Everything That Went Wrong With My Launch And More
Disaster is a strong word. It's the right one, but strong. My recent launch was two things: it was a great example of why the proven self-publishing strategies are what they are, and it was a frustrating case of yet another Amazon glitch demonstrating how little control we sometimes have over our publishing careers. So before I get to the meat and potatoes of what happened, I need to give a sort of crash course on the generally accepted understanding of how to promote a book on Amazon. To understand why this glitch was so devastating, you have to understand the plan. After all, if you watch an NFL quarterback throw an interception, you don't get why that's bad until you understand the rules of the game, right? This is going to be a sort of generalized, watered down description for brevity's sake. But I'll start with an analogy. Think of self-publishing like pushing an object up a big hill full of trash and debris. The trash and debris is always changing and shifting, so you never know exactly what shape of object is the best to try to push up that hill. You can study trends and patterns. Maybe this time last year, really small shapes seemed to do better because all the debris was big and easy to avoid. Or maybe last year it was best to have really big shapes because there was a bunch of trash that small stuff got tangled up in. The launch of your "object" or book is the process of trying to push it as high up the hill as you can. You won't know if you chose the right shape until the moment you commit and see it in action. But you know you have to push hard, because even the perfect shape won't make any progress up the hill without some initial energy invested. If you fail, your object will run out of steam before it reaches the top of the hill and roll back down towards you. If you succeed, it'll reach the top and roll down the other side (your goal). The more effective your shape, the easier the path up the hill and the easier time it has coasting for a long time on the good side of the hill. The worse the shape, the sooner it'll stop rolling on the other side, even if you manage to get it over the hill. Super long analogy there. Or maybe it's a metaphor at this point, but who cares. The point is the "object" is your book. The debris and trash are the market and trends at the time you publish. The shape of your object is your title, blurb, cover, and reviews. Your "package" if you will. The act of pushing the object up the hill is your initial salvo of promotions like newsletters, daily ads, social media interaction, and so on. But that's essentially how this all works. You launch a book and you have to try very hard in the first few days to push it with money. The better job you did on the package, the more efficient each dollar you spend is. A wonderfully packaged book might sell 5 copies for every dollar you spend. A badly packaged book might only sell 1 copy for every 30 dollars you spend. Most often, you're going to be spending more than you earn for a few days to give a book a real shot at getting over the hill. Unfortunately, sometimes the book will just not convince people to purchase once they land on the page. So no matter how amazing you do at getting eyes on your product page, the product still has to sell once they get there. Now the other factor here that was harder to wrap into the analogy is that Amazon's bookstore is operated by a complex algorithm. Nobody knows exactly how it works and the best we can do is look at the evidence available on our end and guess. To make things more complicated, the algorithm is constantly evolving and changing. We get no warning when this happens, either.
At the moment I'm writing this post, my best *guess* for how the algorithm works is that it rewards efficiency. I won't bog this post down with data and details to explain why I believe this, but I can make a quick argument for why it would make sense for the algorithm to operate this way and what that means. Basically, my guess is Amazon records data when a customer visits your product page and tries to evaluate how effective your product is at generating profit. So let me give a few examples where 4 books all get the same clicks but different sales.
Every book gets 1000 clicks (people clicking an ad or searching the book but somehow 1000 people landing on the page) Book A is $2.99 and sells 23 copies. ($68.77 - $2.00 royalty to the author per sale = $22) Book B is $2.99 and sells 219 copies. ($654.81 - $200 royalty per sale = $216) Book C is $0.99 and sells 150 copies. ($150 - $.35 royalty to author per sale =$97.5) Book D is $0.99 and sells 40 copies. ($40 - $.35 royalty to author per sale = $24) So if you were the Amazon algorithm and your job was (I'm guessing) to maximize profit for Amazon, you can kind of get a more clear picture here of which books are going to be pushed by the algorithm and which books are going to get buried. The superstar example is Book B. 219 sales after 100 clicks is over a 20% conversion rate, which would be super effective and probably better than 99% of the competition. A book like this will feel like it's just magically rising in rank and growing almost exponentially as long as it keeps moving copies this way. I think that's because the algorithm starts guiding clicks to it since it's earning the algo so much money per click. On the other hand, a book like Book A is just eating up clicks and hardly moving copies with them. I would guess a book like that with such a poor conversion rate would get almost no help at all from the algorithm because it could find so many more profitable books to push to customers. The only other thing I'll point out is I do think the price of your book matters because you can see how the profit plays out. Even if Book C had sold 300-400 copies, it wouldn't be more profit than