Amazon Ads – Nothing Yet

When last we met, I had thrown in the towel on Kindle lockscreen ads ($20 in ads led to zero sales). I set up a couple regular Amazon ads to see what would happen. Here’s what I’ve learned so far:

  • “Automatic Targeting” doesn’t work. I have a 50¢ PPC bid, which is more than Amazon recommended, and no impressions at all on that campaign.
  • Book targeting doesn’t work. I’m getting a few impressions, but it’s negligible. It seems that you would have to manually adjust the bid for every book, because the recommended bid levels have a huge range. So at a fixed 50¢ bid, I would win on books nobody looks at, and I lose on books that people actually visit. The bid required for bestsellers is bonkers ($3.67, for example).
  • Category targeting does get impressions, so that’s something.
  • So far I have 1,665 impressions, and have spent 89¢. That’s 3 clicks, or a CTR of 0.18%. So basically, no clicks and no sales.

I’m going to suspend those two campaigns and set up two new ones based on this information: manual category targeting only. Stay tuned!


Amazon Lockscreen Ad Verdict


My new novel!

Last time we talked, I had launched some Amazon “Lockscreen Ads” to see if they might be an effective strategy for my latest novel. This one just happens to be a trendy topic with broad appeal, so I figured it’s a good candidate for the Amazon advertising machine. Alas, these ads were a complete failure.

Over the course of a month, the three ads (you can see them in this post) generated 44 clicks at 50¢ apiece, resulting in zero sales. Those 44 clicks were tied to about 16,000 impressions, which means 0.27% click-through-rate (about a quarter of the 1% CTR I’ve experienced with Twitter ads). From this we can theorize:

  • Very few people click on Kindle Lockscreen Ads
  • A lot of those clicks are probably accidental

A conversion rate of zero books from 44 clicks just doesn’t makes sense if these were intentional clicks. Other bloggers have theorized that a lot of the clicks on lockscreen ads don’t even lead to a product page because the device is offline. I don’t know if that’s true, or if those clicks get charged, but it would be consistent with my results.

So what’s next? Straight-up Amazon Ads. The creative for Amazon Ads is the same as for lockscreen ads: just a tweet’s worth of words. I’m going to use the same words I used in lockscreen ads #2 & #3 (those got more clicks than #1, but I suspect that has more to do with targeting than the words in the ad).

For one of them, I’m letting Amazon target automatically. For the other, I am targeting a couple relevant book genres and a whole bunch of individual books that talk about cryptocurrency. I’m bidding 50¢ and letting Amazon lower that bid if it wants (but I’m not letting them raise it). I’ll report back when I get some news.

Trying Amazon Lockscreen Ads

The last time I tried Amazon Ads it was a fiasco. But most of that was related to the book I was selling being filthy, and although Amazon loves to sell filthy things, they don’t so much want you to advertise filthy things. Seeing as my latest couple of books don’t have any sex in them at all, I figured it was time to give Amazon Ads another shot.

When I went to set up an ad, I was given a peculiar choice. In addition to the normal ads on, I could set up a “lockscreen ad” which would appear on people’s Kindle devices. After some blog searching, I learned that everyone advises against making these kinds of ads because they don’t think they work. And while that might be true, an under-utilized ad channel is prone to be a much more affordable ad channel. So I figured why not try it myself?

The setup was pretty simple. You target groups of readers, and come up with a very short tag line. Your ad is your cover with that tag line under it.

I set up three ads:

Ad 1: After being catfished by a reclusive millionaire, will Eileen be a pawn in his high-stakes game, or will she become his queen?

Ad 1 Targeting: Humor & Entertainment: Humor; Mystery, Thriller & Suspense: Mystery; Romance: Contemporary, Romantic Comedy, Romantic Suspense.

The caption here is designed to pique the interest of romance readers.

Ad 2: A consortium of tech and finance companies has created a stable cryptocurrency. What could possibly go wrong?

Ad 2 Targeting: Business & Money: Accounting, Business Life, Economics, Entrepreneurship & Small Business, Finance, Industries, International, Investing, Job Hunting & Careers, Management & Leadership, Marketing & Sales, Personal Finance, Real Estate, Skills, Taxation, Women & Business; Mystery, Thriller & Suspense: Heist, Legal; Other: Computers & Technology.

The second ad is targeting business and technology readers. Those people are likely to know about Libra, which is an actual cryptocurrency proposed by a consortium of tech and finance companies (Facebook, in particular). It turns out that Libra is very similar to the fictional cyptocurrency in this novel. It shares the same key advantage (stability) which is simultaneously its key design flaw. (It just occurred to me that the cryptocurrency is following the super-villain trope that its key strength is also its key weakness.)

