Your branded keywords are starving

The highest-ROAS traffic in almost every Amazon account is the brand’s own name. The lowest-funded campaigns in almost every Amazon account are the ones running on the brand’s own name.

This is the inversion most agencies don’t catch because they aren’t structurally looking for it.

Here’s what it looks like in real data.

A skincare brand I audited had spent over $113,000 in ads over ten months at a 125% ACOS. They were burning $1.25 for every dollar earned. Looked like a complete account failure.

Then I broke the spend down by intent. Their three branded terms had spent a combined $614 and generated $5,146 in attributed sales. Branded ROAS was over 8x. Non-branded ROAS was under 0.8x. They had effectively starved the only thing that was working in favor of bidding aggressively on terms that didn’t convert.

An energy drink brand showed the same pattern at scale. Branded ACOS was 19%. Non-branded ACOS was 125%. They were spending more than 70% of their budget chasing a 0.8x ROAS while their own brand defense was the cheapest, highest-converting traffic in the entire account.

Same pattern in apparel. Same pattern in baby goods. Same pattern in deodorant. Different categories, same structural inversion.

Why this keeps happening

Two reasons.

First, agencies sell growth. Branded campaigns aren’t growth. They’re defense. The pitch deck doesn’t have a slide that says “we’ll protect the customers you already have from being intercepted by competitors.” It has a slide that says “we’ll find you new customers.” So budget flows to non-brand acquisition by default, even when the math says brand defense is the better dollar.

Second, branded campaigns are politically uncomfortable. The brand owner sees an ad campaign running on their own brand name and asks “wait, are these customers we’d have gotten for free anyway?” The answer is partially yes. The agency doesn’t want that conversation. So the campaigns get underfunded.

The second point is the more honest one. Brand defense converts so well partly because the customer was going to buy regardless. The increment is the customer you didn’t lose to a competitor bidding on your name. That increment is real and it’s measurable and most agencies don’t measure it because measuring it forces them to charge less.

What it costs

The math on most accounts is: $5,000 per month behind structured branded campaigns at sub-20% ACOS would generate $40,000 per month in attributed revenue. That’s a 7-to-1 return.

Most accounts get a fraction of that because the campaigns aren’t set up properly. Sometimes the branded campaigns run in the same bucket as non-brand and get pulled down by the average. Sometimes they exist but they’re capped at $200 a day for no defensible reason. Sometimes they don’t exist at all, on the theory that “if they’re searching for our brand they’ll find us anyway.”

The “they’ll find us anyway” assumption is the most expensive assumption in the entire account. Competitors bidding on your branded terms intercept a measurable portion of buyers who searched for you specifically. The number varies by category but it’s never zero.

What to do

Three steps.

First, pull the spend breakdown by branded vs non-branded keywords. Look at the ACOS for each. If branded ACOS is under 25% and non-branded is over 80%, you’re underfunding the wrong layer.

Second, set up structured branded campaigns separately from non-brand. Different campaigns. Different budgets. Different targets. Pull them out of the blended bucket so they can be measured and scaled on their own.

Third, fund them appropriately. If branded ACOS is 19% and non-branded is 125%, the marginal dollar should be going to branded until you cap the demand. Most accounts have $3,000 to $10,000 per month of unmet branded demand sitting there.

Your branded keywords don’t need a campaign overhaul. They need a budget overhaul.