Learn from My Mistakes: A Guide to Self-Publishing

In May 2026, I published my first novel, Ahelno, on a budget of $250. Come on the journey with me to learn what I did to publish the book, including what worked and what didn’t. 

If you’re here because you’re hoping to self-publish, welcome! I hope this post is helpful. If you’ve already successfully self-published, pay it forward, and drop a comment, and let the community know what worked for you. (Any comments posted attempting to sell a product or service to authors will be deleted)

1. Wrote the book. That’s a whole other blog post. I drafted the book in Google Docs (free) and used Reedsy (also free) to typeset it and export it for publication.

2. I chose a pen name. I decided I didn’t want to publish under my government name, mostly because the book I wrote has some pretty spicy scenes, and I don’t want my parents to read it. Or my friends. Or people I work with. Or basically anyone who knows me in person. IYKYK

3. At first, I decided to only publish through Amazon and Kindle Unlimited (KU). As a new author, I felt it would be the easiest way for me to reach a wide group of readers. I had done some Reddit research, and it sounded like getting bookstores to actually carry a printed book would be unfeasible for a first-time, self-published author, so an e-book seemed the best route. Plus, Amazon dominates about 80% of the e-book market, so it made the most sense to invest my time there. I did end up uploading a paperback for fun (mostly because I wanted a copy).

Of course, I later changed my mind and decided I wanted to take the paperback wide so I can see if I can sell it in a few local bookstores, but that’s a different journey. Learn from my mistake: Buy the ISBNs so you can sell your book anywhere. If you are only publishing on Amazon, you can use an Amazon identifier (ASIN) instead of an ISBN, which is an international identifier for your book. But that ASIN can only be used on Amazon. If you decide you want to sell the book elsewhere, you need your own ISBNs. You have to purchase separate ISBNs for each format (ebook, paperback, hardcover). In the US, ISBNs can be purchased only through Bowker. Do not fall for scams that claim they can sell them cheaper.

4. I set up a separate gmail account (free) for my pen name. At some point, I will probably spring for the "author@mkstephenson.com" email through this website, but that seems like an unnecessary expense at this point.

5. I went to Kindle Direct Publishing (KDP) and signed up for an account (free). KDP is tied to your existing Amazon account.  If you don’t have one, you can create a new Amazon account that will also be used for your KDP account. Honestly, I wish Amazon would change this. I would prefer to keep my Amazon consumer account separate from my author account, but for some reason, Amazon doesn’t want you to do that. I don’t know if you could pretend to not have an Amazon account and create a new one altogether under your pen name, but if I could go back in time and try it out, I would.

   
   

   Learn from my mistake: Do not use Kindle Create. I originally tried to use the Kindle Create app to publish, but you can't export files to PDF or as a docx file if you made any changes. If you're doing an ARC campaign (see item #12) or want to publish a paperback or hardcover, you'll need a typeset PDF file, and Kindle Create doesn't do that.

   
   

6. Reedsy (free) is a pretty basic editor, but will export your file as an ePub for ebooks and will typeset it into a print-ready PDF. It will also allow you to back up your book as a .docx file. Overall, I found it very easy to use. Trying to typeset your book (which is a specific type of formatting) out of MS Word/Google Docs, etc can be done, but is a hassle. Reedsy does everything for you automatically. Kindle Create is fine if you're publishing an ebook on Amazon only. However, that was my original plan, and I changed my mind and decided since I was already publishing an ebook, I may as well go ahead and publish a paperback as well. It's not that much more effort. Reedsy is free and allows you to publish in multiple formats. Vellum, Atticus, and Scrivener offer more features, for a price. And a lot of authors use Draft2Digital if they want to publish an ebook outside of the Amazon ecosystem.

   
   

   To clarify, Kindle Create is a specific app from Amazon to format ebooks. You don't need Kindle Create to publish with KDP (Kindle Direct Publishing). Whatever program you use, you'll upload them to KDP (see item #9). You upload an ePub or KPF (Kindle Create format) book for ebook or a typeset PDF for paperback/hardcover.

