Amazon PPC PDF: What You're Really Looking For

Why the PDF Itself Is the Wrong Thing to Want

A PDF is a snapshot. Amazon Ads is not static — placements, match types, and auction mechanics shift every few months, and a document you saved in March can quote a bid range or a metric definition that's already wrong by summer. The people who search for a PDF usually want one of two things: something to read once and remember, or something to hand to a junior hire. Neither actually needs a file. Both need the underlying formulas and the judgment calls that sit on top of them, which is what the rest of this page is.

If you do want something to save for later, save the formulas and the worked example below, not a marketing PDF with a lead-capture form in front of it. Nothing here requires an email address.

The Four Numbers That Actually Run an Account

Everything in Amazon PPC comes down to four measurements, and most accounts only track one or two of them. Here's what each one tells you and where it lies to you if you use it alone.

• ACOS (ad spend ÷ ad sales) tells you campaign-level efficiency, but a low ACOS on a low-margin SKU can still lose money.
• ROAS (ad sales ÷ ad spend) is the same relationship flipped, useful for quick comparison across campaigns, blind to margin.
• TACoS (ad spend ÷ total sales) tells you how dependent the whole listing is on paid traffic — rising TACoS with flat organic rank is a warning, not just a cost line.
• Contribution margin after ads (revenue minus COGS, fees, and ad spend) is the only one of the four that tells you whether you actually made money.

A Worked Example, With Real Numbers

Take a $40 product. COGS is $12, Amazon fees (referral plus FBA) run $8, leaving a contribution margin of $20 per unit before ads — 50%. Over a week, the campaign spends $500 and generates $2,500 in ad sales.

ACOS is 500 ÷ 2,500 = 20%. ROAS is 5. So far it looks efficient. Total sales for the week, including organic, are $6,500, so TACoS is 500 ÷ 6,500 = 7.7% — the account isn't overly dependent on paid traffic. Now the number that matters: $2,500 in ad sales at $40 a unit is roughly 62.5 units. At $20 contribution margin each, that's $1,250 in gross contribution from ad-driven sales, minus the $500 spent to get it — $750 net. A 20% ACOS on this SKU is profitable. On a SKU with a $6 margin instead of $20, the same 20% ACOS would lose money outright. The ACOS number never tells you that. Only the margin math does.

The Structure Underneath the Numbers

Campaign structure exists to make these numbers legible, not to look organized in a screenshot. At minimum, separate branded from non-branded, exact match from broad discovery, and hero SKUs from everything else. A campaign launching a new ASIN and a campaign defending a branded term should never share a budget or a target ACOS — they're doing different jobs and judging them the same way hides the problem instead of fixing it.

One discipline separates accounts that scale from accounts that stall: every bid or budget change should carry the evidence behind it, a measurement plan, and a rollback trigger before it goes live. If a change doesn't have all three, it's a guess wearing a spreadsheet.

The Mistake We See Most Often — and Have Made Ourselves

The most common failure isn't a bad bid. It's judging a launch campaign, a defense campaign, and a harvest campaign against the same blended ACOS target. A launch campaign is supposed to look inefficient early — it's buying search-term data and early conversion signal, not profit. Killing it at week two because ACOS is high throws away the exact information you paid for.

We've made this mistake too: letting a broad-match discovery campaign run through a weekend with a bid ceiling set for a mature, exact-match campaign. The fix wasn't turning broad match off — it's a lower ceiling, a shorter negative-keyword review cadence, and a hard budget cap while the campaign is still proving terms, not scaling them.

When the Number Comes Back Wrong

Sometimes the fix doesn't work. ACOS stays high after you've trimmed the obvious waste. Here's the order to check things in, not the order most audits suggest:

• Check the target first, not the bid. A 15% ACOS target on a SKU with an 18% margin was never achievable — the target was wrong, not the campaign.
• Check the listing before the campaign. Traffic converting below category average usually means an image, price, or review problem, not a keyword problem.
• Check for cannibalization. Rising ad sales with flat total sales often means paid clicks are replacing organic ones you'd have gotten anyway.
• If none of that explains it, the search terms themselves may be wrong — pull the search term report before touching another bid.