Ad 3: Between online dating and stopping a cryptocurrency-induced economic meltdown, Eileen’s life is about to get interesting.

Ad 3 Targeting: Business & Money: Economics, Finance, Investing, Taxation; Literature & Fiction: Action & Adventure, Contemporary Fiction, Genre Fiction, Literary Fiction; Mystery, Thriller & Suspense: Conspiracies, Legal; Romance: Romantic Suspense.

This ad is a shotgun approach, trying to hit a wide variety of kinds of readers with a fairly generic message.

With those ads defined, the next thing I had to do was pick a CPC (cost-per-click) bid. I started at 20¢. Here’s what I’ve learned so far:

  • The Amazon Ads dashboard says “no data available” when you are getting zero impressions. I wasted a week waiting for the data to become available before I increased my bid, and then started getting data almost immediately.
  • You cannot get any impressions with a CPC bid below 50¢.
  • Click-through rates are quite low. So far, I’m seeing 1 click per 750 impressions. In contrast, my Twitter ads typically have a click-through-rate of 1% (1 click per 100 impressions). Twitter ads have 7.5 times the rate I’m getting on lockscreen ads.

Three weeks in, I have two clicks (so I’ve spent $1) and no sales from this channel. That’s not great, but also not terrible. And I like the idea of getting all those impressions for free. So I’ll leave these running awhile, and I’ll report back when I have more data.



My new novel!

My new novel Singularity dropped on May 14. Work (the real job in which I actually earn money) has been crazy, so I haven’t had a chance to post here about the launch. Until now.

I’d say the launch was successful. I promoted it exclusively through my Twitter and Facebook accounts, and I’ve sold about 20 copies in the first 10 days. That’s on par with my other novels. Selling books is definitely a marathon, not a sprint.

I’ve been doing the writing thing a while now. I have five novels out and two short stories. I’ve written Erotic Romance, Women’s Fiction, and now a Financial Thriller. Although changing genre may not be a good idea from a career/brand standpoint, I’ve really enjoyed it, and I’ll probably keep doing that going forward. I like the idea of being an author that people can follow from genre to genre. I read all different kinds of fiction, myself.

I just ran the numbers for the year so far, and here are some lifetime totals:

  • Books sold: 1,075
  • Gross revenue: $2,110
  • Net income: $562

Obviously, this is no way to make a living. Those are actual sales. I have used perma-free and free promo days to put several thousand copies of my first novel and my short stories into the hands of readers. But I figure if people didn’t pay for it, I shouldn’t count it.

If you are wondering about whether this new novel will appeal to you, I’d say there’s a good chance. The beta feedback and early reviews have all been outstanding. It’s a fast read, and a lot of people have told me they couldn’t put it down. I did a thread on Twitter where I described the story, so if you want to know more, go here:

If you do decide to read it, please, please, please write a review. I need a lot more of those before it’ll make sense to do any advertising. Just a star rating and three adjectives would be tremendously helpful.

My Next Novel is Coming Soon

Hello, my name is Joshua and it’s been almost a year since my last blog post! Wow.

I’m getting ready to launch my next novel. I’m switching genres yet again, because why not? As is the case for all authors and all books, there is no perfect fit in choosing the genre, but this time I’m closest to “Financial Thriller,” which conveniently is not one of the genres Amazon lets you choose, because of course it isn’t. Sigh.

My print titles made the move from CreateSpace to KDP Print last year without a hitch, but this is the first time I’ve had to create a print edition from scratch in KDP Print. It was fairly straightforward, except that they wanted the cover in a slightly different format. They give you a PNG template but want you to upload a PDF, and getting your PNG into that PDF has to be exactly perfect or it doesn’t work right. Sigh. Anyway, that’s all sorted out, and my print proof cover looks perfect.

The site I use to make universal links (so people clicking a link end up in the right Amazon store for their geography) is still around and still free, but you have to know the secret URL to find it. If you just go to, you end up at the parent company’s pay site, which is ridiculously expensive at about $120/year. The trick is to go to That gets you to the place where you can still make short universal links for free.

The trick I discovered to post reviews during pre-order still works despite the switch from CreateSpace to KDP Print. The only difference is that you can’t unpublish immediately after publishing, because that option isn’t available. You have to wait for Review/Approval to finish, and then you can unpublish.

Also, I discovered that KDP Print requires you to use HTML to get extra line breaks into your description, while plain old KDP happily takes blank lines. Kind of a weird difference, but whatever. I’m waiting for my update to fix my description to go through so I have the line breaks I want. Seems like they’re taking the whole 48 hours they threaten.

