Substack Affiliate Marketing: Turn Readers Into Buyers

Substack is an unusually good place to run affiliate recommendations and an unusually easy place to destroy the trust that makes them work. The mechanism that makes it good is the same one that makes it fragile: people read newsletters in a personal context, with an expectation of editorial voice rather than commerce.

This is a guide to doing it without wrecking the thing that made it work.

Why the economics are better than they look

The obvious comparison is subscriber count against a social following, and by that measure newsletters look small. That comparison is wrong.

• Open rates on an engaged newsletter dwarf the reach rate of any social platform. A meaningful fraction of your list sees each send. A meaningful fraction of your followers does not see each post.


• The context is high-trust. Email arrives in a space the reader controls, from a sender they chose. A recommendation there carries weight that a feed post does not.


• You own the relationship, partially. Substack lets you export your list, which distinguishes it from platforms where the audience is not portable. This matters enormously and is worth checking on any platform before you build on it.


• Recommendations compound within the network. Substack's recommendation system between publications is a genuine growth mechanism.

The result is that a few thousand engaged newsletter readers can out-earn a much larger social following, provided the recommendations are good.

What Substack allows

Substack permits affiliate links. The practical constraints are about presentation rather than permission:

• Disclosure is required by law regardless of platform policy, and belongs near the recommendation rather than in a footer.


• Paid subscription content and affiliate content can coexist, but readers paying you for content react badly to feeling sold to. Be deliberate about which posts carry recommendations.


• Substack takes a percentage of paid subscriptions, not of affiliate revenue. Affiliate income is often the better-margin line for a mid-sized publication.

The formats that work

The embedded recommendation

A product mentioned inside a piece that is about something else, because it is genuinely part of the answer. This is the highest-converting and least intrusive format, and it only works when the recommendation is incidental to a genuinely useful piece.

The tools issue

A periodic post covering what you actually use. Works because the framing is honest: this is what I use, here is why, here is what I stopped using and the reason.

The version that fails is the one that lists everything positively. If a tools post has no negative opinions in it, readers correctly infer it is an ad.

The deep review

One product, examined properly, including who should not buy it. Converts at the highest rate of any format because it does the work the reader was going to have to do.

The requirement is that you actually used the thing. Reviews assembled from other people's reviews are transparent to readers and worthless as recommendations.

The comparison

Two or three options, with a clear statement of which situation each suits. Excellent for search discovery if your posts are public, and excellent for conversion because the reader arrives already deciding.

What does not work

Roundups of twenty products with a line each. No trust is transferred because no judgment was exercised, and the reader can tell.

Getting the links to work

The tracking problem in newsletters is specific and costs people money.

Email platforms rewrite links for click tracking. If that rewrite strips query parameters, your affiliate identifier never arrives and the sale is credited to nobody. This is the most common silent failure in newsletter affiliate marketing.

Test it before you rely on it:

1. Send yourself a test issue containing a tracked affiliate link.
2. Click it from the received email.
3. Check the final URL after all redirects resolve.
4. Confirm your identifier survived.

If it did not, route links through your own domain so the identifier sits in the path rather than the query string. This also gives you a first-party click record, which is the only reliable basis for knowing which issue produced which sale.

The trust arithmetic

The thing worth understanding is that every recommendation spends a small amount of accumulated credibility, and a good recommendation returns more than it spent.

That framing makes the decisions obvious:

• Recommend rarely enough that each one is an event. A newsletter with a promotion in every issue has taught readers to skip them.


• Recommend things you would recommend without a commission, and say so.


• Name the drawbacks. Specific criticisms are the clearest signal that you are not simply reading a brief.


• Say when something is not for the reader. Telling a segment of your list not to buy something is the single most credibility-building thing you can do, and it improves conversion among the segment it does suit.


• Follow up honestly. If you recommended something and it disappointed you later, say so. This is rare and readers remember it.

Recurring beats one-off

For a newsletter specifically, recurring commissions are worth disproportionately more than the headline rates suggest.

A newsletter audience is stable and grows slowly. A one-off commission means the same subscriber can only pay you once. A recurring commission on a subscription product means your income grows with the accumulated history of your recommendations rather than with this month's send.

Software, memberships, hosting and subscription services are where this applies. Over two years, a twenty percent recurring commission usually beats a forty percent one-off substantially. The recurring revenue models piece covers the comparison in detail.

Growth, which is the actual constraint

Affiliate income scales with subscribers, so the honest answer to "how do I earn more from Substack affiliate marketing" is usually "get more of the right subscribers".

What works on Substack specifically:

• The recommendation network. Recommending publications you genuinely read, and being recommended in return, is the highest-yield growth channel on the platform.


• Public archives and search. Posts that stay public accumulate search traffic over years. Paywalling everything removes this entirely.


• Guest posts and cross-posts with publications serving the same reader.


• One genuinely excellent piece that circulates, which does more than a year of consistent adequate ones.

Measuring it

Substack gives you opens and clicks. Neither is revenue.

What you want is revenue per subscriber and revenue per issue, which requires connecting the click on your side to a settled sale on the merchant's side. Without that, you will optimise for clicks, and the issues that generate the most clicks are frequently not the ones that generate the most money.

For the underlying calculation, revenue per visitor covers the method, and best affiliate email marketing platforms covers the tracking configuration in more depth, including the link-rewriting problem above.

The failure mode to avoid

The predictable way this goes wrong is gradual. Recommendations get slightly more frequent, criticism gets slightly softer, and the sponsor-friendly framing creeps in. Nobody unsubscribes over any single issue. Open rates drift down over months and the newsletter quietly stops working.

The defence is a rule you set in advance and keep: a fixed maximum frequency of recommendations, and a standing commitment that every review names something the product does badly. Both are easy to hold and both are what the whole model depends on.