Most authors I work with open KDP the same way: once a day, straight to the graph, a quick look at whether the line went up, and then the tab closes with a feeling and no information. The feeling is usually bad, because a daily line for one book is mostly noise, and noise always seems to trend down when you’re the one staring at it.
So this is the read I do when a client hands me their KDP access. Five numbers, pulled in a set order, and a short list of the ones I skip entirely. You don’t need a spreadsheet for it, though I have one, obviously.
Set the date range before you read anything
The dashboard defaults to a short window, and a short window is where the panic lives. Change it to the last 90 days. That’s long enough to see a shape and short enough that a launch spike, if you had one, stops distorting the average. Then move off the graph and into the reports underneath it: Orders, KENP Read, and Payments. The graph is a summary of those three. The reports are the actual record.
One more setting before you start: filter to one title at a time on the first pass. A catalogue view hides a book that has stopped selling behind a book that hasn’t, and the whole point of this exercise is to find out which is which.
Number one: net units ordered, per title, per month
Units ordered is the count everyone looks at, and the version that matters is the net figure, after refunds and cancellations come off. A book that shows 40 orders and 9 refunds in the same month is telling you something the gross number hides, and in my experience it’s usually a mismatch between what the cover and blurb promised and what chapter one delivered.
Read it by month, never by day. Fiction sells in lumps: payday, a newsletter send, a group post that took off, a weekend. Daily units for a mid-list title swing between zero and six for no reason you’ll ever identify, and if you let those swings set your mood, you will make a pricing decision on a Tuesday that you regret by Friday. Monthly totals for the last three months tell you whether the book is growing or fading, and that’s the only question this number can answer.
If you ran a free promotion, those downloads sit in a separate free units column. Keep them separate in your head too. A free run is a visibility tactic, and its success shows up in the paid units and page reads of the weeks that follow, never in the free count itself.
Number two: KENP read, as a trend and as a page count
If your book is in Kindle Unlimited, KENP Read is the number that pays you, and it deserves more attention than units. It’s hard to read for a reason. The per-page rate is only set after the month closes, so the royalty figure beside it is an estimate until Payments confirms it. And the raw total means nothing on its own, because 60,000 pages read is brilliant for a novella and worrying for a 600-page epic.
The fix is to divide it by your book’s KENPC, the page count Amazon assigns to your title (you’ll find it on the book’s promotion page in your Bookshelf). A 400 KENPC book with 60,000 pages read in a month has roughly 150 full reads in it. Now you have a number you can compare across books, across months, and against your paid units. When the full-read count climbs while units stay flat, KU readers have found you before buyers have, which changes where your next dollar of marketing should go. One client’s page reads grew fifteen times over in seven months and her unit sales barely moved for the first three, which would have looked like failure on the graph and was the opposite.
Number three: read-through, which you have to calculate yourself
KDP will not show you this one. You build it from the two numbers above: take book two’s full reads (units plus KENP full-read equivalents) and divide by book one’s, over the same 90-day window. 200 reads of book one and 90 of book two is a 45% read-through, and that single figure tells you more about your series than any marketing metric on the page.
Where it sits decides what you fix next. A low read-through with strong book-one numbers points at the ending of book one, the back matter, or a book-two cover that doesn’t look like it belongs to the same series. A read-through that holds above 60% with weak book-one numbers tells you the story is working and the front door is the problem, so the money goes to the book-one listing and ads. I wrote about what read-through rate tells you and what makes it drop if you want the long version.
Number four: royalties paid, by marketplace
The Payments report shows what Amazon sent you, by marketplace, after the exchange rate and after the 35% or 70% band did its work. Read it once a month and read it by country.
Price and geography both show up here and hide everywhere else. Price first. A book at $0.99 can top the units chart and still earn a third of what the same units would at $2.99, and only this report makes that visible. Then geography. I’ve watched an author dismiss the UK and Australia as rounding errors for a year, and when I pulled the Payments report those two markets were paying a quarter of her income. She’d never run an ad outside the US. If you’ve been working out whether KU or wide is the right call, this report is where the answer starts, because it shows where your readers already are.
Number five: total monthly royalty across the whole catalogue, six months back
Now widen the filter to all titles and look at one line: total royalty per month for the last six months. This is the career number. Individual books rise and fall, a new release pulls the backlist up for a while and then settles, and none of that matters if the total is moving in the right direction across two quarters. A flat total with a growing catalogue is the signal I take most seriously, because it means each new book is only replacing the income the last one lost, and that pattern is exhausting to keep feeding.
The numbers to ignore
Sales rank. It refreshes hourly, it’s relative to every other book on the store that hour, and it will ruin your afternoon. Look at it on launch day if you must, then close the tab.
The daily graph, for all the reasons above. The royalty estimate on KENP before the month closes. Lifetime totals, which flatter everyone and inform no one. And the screenshot someone posted in a Facebook group, because you don’t know their genre, their price, their ad spend, or whether they cropped the refund column.
The whole read takes about twenty minutes once a month. Pull the five numbers, write them in a note beside last month’s, and only then decide whether anything about the price, the cover, or the ads needs to change. Deciding on a Tuesday afternoon because the line dipped is how most authors end up changing a cover that was working.
If the numbers are telling you that readers arrive at the listing and leave without buying, that’s the part I fix. My KDP setup and blurb writing service rebuilds the blurb, categories and keywords so the page does the selling the dashboard says it isn’t doing. Send me the book and tell me what feels messy.
Frequently asked questions
Which KDP report shows what I was actually paid?
The Payments report. It lists the royalty Amazon sent you by marketplace and by month, after the exchange rate and after the royalty band is applied. The figures on the dashboard graph and the KENP Read report are estimates until this report confirms them.
How do I turn KENP read into a number I can compare?
Divide the month’s KENP read by the book’s KENPC, the page count Amazon assigned to the title. That gives you an approximate count of full reads, which you can then compare against paid units and across books of different lengths.
How often should I check my KDP dashboard?
Once a month, over a 90-day window, is enough to make good decisions. Daily checks show noise for most titles and lead to pricing or cover changes that a monthly view would have argued against.