How to Choose KDP Categories: A Step-by-Step Method
A repeatable method for choosing your three KDP categories by relevance, buyer fit and competition, with a worked example and a ve…
A KDP ads agency takes over your Amazon Ads campaigns for a monthly fee on top of your ad spend. Here is exactly what that fee buys, the breakeven math behind it, and when doing it yourself wins instead.
Strip away the sales page language and a KDP-focused ads agency does five things for its retainer: it builds and launches your Sponsored Products and Sponsored Brands campaigns inside Amazon Ads, it reviews the search term report on a schedule, it adds negative keywords to stop wasted clicks, it reallocates budget toward whatever is converting, and it sends you a report translating the numbers into plain English. That is the whole job description, whether the invoice says $400 a month or $2,000.
Most agencies working with authors are small operations — two or three people managing dozens of accounts, not a specialized publishing division of a holding company. A few run KDP campaigns as one service line inside a broader Amazon-seller ads shop, which means the person touching your account may be more familiar with FBA product listings than with book categories, KENP reads, or how a 99-cent price point changes campaign economics. Ask directly who will handle your account day to day, not just who signed the contract.
What an agency does not do, no matter how the pitch is worded: it does not write better ad copy than your book cover and blurb already communicate, it does not fix a weak category, and it cannot make Amazon's own algorithm favor a book that is not converting once a shopper clicks through. If the underlying problem is category or niche selection rather than ad delivery, an agency will keep spending your budget without solving it. Ad management moves traffic toward a listing. It does not repair a listing that is not converting that traffic into sales.
Here is the arithmetic a retainer has to beat before it is worth paying. Take a $9.99 paperback in a standard 6x9 black-ink trim, roughly 200 pages. On Amazon marketplaces, KDP pays 60% of list price minus the printing cost, with 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 and the split you see there may differ from this one by the time you read it.
60% of $9.99 is $5.99. KDP's published printing-cost formula for a standard trim, black-ink interior (as of September 2026, confirm the current fixed and per-page fees on KDP's printing cost page) works out to roughly $0.85 fixed plus $0.012 per page. At 200 pages that is about $3.25 to print. Subtract the printing cost from the royalty share: $5.99 minus $3.25 leaves about $2.74 net per copy sold, before any ad spend.
That $2.74 is the ceiling. If an ad click costs you more than $2.74 in total ad spend per resulting sale, the book is losing money on that sale, full stop. Expressed as ACOS — ad spend divided by ad-attributed sales revenue — $2.74 of royalty against a $9.99 sale price means you can spend up to roughly 27-28% of the sale price on ads and still break even. Above that ACOS, every additional sale the ads generate costs you more than it pays back. Below it, the campaign is profitable on royalty alone, before counting any halo effect on organic rank.
An agency retainer sits on top of that ceiling, not inside it. A $500-a-month flat fee needs roughly 182 profitable book sales a month (500 divided by 2.74) just to cover its own cost, before the agency's ad spend recommendations have paid you back a cent. For an author with one or two titles generating a few sales a day, that math rarely closes. For a publisher running 30-plus active titles with real ad budgets behind them, spreading that same $500 across the catalog is a much smaller ask per title, and the time saved reviewing dozens of accounts by hand becomes the real justification, not the ad performance itself.
The ranges in the table below reflect what KDP-focused ad agencies commonly quote in seller forums and public rate sheets as of September 2026, not a single published price list — get any specific agency's actual fee and minimum spend in writing before comparing it to these figures.
| Factor | Typical KDP Ads Agency | Managing It Yourself |
|---|---|---|
| Monthly fee | $400-$2,000 flat retainer or 10-20% of ad spend, on top of the ad spend itself | $0 — your time is the cost |
| Typical minimum ad spend required | Often $1,000-$3,000/month to take on a new account | None — you can run a $5/day campaign |
| Contract length | Commonly a 3-6 month minimum before you can leave | None, stop or pause anytime |
| Who holds campaign access | The agency, via an invited user role on your Amazon Ads account, unless you insist otherwise | You, always, by default |
| Reporting cadence | Usually one summary a week or month, written for you | Whatever you check, in real time, inside the Amazon Ads console |
| Who absorbs the learning curve | The agency already knows it; that knowledge is what the fee buys | You, unpaid, but you keep the skill afterward |
The one line in that table worth sitting with is contract length. A campaign that is losing money under agency management for three months before you can cancel costs more than the retainer alone — it is retainer plus three months of ad spend at whatever ACOS the agency was running. Ask for month-to-month terms before signing anything with a longer minimum, especially on a first engagement.
Paying an agency buys back your time, but it also hands someone else the keys to a lever that directly moves your income. You lose the pattern-recognition that comes from watching your own search term reports week over week — noticing, for instance, that a keyword converts well in December and dies in February, or that a particular cover variant pulls a lower click-through rate than the others in a series. An agency managing forty accounts is not going to catch that nuance on your specific backlist the way you would.
