The Notebookery
What to publish, how it looks
Covers and interiors (Canva, self-taught)
Keywords, listings, ads, pricing
Reviews, customers, print, the legal side
751 reviews, 4.8 stars, still number one in 2026
Profitable in 8 months, from a few hundred dollars
I started publishing books on Amazon because it was the one business that didn't need money up front.
Six flops and one sideways notebook.
Amazon prints a book when somebody orders it, so there's no inventory and nothing to buy up front. I taught myself Canva and put out six books: workbooks, journals, coloring books, a US citizenship study guide (I recorded the audiobook for that one myself; I still have the microphone). None of them sold.
The seventh was a kindergarten writing pad, and I made it landscape, because I'm left-handed and I've spent my life using things designed for other people, and I figured a wide page would be easier for a five-year-old. That was the whole theory. No testing.
It turned out to matter for a reason I hadn't thought of. On Amazon every cover is tall. Mine was wide, so people clicked on it before they read a word. Within days it was selling five to ten copies a day. It was the first landscape kindergarten writing book on the site. Nobody had thought to do it.
One star, eight reviews, and $900.
Kindergarten books sell in January and August, around the school semesters. In between, nothing. I was in Batumi, Georgia, living on a few hundred dollars a month, and the writing pad was the only thing making money. It had eight reviews. Then someone left a one-star review about the line size. She was right, I hadn't explained the specs well, but with eight reviews one star pulls your rating down hard, and Amazon's algorithm doesn't care why. It buried the book. Ten a day became one or two.
I was scared. I'm not going to dress that up. But I'd been watching the ad numbers for weeks and I knew which campaigns worked. I just couldn't afford to run them. When the COVID stimulus check arrived I put $900 of it into those campaigns, timed for the January window, to get sales, get reviews, bury the one star and climb the keywords all at once. Then I raised the price.
1,299 copies that month.
It was the first time since leaving the US that I wasn't living off savings.
Mead copied it. Mine's still number one.
After that month the book had momentum: sales lift rankings, rankings bring traffic, traffic brings reviews. Mead made a landscape version. So did School Smart. Real companies with design departments looked at my Canva file and did the obvious thing, and six years later mine still leads the category.
It's sold 27,914 copies, 18,313 of them with no ad behind them. 751 reviews at 4.8 stars. At the peak it was 1,900 copies a month. Today it's a couple of hundred a month with nobody touching it, through several price increases and a few Amazon takedowns, which happen to everybody and are terrifying every time.
The rest of the catalog.
I ran the same play on more books: find the gap in the keywords, read the reviews for how parents actually talk, make a cover that works as a thumbnail, write the listing in their words, manage the ads one book at a time. I learned CMYK because screen colors don't print the same and I couldn't afford to find that out twice, and enough copyright law to stay out of trouble because I couldn't afford a lawyer. Bucharest, Tbilisi, Zagreb, Cairo. The laptop was the office. Altogether, about 37,000 books.
The margins, and doing it alone.
After Amazon's cut, printing and ads, most of the titles barely broke even. I didn't understand that until the business was already built around it. If I did it again I'd check the margin before falling in love with the product, and I'd build an audience somewhere Amazon couldn't take away.
And I'd get a team sooner. By the third year I deeply, deeply wanted people to build something with. I flew home the day Russia invaded Ukraine (I was living in Georgia at the time) and went looking for something physical to work on with other people. That turned out to be an internet company in the mountains.
That notebook is where I learned what I actually do: get someone to stop and look.