   
   

7. I found a graphic designer on Fiverr to design my cover art for $75. Now, this is where things get controversial. The designer I chose used generative A.I. to start the cover art and then edited it. I was fine with that, because no one is hand-drawing a book cover for $75. If you want a professionally-designed cover that doesn’t use A.I., you can expect to pay between $300-$1,000. That is absolutely, in my opinion, a fair price to pay for that amount of work and talent. But this was my first book, and I didn’t know if I’d sell even one copy. 

   
   

   There are a lot of readers who do not want to read books that have used A.I. in any way (including the cover design), and I respect that. I do think transparency is important. KDP has a checkbox asking if you’ve used A.I. in your cover art, so of course I checked the box to say that I did. I don’t know if readers are able to filter that out, though. 

   
   

   Since publishing, I've learned that there are many sites to get lower-cost covers. GetCovers is the one I see most recommended. The covers are generally $10-$35. I will be honest. I've seen some great covers that authors have posted using GetCovers. I've also seen some really horrible ones. I looked into it, but you have to sign up and pay before you can look at the artists' portfolios. I wasn't comfortable with that, so I went a different route. But if you're on a budget, it could be a good option. It's not hand-drawn art, but the GetCovers artists commit to using stock photos only and not using any generative A.I.

   
   

   For those who are wondering, I did not use A.I. to write any part of my book. All those words (the good, the bad, and the ugly) are my own. I am happy to report that now that I’ve had a few sales, I have engaged a different artist to hand-draw the cover for the new paperback version that will be coming out later this year. 

   
   

   No  A.I. was used to write this or any of my other blog posts. #transparency

   
   

8. I set up a website on Wix. I selected the two-year Core plan--including my own domain name of mkstephenson.com--for a total of $350. That’s $175 per year, which I think is a reasonable price. Certainly, there are cheaper options out there, but it had all the features I was looking for, and it’s easy to use. This is where the bulk of my money went. You can obviously go the free route or even just set up something like a Facebook page, but I think having your own domain does give the impression that you’re a serious writer. However, if you want to see if your book sells first before investing in a site, I think that also makes sense. The website allows me to easily set up pages, blog posts, reader magnets, subscriptions, emails to the mailing list, etc. 

   
   

   You may ask why I was willing to spend $350 on a site before publishing but not that amount on cover art. The website is part of a longer-term marketing plan that could be successful even if the first book fails. I have four other books I’m currently working on, so I’m hoping to sell 4-6 books in the next two years. That works out to about $50-$90 per book. And I can always pay for new cover art later if a book sells well.

   
   
   
   

9. When everything was ready, I uploaded the book and cover design to KDP and set a publishing date (May 27, 2026). There’s a lot of debate about best times of the year to publish books. I don’t have enough experience yet to give advice on that. Based on what I read, I knew to set the publishing date at least a month out (more below), but not too late in summer, when it seems readers might not be as active.

   
   

   There are a lot of options to sort through when you upload your book through KDP. I selected a price of $3.99 for the e-book, and enrolled in Kindle Select. Only books enrolled in Kindle Select are included in Kindle Unlimited. Kindle Select also offers free promotional opportunities for your e-book for 90 days. During that 90 days, you agree to not sell your e-book on any other site. If you’re having trouble figuring out the marketing options and royalties in KDP, I highly recommend this post by boundlessbookwriting.

   
   

10. In case you’re wondering, I went through five rounds of editing, revising, and proofreading AFTER I had uploaded what I thought was my final manuscript. Don’t recommend that, but hey --first time author problems. I have since learned there are actually four different levels of editing: developmental, line editing, proofreading, and copyediting. Professional editors all charge between a penny and six cents per word--about 10-20 cents for all four. So a full edit of my book would cost a minimum of $1,600. That’s just not in the budget right now.