That’s it for now. Once all the book pages are up and correct and linked and whatnot, I’ll be back to post an announcement of pre-order availability.

How to Have a Wildly Successful Promotion that Still Loses Money

I promised an update after I ran my big promo, but it turns out that it takes a lot longer than I expected to fully quantify the return on investment. Let’s start by summarizing what we are dealing with:

  • Entropy is a trilogy of romance novels. The first was published in August 2015 and the last in March 2017.
  • Entropy has 55 reviews, averaging 4.5 stars on Amazon, and has proven fairly easy to sell if I can get people to its page on Amazon.
  • None of the books had sold more than a couple a month for the past few months.
  • All the books are in Kindle Unlimited.
  • I’ve never run a free book sale on any of these titles before. I have run $0.99 sales a couple times, and those did okay.

And here is what I did:

  • May 3-7, Entropy was free
  • May 4-10, Duality and Gravity were $0.99/£0.99
  • I tweeted several times a day about the sales, starting on May 3
  • I ran $423 worth of newsletter placements on May 4 (details here)

By any reasonable standard, the promotion was wildly successful:

  • 5,846 downloads of Entropy
  • Entropy hit #26 on the overall Amazon best-seller (free) list
  • Entropy remained in the top 100 free May 4-6
  • 118 downloads of two perma-free shorts I have on Amazon
  • 7 full-price ($4.99) purchases of books from the trilogy (before/after the sale)
  • 3 full-price ($2.99) purchases of my newest novel (not part of this promo at all)
  • 225 purchases of Duality & Gravity at $0.99/£0.99
  • 12,811 pages read in KU (so far)

In hindsight, it turns out I did some very smart things completely by accident. First, I set up the newsletter promos to be stacked on the second day of the sale. I did this because I wanted time to contact Amazon if the sale didn’t start on schedule. Of course, it did start on schedule because this isn’t Amazon’s first rodeo. So during that first day, I promoted it on Twitter. I expected a dozen downloads. I got 300. What that did was push my book up into the top 500 free books. Since Amazon takes history into account when computing bestseller ranks, being at this high level the day before my big promo let my book reach its peak bestseller rank much quicker than otherwise would have happened. I never expected to break into the top 100, much less go all the way to the second page (Amazon shows 20/page on the top 100 list). While I think category bestseller lists are completely pointless, getting into the overall list certainly would seem valuable on getting a virtuous cycle going.

The idea is that if you get a high rank, then people who are browsing for something to download might notice it in the list. That leads to more downloads, which takes you higher on the list, and so on. Did that happen? There’s literally no way to know, because Amazon won’t tell authors where their buyers come from.

Lesson 1: Start the deal a day early and promote for free on Twitter the day before your big newsletter promos drop

The next genius thing I did completely by accident was run the 99 cent sales on my second two books. My thinking was that a few people who grab the free book might actually read it right away (I firmly believe most people who download free books never read them). And if they like the first book, they might want the rest of the series. But since they are smart/frugal enough to wait for books to be free, the chances of them paying $4.99 for the sequel struck me as slim. I figured I’d have a much better chance of getting those readers if the book was as cheap as possible, hence the sale price. I started it a day later than the free day, figuring it’d take people at least a day to read the first one and want the second. My logic was completely wrong.

Very few people came back to get the second two books after reading the first. What people did was buy all three books at once! They came to grab a free book, saw that they could get all three for 1.98, and about 110 of them took the bait. If I had understood this dynamic, I would have had the 99 cent sales start on the same day as the free sale, so I could have grabbed a bunch of those first-day Twitter users’ money, too.

Lesson 2: Put other books in the series on sale for 99 cents when you make the first book free

The last thing that really surprised me is the KU behavior. If a person subscribes to KU, they don’t need to wait for a book to be free. They can read those books any time for free. So why would a free book deal draw KU readers to read the free book? I have no idea. But it does.

The pages read are clearly tapering off, but they still haven’t stopped. I suppose I’ll have a few slow readers keeping that chart busy for the rest of the month. The great thing about this phenomena is that even though the book was free, I still get paid for KU reads. Last month the rate was 0.448¢ per page. Assuming it’s close to that for May, those 12.8K pages read should earn me about $57.

Lesson 3: The boost in KU reads is significant and helps offset the cost of the promo

This brings us to the bottom line. I spent $423 on the promo. I earned $182 on sold books, and $57 on KU page reads, for a total of $239. My wildly successful promo left me $184 in the hole. Sigh.

At this point, most authors would congratulate themselves because of all the exposure or branding they got for that $184. Sorry, folks. That’s nonsense. I’m an independent author with 4 novels. I’m not a brand.

