Is KDP Still Profitable in 2026? Seller Claims Tested
The 'KDP is dead' claims mix real problems with outdated ones. Here is which hold up against Amazon's royalty structure, AI disclo…
Impressions and clicks feel like progress, but only two numbers on your ad report tell you whether a campaign made or lost money: ACOS and your book's break-even ACOS. Here is how to separate the metrics worth watching from the ones that just make the screen look busy.
Open the Advertising tab inside KDP and you land on Amazon's Sponsored Products reporting console, re-skinned for books. Every campaign, ad group, and targeted keyword gets a row, and each row carries the same set of columns: Impressions, Clicks, Click-Through Rate, Spend, Cost-Per-Click, Orders, Sales, and ACOS. You can filter by date range — last 7 days, last 30, or a custom window — and sort by any column.
Most authors scan left to right and stop at whatever number is biggest or greenest. That habit is the entire problem. Two of those eight columns are cost inputs, three are engagement signals, and only one is a direct read on whether the campaign paid for itself. Knowing which is which changes how you act on a Monday-morning check of your kdp ads results — you stop reacting to noise and start reacting to signal.
One column trips up almost everyone the first time they see it: Sales. It is not your royalty. It is the list-price total of every unit Amazon credits to that ad — the same number a hardcover thriller selling for $24.99 would show if one copy sold. If you glance at Sales and feel good because it is bigger than Spend, you have compared your ad cost to revenue you do not keep, not to the money that actually lands in your KDP payment.
Here is what each column on your kdp ads dashboard measures, and whether it tells you anything about profit on its own.
| Metric | What It Actually Measures | Vanity or Profit Signal | Read It Like This |
|---|---|---|---|
| Impressions | How many times the ad was shown | Vanity | Reach only — tells you nothing about buying intent |
| Clicks | How many times someone clicked through | Vanity | Traffic you paid for before it earned anything back |
| CTR | Clicks divided by impressions | Mostly vanity | A relevance score for your cover and targeting, not a profit score |
| CPC | Average price paid per click | Cost input | Feeds the spend side of the equation, says nothing about the sales side |
| Spend | Total dollars paid out over the report window | Cost input | Meaningless on its own — it needs a royalty number to compare against |
| Orders | Units Amazon attributes to the ad | Progress signal | Confirms the ad converts; still doesn't confirm it's profitable |
| Sales | List-price value of attributed orders | Misleading if read as profit | This is revenue at cover price, not the royalty you're paid |
| ACOS | Spend divided by Sales, times 100 | Profit signal | Spend as a percent of sale revenue — the only number here that predicts profit |
Notice the pattern: five of those eight are describing activity, not outcome. Activity matters for diagnosing a problem — a campaign with plenty of impressions and almost no clicks has a targeting or cover issue, not a pricing issue — but activity numbers cannot tell you whether the campaign is worth running. Only ACOS, compared against a target you calculate yourself, does that.
Generic advertising advice says an ACOS under 30% is healthy. That rule comes from product categories with 50% or higher profit margins, where an advertiser can absorb a 30-cent ad cost on every dollar of revenue and still walk away ahead. A paperback rarely works that way. KDP pays 60% of list price minus the printing cost on Amazon marketplaces, and 40% through Expanded Distribution — check the current rate on KDP's royalty help page before you price, since Amazon has adjusted low-price bands before. After printing cost, a typical paperback's real margin against list price often sits well under the 50% a generic ACOS benchmark assumes, which means the "under 30%" rule of thumb can still lose you money on a book.
The number that actually applies to your book is your break-even ACOS: the ACOS at which ad spend exactly equals the royalty you earn from the sales that ad produced. The formula is simple once you have your royalty per copy:
Break-even ACOS = (Royalty per copy ÷ List price) × 100
Anything below that number means the campaign nets you money after royalty. Anything consistently above it means you are buying sales that cost more than they pay back. Pull your exact royalty figure from KDP's own royalty report or run it through a free royalty calculator before you do this math — printing cost changes with trim size and page count, so a number you eyeballed last month may already be stale.
| List Price | Royalty per Copy | Royalty Margin | Break-Even ACOS |
|---|---|---|---|
| $6.99 | $2.50 | 36% | 36% |
| $8.99 | $3.00 | 33% | 33% |
| $12.99 | $4.50 | 35% | 35% |
| $16.99 | $6.00 | 35% | 35% |
These figures are for illustration, built from royalty numbers you would plug in from your own book, not a published Amazon rate. Notice that a 35% break-even ACOS is close to the generic "under 30% is fine" advice, which is exactly why authors following that generic number often run campaigns that look healthy and quietly lose money. Run your own book's numbers before you set a bid ceiling.
One more wrinkle: your break-even ACOS is a per-book ceiling, not a per-keyword target. A keyword sitting at 55% ACOS inside a campaign that averages 25% is not automatically a loser — check its Orders count first. A single early sale on a new keyword can post a triple-digit ACOS purely from small-sample noise.