If you’re publishing a paperback, do pay for an author’s proof of your book (mine was about $10) before publication. When I went through it in print, I probably caught about 50 more errors. So embarrassing.

11. I created an account on Goodreads and claimed the author page for my book (free). If you post your book on Amazon, it will automatically populate into Goodreads within a day or two. Search for your book in the search bar, find the book, then click on the author name. On the author page, scroll down and find the link that says “Is this you? Let us know” and then complete the form.

Fun fact, Goodreads denied my first request. They will send an automatic email that lists possible reasons for denial, but not give you the specific one. I believe it’s because I didn’t have my gmail address fully spelled out on my website. I did that so my email address wouldn’t be scraped and spammed by bots. But Goodreads looks to see if it’s there to verify your author account email matches. I added my email to the site and resubmitted and was approved.

12. I started an ARC review campaign. ARC reviews are “Advance Reader Copies”. You give people a free copy of your book before it’s published in exchange for them leaving an honest review on Amazon, Goodreads, Bookbub, etc. There is a strong correlation between Amazon reviews and sales. You can build your own ARC reader list (obviously time consuming), or you can use a service that farms your book out to ARC readers already interested in your genre.

I chose BookSirens because they were the site most frequently mentioned for romance ARC reviews, but there are plenty of others out there. You pay $10 for them to promote your book and $2 per reader who downloads your book (you can limit the number of ARC readers). They also have an annual plan of unlimited ARCs for $100. Because I have other books in the pipeline, I tried to sign up for that, but was denied. BookSirens very politely told me that because I was a new author, and they had no way of knowing if my book would be of interest to their readers, they weren’t willing to accept me in the annual plan. But, if your book does well and you want to sign up for the annual plan later, they’ll deduct the cost of your first promotion from the annual fee. 

The payment is an admin fee that goes to BookSirens. The ARC readers are volunteers and do not get paid for their reviews.

Readers can leave reviews on some sites, like Goodreads, prior to the book’s publication date. Amazon does not allow this, so a reminder will go to the readers who downloaded the book when the book is published asking them to post their review. You have to upload your book to BookSirens in epub and .pdf format.

ARC results: 1398 impressions, 430 clicks, 15 readers, 11 reviews

Total cost: $40

Two things I learned:

• The cover design performance was phenomenal. This is the first thing people see, so your cover art needs to draw people in to click on it. 1398 people were shown my book’s cover, either through a BookSirens email or on their website. Thirty-one percent (31%) clicked on it  That is fantastic performance. However, only 15 of those clicks converted to reader downloads (3%). That’s about half of what it should be, which means the blurb didn’t convince people to download. 

Honestly, I expected that. The book deals with sexual assault and suicidal ideation, and that’s not the lighthearted alien romance a lot of people are looking for. I tried tweaking my blurb (and asking for help on a subreddit), but most people agreed the topic just isn’t to everyone’s taste, and that’s ok.

• I didn’t realize that I needed to start the ARC campaign 60 days before the book launch date. The campaign lasts 30 days for Booksirens to advertise the book to readers for download, and then they need ANOTHER 30 days to read the book and post their reviews. I set my ARC up three weeks before the book’s publication date. Don’t do that. Eventually, eleven of my ARC readers published their reviews, but only seven did so within the first week of launch. Still a good outcome. I was very pleased. Thank you, ARC readers!

13. I set up an Instagram account under my pen name (free). This was a tough decision. #booktok is a powerful thing, especially for the romance genre. However, I’m using a pen name and not interested in releasing videos of myself.  So I started with Instagram. After about a week of research, I started a TikTok as well. I do faceless content, mostly book reviews, funny memes about being a self-published author, and, of course, book promotion.

Promoting yourself as a new author is tricky. No one cares about your book, because they’ve never heard of you. So you have to build a presence. I started by following 25 authors I liked. I posted a mix of content: book reviews, ideas for book lovers, and day-in-the-life of an author. 