book B because of the lower selling price. Of course there are likely several other factors that get inserted into the algorithm's equation. It wouldn't shock me if something like the liklihood of readers to go on and purchase more books after buying a book is included. Or maybe how likely shoppers are to buy another item before finishing their Amazon session. If, for some reason, a lot of people tend to buy the book and then go on to buy electronics or lawn mowers or something, that would also arguably make the book more valuable to push. But basically, it seems like two things are important above all else. 1) Time. Your book needs to prove its worth relatively soon after launching unless you get some extraordinary influx of numbers later down the line. Generally, you are feeding data to the algorithm to help convince it the book is worth its time and advertising effort. If you fail to prove your book's value in those first few days of generating sales, you are just accumulating more and more negative data that is going to weigh down anything positive you do going forward. So having 10 days of 2 sales a day followed by 1 day of 500 sales is going to be a lot less impressive to the algo than having 500 sales on day 1. 2) Efficiency. The more efficient your book is at converting sales once people land on the page, the easier everything will be. Any advertisements you run will lead to more sales because a greater percentage of the people you drive to the page will buy. Any promos will mean more sales. Any help from the algo will mean more sales, which will reinforce the algo's decision to push the book. So the ideal book launch starts off strong and stays strong with a sort of "artificial" push from ads, promotions, newsletters, social media, and so on. It often can feel like it takes about 5-7 days for the algorithm to gather enough data on the book to start either pushing it or burying it. That means the artificial push has to sort of survive that period. You generally can't just have one great day and convince the system your book is a gem. You have to give it consistent data over a period of a few days for it to get enough of a feel for your book. That's why most authors spread out a stack of promos over the first couple days with their launch and then rely on paid daily ads to float it after that. Okay, that was more like a lesson than a crash course, but I can finally get to what happened with my launch now. When my book went live last week, there was a bug going around. New books weren't getting a rank and old books' ranks seemed to be stuck. So my book launched with no rank on day 1. I set up my ads and made all the preparations for my promo stack over the next 3-4 days. Day 2 I sent my NL and had some social media posts. The book was selling well and I was hopeful the rank would come in at some point around the 100s or maybe even the top 100 if I was lucky. Day 3 I had the next best promos run and had fewer sales, which is normal for me on the day following my NL, but I was still feeling good about the pattern. I knew based on the number of clicks I'd brought to my book that the sales seemed really healthy. That gave me confidence based on past books that I was creeping into the top 50 territory with sales, but I still had no rank.
I was starting to get pretty worried by this point. The rank is just one tool the algo uses to bring eyes to your book. Rank 50 means you'll be near the top new release spot in most categories and the top overall book in some other categories as well. It means when people go to the amazon store page you'll be one of the first 10 or 15 romance books showing at the top of all the lists and you get a lot of exposure just from being there. Basically, it brings in a lot of sales, which further helps convince the algorithm that your book is doing well and deserves a push. But the next day came and I was down to the fumes on my newsletter and promo sends. The sales were significantly lower at this point than I'm used to, but I figured that was because I normally have a rank by that point and usually a pretty solid one that brings a lot of visibility. I figured maybe the rank was delayed and I hoped when it showed up I'd still see the story of my rank updates, just a few days behind. It wouldn't be ideal, but I was relatively confident the book could still survive it because that would mean by day 6 or 7 it would probably hit top 50, based on the sales pattern I'd seen. It was 4 full days after launch when my rank finally showed up. My sales were near rock bottom at this point. I had virtually no organic visibility going for me and the only views on my book were ones I could directly drive to it through paid ads. And when my rank finally showed up, it was 474,000. For reference, that's what you would expect if you were selling maybe 1 copy of your book every 2-3 days, if that. I figured it was okay because I thought maybe that was just showing the rank my book was when it first launched before I had ads set up and had very few sales that first day. I thought for sure by the next day I'd seen the rank jump from when I sent my NL and sold hundreds of copies. But the next day came and my rank slow crept from 474,000 to about 150,000. Again, that's probably still not even selling a full copy a day in that rank range, and I was still in the 100,000s by the end of the night. That made no sense, because if it was simply delayed, then I should've seen a huge jump because that would represent the day I sent my NL and had a lot of sales. The next morning came and the rank was something like 1500. And keep in mind at this point I was still getting terrible sales every day. I'd just spent my whole promo stack while my rank was invisible and then the book was essentially flying blind with no traffic except traffic I paid to send through FB ads, which is very expensive and not