Running it yourself costs real hours, and the honest number is higher than most "just spend 15 minutes a week" advice suggests once you have more than one or two titles live. The table below is time per batched review session, meaning you sit down once and work through every active campaign in that sitting rather than opening the Amazon Ads console separately for each book.
| Task | How Often | Time Per Batched Session (Whole Catalog) |
|---|---|---|
| Search term report review | Weekly | 15-20 minutes across all active campaigns together |
| Bid adjustments on existing keywords | Weekly | 10-15 minutes |
| Negative keyword additions | Weekly | 5-10 minutes |
| Budget check and dayparting review | Every two weeks | About 10 minutes |
| New campaign setup | Per new title published | 20-30 minutes, one time |
Those figures are catalog-wide per sitting, not per campaign, which is the detail that makes the monthly total work. A weekly batched review of the first four rows runs roughly 30-45 minutes regardless of whether you have three active campaigns or eight, since you are reading one consolidated search term report and adjusting bids in one pass rather than opening each campaign separately. Multiply that by four weeks and add an occasional new-campaign setup, and stable DIY management across a five-title catalog lands around three to five hours a month — not three to five hours per title.
That number climbs fast once your catalog grows past roughly ten to fifteen active titles, because the search term report itself gets longer and harder to scan in one sitting, even though the review is still done in a single batch. Past that point, either the agency math above starts to close, or you need your own system — saved search filters, a recurring calendar block, a simple spreadsheet tracking ACOS by title — to keep the batched-review model from breaking down.
Look under Account Settings for a Users or advertising-access section inside Amazon Ads, invite the agency's email, and assign a manager-level role rather than full administrative control. Exact menu labels shift as Amazon updates the console, so confirm the current path at advertising.amazon.com before you do this — the principle matters more than the exact click sequence: you should always remain the account owner, with the ability to remove access unilaterally.
Request two or three anonymized before-and-after ACOS screenshots from actual book campaigns, not a generic e-commerce case study. An agency that only shows aggregate revenue growth without ACOS or spend context is showing you a number that is easy to inflate by simply spending more.
Flat retainer, percentage of spend, and percentage of sales are three different incentive structures. A percentage-of-spend fee rewards the agency for spending more of your money regardless of results — ask which model you are agreeing to and why.
Agree on a specific ACOS target and a date to review it before the retainer auto-renews. Sixty days is enough time for Amazon's algorithm to move past the initial learning phase on new campaigns without letting an underperforming engagement run for a full quarter unchecked.
Since you retained ownership in step one, the campaign structure, search term data, and bid history stay in your account when the engagement ends. Confirm that in writing too — it is the difference between switching agencies cleanly and starting over from zero.
Below roughly five active titles, the fixed retainer rarely earns itself back inside the ceiling calculated above. The 30-45 minutes a week is a better trade than a monthly fee that needs hundreds of profitable sales just to break even.
Past roughly twenty to thirty active titles with real budget behind them, the time saved on daily bid management and search term triage across that many accounts starts to outweigh the retainer, especially if you are also writing, formatting, or running a second income stream.
If you are still doing active keyword research to find which niches or categories work for your backlist, running your own campaigns teaches you what converts faster than a monthly summary report does. That knowledge compounds across every future title.
Launching five or more titles in a tight window, where campaign setup alone eats several hours you do not have that month, is one of the few scenarios where even a smaller catalog benefits from short-term agency help — ideally on a month-to-month basis rather than a long contract.
A middle path exists that most pitches skip entirely: hire someone for a one-time campaign audit and setup, then run it yourself from there using the checklist above. That converts an ongoing retainer into a single fee, and it is worth asking any agency whether they offer it before assuming the only option is a monthly contract.
Flat retainers commonly run $400 to $2,000 a month, or 10-20% of ad spend, on top of the ad spend itself — that spend is separate money paid to Amazon, not included in the fee. Some agencies also charge a percentage of ad-attributed sales instead. Always ask which of the three models applies before signing, since each rewards different agency behavior.
Usually not on the numbers alone. A single title rarely generates enough profitable sales volume to cover even a modest retainer, based on the breakeven math above. New authors are typically better served spending that same money on more ad budget under their own management, or on improving the listing itself — cover, categories, and sample pages — before paying someone else to drive traffic to it.
Track ACOS against the breakeven ceiling for each title, calculated the way the cost comparison above walks through, not against the agency's own summary language like "strong performance." Ask for raw search term reports, not just a dashboard screenshot, and confirm bid and budget changes are actually happening on the schedule you agreed to.
Yes, if you kept account ownership and only granted the agency a manager-level user role, as the vetting checklist above recommends. Removing their access and continuing to manage the existing campaign structure yourself is straightforward. It becomes much harder if you let the agency create and own the account itself.
No legitimate one does, and a guarantee of sales or ranking is a reason to walk away from the pitch. Amazon Ads is an auction system with no promised outcome; an honest agency will quote a target ACOS range and a timeline to reach it, not a sales number.
Many authors start meaningful testing around $5-$10 a day per title, adjusted after the campaign has enough clicks — generally a few hundred — to read the search term report reliably. Below that spend level, results are mostly noise, whether you are managing the account or an agency is.
Octozia's KDP Creator builds the numbered puzzle-book page images and manifest you assemble into your interior — 3-day trial, card required, so you can see the output before committing a subscription.