Back to Blog
GuidesNovember 5, 2026· 7 min read

How Much Should You Spend on KDP Ads? Budget From the Royalty, Not From a Number

I

InteliAds Team

Amazon KDP PPC

How Much Should You Spend on KDP Ads? Budget From the Royalty, Not From a Number

Short answer: there is no right monthly number. One real KDP account ran profitably at $818.96 in a month and at $2,329.28 in another. What made both right was the same rule: the budget was set from how many orders the books could afford to buy, and ads were kept under the ACOS the royalties allowed. This article shows that arithmetic with real figures.

The wrong way, and why it is so common

Most authors start with a number: $5 a day, $10 a day, $300 a month, because someone in a group said so. Then they judge the result against a feeling. A budget chosen this way has no answer to the only question that matters: is each order making or losing money? Amazon will happily spend $300 a month on a book that loses $1.50 per sale, and the dashboard will show "sales".

Two real months as bookends

December 2025October 2026
What was runningOne US profile, one seriesWhole catalogue, launch month
Ad spend$818.96$2,329.28
Orders303388
Cost per order$2.70$6.00
ACOS16.81%35.5%
ACOS ceiling the royalties allowunder 47%under 47%
Royalties after ads (October)$3,896.07 earned, $1,566.79 net

October spent 2.8 times more and paid 2.2 times more per order, and it was still the right budget: the ACOS stayed inside the ceiling, and the month closed $1,566.79 ahead after ads. December was tighter because a mature series with reviews converts better. Neither month would have been improved by copying the other's budget.

The three-step budget

1. Break-even ACOS, per book

Net royalty divided by list price. A $14.99 paperback with a $4.50 print cost at 60% nets $4.49, so its ceiling is 30%. A $4.99 ebook at 70% nets about $3.39, ceiling 68%. Use the break-even ACOS calculator for each book you plan to advertise; the budget belongs to the books with the highest ceilings.

2. Expected cost per order

Until you have your own data, use about $0.86 a click and five clicks an order (this account's three-year averages), so roughly $4.30 per order. A new book with no reviews will cost more; October above was $6.00. A book with 50 reviews and a strong cover may be nearer December's $2.70.

3. How many orders you want to buy

Budget = orders x cost per order. Want 100 ad-attributed orders a month on a book with a 30% ceiling? At $4.30 each that is $430, and it only works if the royalty on 100 orders ($449) is above the spend. It is, by $19. Want 100 orders on the $3.39 ebook? Same $430 spend, $339 royalty, a loss of $91 unless the ebook is the door to a series.

Worked example: a three-book series

  • Book 1, $2.99 ebook, nets $2.03. Ceiling 68%, but at $4.30 an order it loses $2.27 per sale on its own.
  • Books 2 and 3, $4.99 each, net $3.39 each. If 40% of book 1 buyers read on to book 2 and 70% of those to book 3, one book 1 sale is worth $2.03 + $1.36 + $0.95 = $4.34.
  • So the series as a whole breaks even at $4.34 per order, just above the $4.30 expected cost. Advertise book 1 only, budget 60 orders a month = about $260, and measure the read-through monthly. If read-through is below 40%, the budget is too high, whatever the ACOS on book 1 says.

Launch versus steady state

A launch month should cost more per order and you should let it. October cost $6.00 an order because new titles had no reviews and were bid up on purpose to get them. The ceiling, not the budget, is what you protect: October's 35.5% ACOS was still under 47%. Had it crossed the ceiling, the rule on the account would have cut bids regardless of the launch. Plan a launch at roughly double the steady-state cost per order for 4 to 6 weeks, then expect it to fall as reviews arrive.

The monthly checkpoint

  1. Spend and orders for the month, per book. Cost per order against each book's ceiling.
  2. ACOS against the ceiling. Over it two months in a row means the budget is buying losses.
  3. Keywords with spend and zero orders in the last 60 days, paused. On this account that filter has caught 1,281 keywords and $2,277.91 over three years, 4.6% of all spend; running it monthly keeps the budget honest.
  4. Only then: raise or lower the budget, and by orders, not by a round number.

FAQ

Is $5 a day enough for KDP ads?

At about $4.30 an order it buys one order a day, which is enough to learn whether a book converts, and not enough to judge keywords quickly: a keyword needs 10 or more clicks before its zero orders mean anything, and at $5 a day across 20 keywords that takes weeks. Fewer keywords, or more budget, not both small.

Should I spend more during a launch?

Yes, and expect a worse cost per order while you do. Protect the ACOS ceiling, not the daily figure, and set a date to review.

When do I stop increasing the budget?

When the next increase raises cost per order faster than it raises orders, or when ACOS reaches the ceiling. Amazon does not run out of clicks; your royalty runs out of room.

Sources: InteliAds dashboard on the founder's KDP account, monthly views December 2025 and October 2026, all-time account overview read 29 October 2026. One account, stated as such. Royalty examples are illustrative, not customers. Related: Amazon Ads for authors: what they cost, and when they are worth it.

Put these rules to work

Set IF/THEN conditions, run them every hour, and revert any batch when you need to.