something you can rely on to make a profit by any stretch. I had one last promo run and sold something like 70 copies at $2.99 that day, but my rank sat in the 900s all day. Again, for reference, I've had books in the top 50 on the store selling 40-50 copies a day and holding rank partly because they had the proven track record from that initial boost of sales to get them there. But in this case, the rank just wasn't responding to my sales. It also clearly hadn't counted my big boost of sales during my promo stack. There was just no way with the volume of sales I had. Rank 900 at $2.99 would be more like 10 sales a day or something is my guess, but somehow my best rank was 900 when I'd sold hundreds of copies a day during my promo stack and just had a day with 70 sales. So what happened? My best guess is those days with no rank were simply lost in the system. Instead of a delay, it seemed more like they just never happened. Or possibly even worse, it seemed like they were counted as 0 sale days. So when I came out of that 4 day period my 600 or 700 sales counted as 4 days of 0 sales. So then when I was only selling 20ish copies a day, I was getting penalized because those 20 sale days were also being averaged in with 4 terrible days. By the time I had my 70 sale day it was heavily weighted by all the weak days before it. I still don't know how 70 sales didn't get me past rank 900, but all of that is the only explanation I can think of. So I considered unpublishing the book at this point. I thought maybe I could just call the launch a bust and hope in 5-6 months I could re-launch it, avoid a bug, and hope the book did better. But Amazon is pretty picky with republishing books and I didn't want to risk walking the line wrong. I decided instead to try a last-ditch plan and discount the book to $0.99 for about a week. I'd send to my list again and try to book any promo could get in hopes that I might get a second try to push my book up that hill. It *sort* of worked. Everybody I reached out to was really kind and supportive about how the launch had gone and seemed to go out of their way to grab the book to help out. It sold something like 700 copies the day I sent my NL again at $0.99 and bounced from rank 900 to 120 at its best. Of course, if that 700 sale day at $0.99 had come on day 1 of launch, it probably would've also been enough to push the book straight into the top 100, but I was dealing with 7 days of abysmal sales in the eyes of the algo because of the glitch at that point. The sales didn't stay that strong the following few days and the book has drifted to around rank 300 or so last time I checked. All in all, I'm grateful it at least didn't just drop like a boulder as it appeared to be on its way to doing. But I'm left feeling really frustrated by the experience. Self publishing is sort of like having a split personality. For the vast majority of my time, I'm author me. My work every day is to get words down on the story and keep making progress. When I'm in that mode, I often think about the launch period to motivate myself. I imagine the book doing well and getting to read feedback from people who liked it. I imagine the sort of emotional payoff of possibly seeing the book rank well, which always makes me feel like I did a good job as a writer. In this particular case, I had a ton of trouble writing this book. I started and scrapped two versions of it and eventually had to shift to another storyline for it to feel right. Even then, I wound up literally getting sick four different times while writing this book and dealing with sickness going through my family when I wasn't sick. We had a vacation for a week and had to put our dog of thirteen years down after a long battle with what felt like countless illnesses. It just felt like my personal life was in pure chaos for 3-4 months and the project of trying to finish the book in the middle of it all was a constant source of guilt and stress and disappointment for me. But I finally managed to get my shit together and not just write the book, but write it in a way I was really proud of. So then I turned things over to the business side of my brain and felt like it was finally time to cash in the emotional reward I'd worked so hard for. And just like that, it all kind of went down the drain. But that's just how this business is sometimes. Sometimes it's great and you feel like the luckiest person in the world. Other times you feel like a cog in a cold and uncaring machine. If nothing else, I always try to learn from both my successes and my failures. In this case, it was at least interesting to see just how important rank and visibility is for sales. It's usually impossible to really measure with any accuracy because rank is always there and moving. But this time, I got to see what a book that converted above average to my list and regular promo would do without any rank or organic help, and it wasn't pretty. Once I was down to just Facebook Ads and Amazon Ads, I saw I was really only selling about 20 copies a day from roughly $300 a day in spending. It has been hard to ever really know how much of my spending is directly creating sales because both ad platforms give shaky information at best. But my takeaway here is that I've been wasting my money for the most part. The majority of my sales were coming from momentum after my newsletter and promos, but I was assuming they were partly because of my ad spending. Do I wish I didn't have to spend 4 months and more of my life toiling to write a book to learn this lesson? Yes, absolutely. But I think I'll take it to heart anyway. So I hope this was interesting at the very least. And if you haven't already, feel free to become one of those clicks. Check out my book and decide for yourself if it's worth a purchase! Feed the algorithm!!! Once Upon A Grump by Penelope Bloom - Learn More xx Penelope