I did get one brilliant review on Goodreads. But I also got a couple one-star no-review ratings there, because people are jerks, I suppose.

Once those KU reads die out, the book will be exactly as dormant as it was before I ran this promotion. And that glimmer of hope I had that this might have a knock-on effect of getting new readers for my latest novel? Seems that’s not happening either. (I sold three copies during the promo, but I think that was just a coincidence because they happened right at the beginning.)

Would I do this again? Definitely not. It is a bad idea to pay people to read your books. But I might do something different. I might do the Entropy is free / Duality & Gravity are 99 cents thing in a few months, and look for ways to promote it that are more cost effective, for example.

Chasing the “Big Bump”

For those of you who are new around here, let me catch you up. I’ve been writing novels for about three years. My first one was an erotic romance called Entropy, which has done pretty well for a first novel from an indie author who refuses to spend more on marketing than he makes on his writing. (I’ve sold about 400 copies of that one across all channels.) I followed that up with a sequel, Duality, and then finished out the trilogy with Gravity. The sales of the latter two books were not as robust, being sequels. It didn’t occur to me (particularly against the drumbeat of bloggers saying “You have to write a series!”), but sequels limit your audience to people who read the previous book. So while my sell-through rate (40%) is well above average, sequels are never going to be able to match the sales of the first book.

Having learned that lesson, my next novel was a genre-change. I wrote a standalone women’s fiction novel Apotheosis. It’s not erotic (in fact, there is no sex at all), and it sort of mashes up “Women’s Fiction” and “Hero’s Journey” plot lines to make something that should have a pretty broad appeal. That’s just been out a couple months and it is off to a decent start, selling about 50 copies so far across all channels. The reviews are outstanding. I’m having a heck of a time figuring out how to promote it, though. Twitter Ads, which have worked really well for me in the past, aren’t working at all for the new novel. Getting cheap clicks is working (particularly in the UK), but once they get to the page, they aren’t converting to readers at nearly the rate I saw with Entropy. No clue why.

So while I let that new book simmer, hoping it finds an audience on its own, I’ve temporarily turned my attention back to my original series. One thing I’ve never done is make a novel free. I have a couple short stories that I turned into e-books and made free, and although I’ve never spent a dime promoting them, they’ve managed to find about 1,800 readers on their own, just sitting there being free. The free shorts also include the first chapter of Entropy, as a teaser. As far as I can tell, that’s never led to any sales. But each short only took me a day or two to write, edit, and publish, so I don’t feel bad about not making anything back on them.

Despite that evidence that free is not a good strategy, along with my argument that authors should never pay people to read their books, I’m going to go ahead and do a free promo of Entropy. And I’m going to do it right. The book is in KDP Select (Amazon exclusive), so I can easily make it free for five days. And I’ve secured paid spots on a bunch of newsletters to promote the free day:

  • Book Sends: $180
  • Free Booksy: $100
  • Excite Spice: $40
  • Fussy Librarian: $30
  • eBook Betty: $25
  • Book Soda: $20
  • Book Raid: $20
  • Read Freely: $8

Yes, that is an eye-popping $423 I’m going to spend to promote a free book. I can guarantee I won’t get a direct return on my investment, since it’s free. There’s no royalties when it’s free. So without an ROI, why do it? Curiosity, mostly.

The conventional wisdom of doing a free promo is that you get The Big Bump. That is, after the promo is over, your sales go way up. I’ve never bought into this theory because it’s usually predicated on the idea that you get the bump because your bestseller rank goes up on Amazon. Except whenever I see authors talking about their bump, they point to their category rank. Having a high category rank does nothing. I proved this a long time ago (skip ahead to the word “delusional” in that post). If your overall book rank hit the top 10, as might happen if you get a coveted BookBub spot (which I can’t), that should cause a nice knock-on sales effect. But being in the top 10 of some sub-sub-subcategory helps nothing but your ego.

However, I recently read a blog post that had a much more plausible explanation of the bump: Also Boughts. Read that post for the details, but the gist is that if lots of people “buy” your free book, then your book is likely to show up on the pages of people who are browsing Amazon. We know the bump is real. Authors always talk about it as the reason to run free promos. And now that I have a reasonable explanation of it to hold on to, I’m finally ready to try one.

It’s scheduled for May 4, so I’ll do a follow-up after that with my results. Since I’m “stacking” all the newsletter ads together, I won’t have any way to tell which ones worked and which didn’t. But that’s a very solid list of well-respected performers (particularly the two expensive ones), so I’m pretty sure that I couldn’t do much better. (Within the constraints I have: ENT won’t run ads for Entropy because the cover lacks a bare-chested male model, and BookBub is just a big bunch of meanies.)