Click-Through Rate and Cost-Per-Click get more attention than they deserve because they update fast and feel actionable. Both are useful diagnostics. Neither one, alone, tells you if you made money.
Your cover, title, and the keyword or product you're targeting are aligned enough that people who see the ad want to look closer. A cozy-mystery cover on a cozy-mystery keyword search will usually out-click a generic thriller cover on the same search.
Whether those clickers bought. A curiosity-driven click from an irrelevant but visually loud cover can post a great CTR and a zero-order campaign. CTR measures attention, not intent to buy.
You are winning the auction cheaply, usually because the keyword has less advertiser competition. Long-tail, specific keywords tend to run cheaper than broad genre terms.
Whether the traffic converts. A $0.15 click that never orders still adds up, and a $0.90 click that reliably orders can be the better keyword. Compare CPC against orders, never in isolation.
The practical use of CTR and CPC is triage, not verdict. A campaign with low CTR across the board usually has a targeting problem — you're showing up on searches only loosely related to your book, and tightening your keyword list with real keyword research fixes it faster than raising bids. A campaign with decent CTR but no orders usually has a pricing or listing problem downstream of the click, not an ad problem at all.
Here is the step where almost every new advertiser talks themselves into pausing a campaign that was about to work: they launch a campaign, check the dashboard the next morning, see Spend with zero Orders and an ACOS field showing nothing or an error state, and shut it down as a failure.
Amazon does not credit an order to an ad instantly. There is a reporting lag between a click and the order showing up attributed to that click, and same-day numbers on the dashboard are frequently still updating when you look at them. A campaign that has run for two days with a handful of clicks has not generated enough data to judge — you are reading noise, not a result. Give a new keyword or campaign real runway, generally on the order of one to two weeks and enough accumulated clicks to matter, before you decide it isn't working. Amazon Ads Help documents the exact attribution window for your report type; check it there rather than assuming same-day figures are final.
The second version of this mistake is seasonal, not structural. ACOS on the same campaign can swing hard between a slow midsummer week and a gift-buying spike in November and December, because both Spend and Sales move, just not at the same rate. Comparing a July ACOS to a December ACOS and concluding the campaign "stopped working" ignores that demand itself moved. If your book has an obvious seasonal audience — a planner, a holiday-themed activity book, a gift title — plan bid changes around that calendar instead of reacting to a single bad week.
Practically: resist adjusting a brand-new campaign daily. Check it on a fixed weekly cadence, filter the date range to at least seven days, and only make bid or targeting changes once a keyword has enough clicks or spend behind it that the ACOS you're looking at reflects a pattern rather than one lucky or unlucky sale.
Open the Advertising tab, pick your campaign, and set the report window to at least a week. A single-day view amplifies noise and hides the pattern you actually need.
This puts your biggest cost centers at the top, whether or not they've produced an order yet. You want to know where your money is going before you look at what it bought.
For every row with at least one Order, check its ACOS against the break-even figure you calculated for that book. Anything comfortably under is a keep; anything well over, with enough clicks to be meaningful, is a candidate to lower the bid or pause.
Any keyword or targeting row with meaningful Spend and zero Orders after your minimum runway is your pause list. This is where wasted ad budget actually hides — not in the campaigns with a high ACOS, but in the ones with no orders at all.
It's tempting to keep tinkering with bids on keywords already sitting under break-even ACOS. Resist it. Stable, profitable rows are the ones you want to leave untouched while you spend your attention on the flagged rows from step four.
There is no universal number — it depends on your book's royalty margin. Calculate your break-even ACOS as royalty per copy divided by list price, then treat that figure as your ceiling. A generic "under 30%" benchmark from broader ecommerce advice can still be a loss on a book with a thinner margin than that.
Two separate things are being confused. Organic sales — from browse, search, or a reader's own link — never show up in the ads dashboard at all, since it only reports orders Amazon attributes to a click on that specific ad. Recently attributed orders can also lag by a day or more before they post, so a brand-new campaign can look like it sold nothing for its first day or two while data is still settling.
CTR measures whether people who see your ad bother to click it — an engagement signal about your cover and targeting. ACOS measures what you spent as a percentage of what that click eventually generated in revenue — a money signal. A campaign can have a strong CTR and a terrible ACOS, or the reverse, because the two measure completely different stages of the funnel.
Give it real runway before judging it — generally at least a week of data and enough accumulated clicks that one sale or one dry spell doesn't swing the ACOS wildly. A keyword with three clicks and one lucky order will show a misleadingly low ACOS; a keyword with three clicks and no order yet will show an infinite one. Neither is a verdict yet.
Cost-per-click is set by an auction against other advertisers bidding on the same keyword or product, so it rises as more authors target that term or raise their own bids, and it can shift with seasonal demand. A rising CPC on a keyword that still clears your break-even ACOS is not a problem on its own — check the ACOS before reacting to the CPC alone.
Octozia's KDP royalty calculator is free, needs no account, and gives you the exact per-copy number this math depends on.