to the page that sent them. The free conversion rate calculator does the division for you.
What is a good affiliate conversion rate?
A good affiliate conversion rate is one that beats your own baseline for the same kind of page and traffic. Rates vary by niche, merchant, price, cookie length and how close the reader is to buying, and published averages measure different things. Compare a buying guide with other buying guides on your site, over time, and judge changes on earnings per visitor.
Why affiliate CRO is different from store CRO
Affiliate CRO is different because a publisher controls only half the funnel. The sale happens on a site you cannot change or put testing code on, the merchant reports it later and without saying which page or test version it came from, and a click is not money. That changes what you measure and how you test.
- Clicks can mislead. A louder button can pull more clicks from readers who have not chosen a product, and those clicks may not buy. In one Convertica affiliate test, a "top 3 picks" table did not raise clicks overall but did raise sales, because readers were better informed before they clicked.
- Sales data arrives late and in aggregate. Network reports group sales by tracking ID or sub-ID, so a test needs its own IDs to know which version earned what.
- Desktop and mobile behave differently. Comparison tables that work on a wide screen often break on a phone. In the same Convertica test series, one layout won on desktop while the other did better on mobile.
- Your traffic is mostly from search. Changes to the page can affect rankings as well as conversions, so test in a way that is safe for SEO.
Affiliate conversion rate optimization: where review pages lose sales
Affiliate conversion rate optimization starts with the page elements readers use to choose: the top picks box, the comparison table, the product sections and the buttons. Most lost commission comes from readers who cannot quickly see which product suits them, or who have to scroll back up to find the link once they have decided.
- No clear top pick. A short, visible summary of the best options near the top helps readers who have already done their research.
- A comparison table without the deciding facts. Readers need the one or two facts that separate the products. In one Convertica test on an Amazon affiliate site, a product rating was added above the call-to-action buttons in the comparison tables.
- The link is far away when the reader decides. On the same site, a sticky anchor button took readers back up to the comparison table once they had scrolled past it into the full reviews.
- Vague buttons. Say where the click goes and what the reader will see, such as the merchant and the current price check, not just "Click here".
- Tables that break on a phone. Check every table at phone width. Stacked cards often work better than a wide grid.
- Product sections with no link at the end. Put a link where each review ends, not only in the table.
- Slow pages. Heavy images, ad scripts and widgets slow review pages down, and a slow page loses readers before they reach the table.
- Hidden or unclear disclosure. Disclose affiliate links clearly and close to the links.
In two short videos from Convertica's YouTube channel, Kurt Philip explains why comparison tables should carry product images, and why readers should be able to choose a product before they leave for Amazon.
How to split test an affiliate site
To split test an affiliate site, give each version its own page copy and its own affiliate tracking ID, send equal traffic to each, and judge the result on sales reported by the network, not on clicks. Convertica developed a "triple setup" for this, because a standard 50/50 test can leave Amazon sales impossible to match to test versions.
The problem, as Convertica explained in its post on why affiliate site split tests fail: JavaScript testing tools cannot sort some visitors (for example, those with JavaScript or cookies disabled, or on slow connections), so those visitors all see the original page. The original gets more than its share, and the sales data no longer lines up with the split.
-
Duplicate the original page twice
One copy stays an exact duplicate and becomes the control. All changes go on the second copy, the variation.
-
Add a canonical tag on both copies
Point each copy's canonical tag at the original page, so search engines treat the original as the page to rank.
-
Give each copy its own URL and tracking ID
A unique affiliate tracking ID per copy means the network reports the sales from the control and the variation separately.
-
Split traffic 45, 45 and 10
Send 45% of traffic to the control copy, 45% to the variation and 10% to the original, as a split-URL test. Visitors the tool cannot sort land on the original, which is left out of the comparison.
-
Track clicks to the merchant as the test goal
Set the goal on clicks to the merchant's links, then compare sales and commission per visitor for the control and the variation from the network reports.
-
Run it to the planned sample size
Sales are rare compared with visits, so tests take time. Set the sample size before launch with the A/B test sample size calculator and check the result with the statistical significance calculator.
A split test needs traffic. On a smaller site, fix the clear problems in the list above directly and save testing for your highest-earning pages.
Affiliate page conversion checklist
Use this checklist on your top-earning review or comparison page. Each "no" is something to fix or test.
- Can a reader who is ready to buy see your top pick without scrolling far?
- Does the comparison table show the facts that separate the products?
- Is there a link at the end of every product section?
- Can readers get back to the table or top pick once they have decided?
- Do buttons say where the click goes?
- Does every table work at phone width?
- Is each high-earning page tracked with its own tracking ID or sub-ID?
- Do you judge changes on earnings per visitor, not clicks?
