67% of Amazon Sellers Are Losing Margin, and Nobody’s Blaming the Right Thing

Two-thirds of Amazon sellers report a margin decline over the past twelve months. Faced with that number, the first question asked in nearly every account I’ve looked at is: did our advertising cost per click go up? The answer is often yes, and that’s exactly what makes the wrong diagnosis so tempting.

Cost per click has genuinely gone up on Amazon Ads in several categories. Platform fees have climbed too. Those are real facts, but they aren’t the main cause of the margin decline in most of the cases I’ve analyzed. They’re its most visible symptom.

The real culprit is structural, not tactical

A seller who depends entirely on Amazon for revenue has no negotiating leverage against a fee increase or an algorithm change. They can’t redirect budget elsewhere because they never built another channel capable of absorbing volume. They’re captive, and a captive buyer never has negotiating power, whether facing a supplier or a platform.

That dependence, not cost per click itself, explains why the margin decline hits some sellers hard and barely touches others in the same product category. The sellers holding up are almost always the ones generating a meaningful share of revenue elsewhere: an owned brand with its own storefront, presence on other marketplaces, social discovery channels like TikTok Shop that feed Amazon search without relying on its advertising.

Why the wrong diagnosis persists

Blaming cost per click is comfortable because it’s a problem you can measure easily and believe you can fix by adjusting bids. Blaming structural dependence is uncomfortable because it requires admitting you built an entire business on one platform’s rules, and the fix takes months, not an afternoon of tweaks in an ad console.

That difference in comfort explains why so many sellers keep optimizing their ad campaigns with ever finer precision while their margin keeps sliding. They’re optimizing the wrong variable. A business pulling 95% of its revenue from a single channel stays fragile even with the best ad campaigns in the industry, because its fragility comes from structure, not execution.

What the sellers protecting their margin are doing

The cases where margin holds up share a clear pattern. These sellers first built a direct-sales capability, often modest at the start, that lets them sell to a customer without going through Amazon’s fees and rules. Next, they added at least one other marketplace or distribution channel, not necessarily to match their Amazon volume, but to have somewhere to redirect effort if conditions worsen on the main platform. Finally, they invest in discovery channels that don’t depend on paid advertising, like creator partnerships, which feed sales across every platform at once instead of just one.

None of this removes Amazon from the equation. The platform often remains the highest-volume channel for most product categories. The goal isn’t to leave it. It’s to stop being its hostage.

An example that shows the problem clearly

A pet accessories seller pulled nearly all of its revenue from Amazon since launch. When the platform tightened compliance rules on one accessory category, several of its best-performing listings got suspended for three weeks while it fixed its documentation. Three weeks without its main channel, with nowhere else to redirect demand, did more damage to its annual margin than any cost-per-click increase could have done in a full year.

This seller hadn’t done anything wrong on the advertising side. Its campaigns were well built, its acquisition cost was in the right range for its category. The problem only became visible the moment the platform changed a rule it didn’t control, and at that point, no bid adjustment could have made up for having zero backup channel.

When cost per click really is the problem

It would be dishonest to claim cost per click is never the real problem. It is, in certain specific cases: an account whose campaign structure hasn’t been reviewed in years, with poorly segmented ad groups and bids that were never adjusted to actual per-product margin rather than to an overall volume target. In that kind of account, a rigorous ad audit can genuinely recover a meaningful share of lost margin, and it needs to happen before concluding anything else.

The way to tell the two situations apart is simple: if a clean ad audit and a bid restructure bring margin back to an acceptable level, the problem really was tactical. If margin keeps declining despite ad management that’s already rigorous, or if it recovers temporarily and then slides again the moment the platform changes a rule, the problem is structural, and no additional bid adjustment will fix it. Most sellers skip that diagnostic step and go straight to bid tweaking because it’s the most available action, not because it’s the right answer to their specific situation.

How to know if you’re exposed

The simplest test: if Amazon changed its fees or algorithm tomorrow morning in a way that cut your visibility in half, how long would your business survive on the revenue left elsewhere? If the answer is close to zero, the real priority isn’t the next ad campaign. It’s building a second and third revenue channel, even modest ones at first.

One more distinction matters here: structural dependence isn’t the same thing as being small. A seller can be small and diversified, or large and completely captive to one platform. Size is not the variable that predicts who survives a rule change or a fee increase. The share of revenue that comes from a single source is, regardless of how big that revenue number is.

The margin decline two-thirds of Amazon sellers are living through isn’t primarily an advertising problem. It’s a revenue-structure problem. Continuing to optimize bids while that structure stays fragile treats the most visible symptom while ignoring the cause that will keep producing the same result next year.

There’s a version of this argument that sounds like advice to abandon Amazon entirely, and that’s not what this is. For most product categories, Amazon still delivers more purchase-ready volume per dollar of effort than any channel a seller could build from scratch. The point isn’t to walk away from the channel that works. It’s to stop letting it be the only one that does, because a single point of failure eventually fails, whether through a fee increase, an algorithm change, or a compliance rule nobody saw coming.

Diversifying doesn’t mean building everything at once. A seller who only has Amazon today doesn’t need to launch their own storefront, a second marketplace, and a creator program in the same quarter. They need to pick a single backup channel and give it enough time to become real, even at small scale, before adding a second one. The goal isn’t perfect diversification. It’s no longer depending on a single player for the entirety of your revenue.