I did not post about my own book that was coming soon. That gave me three weeks to publish other book-related content and build an audience that might actually be interested when they saw me post about my own book. I also joined Reddit, which has a really helpful /selfpublish subreddit. Highly recommend.

About a week before the publication date, I started promoting my book. Not gonna lie, it’s been slow going. I started from zero and four weeks later I have 22 IG followers and 34 Tiktok followers. I’ve had several posts that have gotten 300-400 views, but nothing that’s gone viral, and most posts average around 30 views. I have discovered I suck at social marketing. Maybe one day I’ll be able to afford a social media intern. Ha!

14. Marketing: In addition to social media, I used this handy list on Kindlepreneur of where you can promote your book. I basically uploaded my book info anywhere that offered a free promotion. The only paid promotion I did was for a week on romance.io, which is the Bible of romance book websites. It’s $40 for an ad. It just started running three days ago, so I don’t have the full data on that yet.

    
    

    After doing a lot of research about marketing, I learned you’re supposed to write a reader magnet (often a bonus scene, or a new story) and include a link to your newsletter for people to sign up and read it. Your newsletter is supposed to be your best long-term marketing strategy. I wrote an additional spicy scene from my book. I’ve had three signups so far. LOL.

In addition to the cost invested, I’m sure you’re curious about time. I’ve seen several posts that will tell you that you should treat writing and publishing like a job. That’s probably true, but I suspect most indie authors have other jobs during the day that are keeping them occupied. So don’t feel bad if you can’t make this your full-time gig yet. My goal with this post is to help save other self-publishing authors time, and to help serve as a case study for what may or may not have worked for someone else trying to make it on a shoestring budget.

In total, I spent a total of about 40 hours over the course of a week on building everything associated with the self-publishing of my first book. Here’s an approximate breakdown:

• 12 hours - Formatting  and editing the book after uploading it into Kindle Creator. Seeing it in a different format will highlight errors you hadn’t noticed. Same with seeing it in actual print.

• 10 hours - Research on marketing and self-publishing

• 10 hours - Content creation - This included my author bio, a blurb, synopsis, a week’s worth of social media posts, and going back and forth with the cover artist on revisions

• 4 hours - Website - I have an advantage in that I already have some experience building websites.  If you don’t, expect this to take considerably longer. The initial build probably took 2 hours, but then I put in another two hours over the week as I changed/added/updated things.

• 4 hours - Technical stuff - setting up all the different accounts. In a lot of cases, you submit your details, upload content, check to see if the account is approved, then go in and edit, etc. So some of these took longer than I expected

Since then, I’d say I’ve spent about 20 hours per week on marketing. Some days it’s more (I like to batch create posts and I just save them on my phone), and other days I don’t spend any time. I've set up a Kindle Countdown deal to happen next week, and I've done another round of setting up promos through the Kindlepreneur list to highlight the countdown. Other authors have recommended the countdown deal (where you discount the book) as opposed to offering the book for free. Those who purchase the book—even at 99 cents—are more likely to rate the book or review it.

I also used a free trial of BookBeam and spent about two hours of research to refine my book's key words. Publisher Rocket is the gold standard, but I'm not ready to buy a subscription.

Now that things are set up to run on their own, I'll probably spend about 5 hours per week on social media & marketing. I need to turn my attention to finishing A.I. Human, which I expect to release in September.

It’s been a whirlwind. I started writing the book in January, and decided on a whim (in April) to see if I could publish it. It’s been one week since my book was published on Amazon and below are my results. I’ll do a followup in a month to update. 

Total cost: $215*

Number of e-books sold: 12

Number of pages downloaded on KU: 5,860 (~14 readers)

Estimated royalties from Amazon: $56.47

Profit:  -$158.53

*I've amortized the website cost over six books, but if you want to include the total spent upfront, then I've spent $515.

[Edited to add: See the results from my first month in this post!]

Questions? Comment below