- Is your affiliate disclosure clear and close to the links?
- Can an AI assistant read your picks, the criteria and the reasons as text on the page?
Affiliate sites and AI answers
AI answers in search and in assistants now respond to many "best product for" questions directly, sometimes citing the review pages they drew on. For a publisher that means two jobs: earn the citation with clear picks, criteria and first-hand testing written as text, and convert the smaller number of readers who still click through, who tend to arrive closer to buying.
Agent readiness is one of the eight checks in Convertica's free CRO audit. For being found and cited in AI answers, see AI search visibility. This is a young field with little long-term data, and we say so.
Proof
Affiliate case studies and what clients say
Convertica has published several affiliate case studies, from split tests on Amazon affiliate review pages to a full write-up of its work on 10Beasts. The quotes below are verbatim from the homepage and Trustpilot.
Affiliate
Product ratings and a sticky anchor
An Amazon affiliate site in the home heating and cooling niche: ratings above the comparison table buttons, and a button back to the table.
Affiliate
A top 3 picks table
An Amazon affiliate review site: a vertical top 3 picks table tested against an at-a-glance list, where sales rose even though clicks did not.
Convertica increased revenue on our site by over 30%. Their stuff works.
I have used Convertica for a couple of affiliate sites and have been really impressed with the quality of work. Highly recommended for any affiliate looking to increase conversion rates.
I hired Convertica initially for three months and then for another three. Their team helped me optimize a number of top pages on my site and increase affiliate revenue.
For one article, they more than doubled earnings. They are experts at optimization and free me up to work on other areas of the business.
I'd highly recommend Convertica's services to any content creator, marketer or even a blogger who wants to get maximum value from existing traffic. I plan to work with them again.
Great results. easy to work with, and do proper split-testing. Even noticed when I accidentally made a change to a page they were testing, and they stopped the test and started again to make sure I didn't distort results. The most important thing to know: they had a positive result on revenue.
The offer
How Convertica helps affiliate sites now
Many of the case studies above were split tests run by Convertica's team. Today the offer is a free CRO audit of your page, then a one page rebuild or a full site rebuild, where Convertica's team builds every fix the audit found.
The free CRO audit
An app with eight checks for people and AI agents, including calls to action, mobile, speed and agent readiness. Enter your website and watch it run live on your page.
The rebuilds
Convertica's team builds every fix the audit found, on your site: the one page rebuild covers one page and the full site rebuild covers your whole site. Both are paid by card through Stripe Checkout, at the end of an audit report or straight from this site with no audit first: Convertica runs the audit after you order.
Affiliate conversion rate FAQ
What are good affiliate marketing conversion rates?
Good affiliate marketing conversion rates are ones that beat your own baseline for the same page type and traffic source. Rates vary widely by niche, merchant, price and traffic intent, and published averages use different definitions, so compare a page with its own history and with similar pages on your site.
How do you calculate an affiliate conversion rate?
Divide conversions by clicks or visitors, then multiply by 100, and say which one you used. Clicks to the merchant divided by page visitors is your click-through rate. Orders divided by those clicks is the merchant conversion rate. Commission earned divided by page visitors is earnings per visitor, the number that combines both.
What is affiliate conversion rate optimization?
Affiliate conversion rate optimization is the work of earning more commission from the visitors an affiliate site already has. It improves how review and comparison pages help readers choose and click through to the right merchant, and it judges every change by commission or sales per visitor rather than clicks.
Why did my affiliate clicks go up but sales stay flat?
Usually because the extra clicks came from readers who were less ready to buy, such as a louder button pulling people who had not chosen a product yet. Convertica has seen the reverse too: a change that left clicks flat raised sales, because readers were better informed before they clicked. Judge changes on sales.
Can you A/B test an Amazon affiliate site?
Yes, but a standard 50/50 test can mislead, because Amazon reports sales by tracking ID, not by test variation, and visitors the testing tool cannot sort all see the original page. Convertica's triple setup gives the control and the variation their own page copies and their own tracking IDs.
How long does an affiliate split test take?
Longer than most site owners expect. Sales are a small share of clicks, merchant reports can lag, and the test needs enough sales in each version to show a real difference. Set the sample size before the test starts, and do not stop it early because one version pulls ahead.
Does Convertica work with affiliate marketing companies?
Convertica works with affiliate site owners and publishers, the sites that earn commission. Authority Hacker and Geniuslink are among its past clients. Brands that run their own affiliate program are better served by Convertica's eCommerce conversion work on product pages and checkout.
Does Convertica build the changes for my affiliate site?
Yes. After the free CRO audit, the one page rebuild has Convertica's team build every fix the audit found on that page, and the full site rebuild covers your whole site. Or you, your developer or your virtual assistant builds the changes from the audit.