My predictions:

  1. I will get a lot of downloads. Let’s guess 4000
  2. I won’t make it to the top 10 overall bestseller list
  3. I will get a bump in my KU reads of all my books
  4. I will see some sales and KU reads of the sequels in a couple weeks
  5. I won’t come anywhere close to making my money back
  6. I will have a loss on my 2018 Schedule C for the first time as a result of this lunacy


Apotheosis: A New Release from Joshua Edward Smith

Maggie Jane Schuler

There are pitfalls people fall in over the course of a lifetime, and Cynthia is one of those folks who has fallen. She’s a midlife divorcee who recognizes she needs a change in order to live her life and fill the empty void. Apotheosis tells the tale of one woman’s journey in search of owning her own happiness. Her sojourn takes her through an emotional cleansing of sorts and mixes her foundational beliefs with a blend of cultural experiences, self-esteem building, and the path to loving oneself and having the love returned.

Joshua Edward Smith takes the Pacific Northwest by storm in Apotheosis. He creates a lost soul in search of the road to salvation in his heroine. Cynthia is never the meek and miserable type woman, but she is struggling to make sense of a life she feels has been unlived. Smith demonstrates a realism in this work. He…

View original post 187 more words

Get a Free Novel when you Sign Up!

I have a new novel coming out soon, so I’ve decided it’s time to do a push to get people to sign up for my mailing list. I thought it might make sense to offer a free copy of my first novel as a way to entice people to sign up (and if they love it, they can buy the sequel and the sequel to the sequel).

But there’s a catch!

All three books are in KDP Select, so that Kindle Unlimited users can read them for free. Amazon has a rule that you cannot distribute books anywhere else if you want to be in that program. And that includes giving them away like this. So what I’ll do is if you sign up now, I’ll buy the book from Amazon and give it to you as a gift. I get my royalty on that, so it costs me almost nothing. I’ll pull Entropy out of the KDP Select program when my contract expires on January 8, so after that I can just email copies.

So let’s do this! Click here to sign up.

Mailchimp will send me a notification that you signed up, and I’ll send the Amazon gift code to that email address.

Hard Tweets Explained: NäN

In the early days of computers, pretty much the only thing everyone could agree on was how to represent small positive integers. Other than that, nobody could agree on anything. They couldn’t agree on how many bits were in a byte. Or how many bytes you should work on together (called a “word”). Or how the bytes in a word should be ordered (biggest to smallest, or smallest to biggest). They couldn’t agree on how to represent negative numbers (sign-magnitude or two’s-compliment). They couldn’t agree on how to map letters to numbers (EBCDIC or ASCII). And they really, really, really couldn’t agree on how to represent real numbers (the kind with decimal points in them).

Over time, all these arguments got settled. Except the word ordering one. They still can’t agree on that. And the final decision on how to represent numbers (both integers and real numbers) was worked out in a committee that produced the IEEE 754 standard. That standard even covers weird things like what’s the value of 1➗0 (+infinity), -1➗0 (-infinity), and what to do with equations that make no sense. For example, if you add +infinity to +infinity you get another +infinity. But if you add +infinity to -infinity, what do you get? It’s not zero. They dump all those strange cases into a bucket called “not a number.”

Think about that: In the standard for how to represent numbers, there’s also a standard way to represent things that aren’t.

“Not a Number,” abbreviated “NaN” is pretty easy to get. 0➗0 for example. Or you can get it by treating something that isn’t a number at all as a number, if you are using one of the modern languages that lets you make mistakes like that.

IEEE 754 didn’t only specify how to store a NaN in computer memory, but it also gave rules for how to handle it in equations. And one of the rules is that it refuses to be ordered. 0<NaN is false. But 0>NaN is also false. A weird result of that rule is that NaN cannot equal itself.

Suppose you are working with a computer language that uses === to mean “is equal to” (yes, such languages exist). In that language if you have a variable X and you test X === X, you’d think that had to be true. But it’s not when X is holding a NaN.

And that brings us to our tweet.

Nän (also spelled “naan”) is a kind of bread they serve in Indian restaurants. Funny story. I was once on a bus at work going from one building to another, and a new Indian restaurant had just gone in. And the bus driver said, “Hey, is that place any good?”

And I said, “Yeah.”

And the bus driver said, “So, what do they serve there? Like maize and buffalo and stuff?”

The bus driver was not joking. Sigh.

Anyway, I digress. So nän is a kind of bread, and NaN is a numberish thing that isn’t equal to itself. And unless you are that bus driver, I’m guessing you can put the rest of those pieces together.

And if you are that bus driver, I think you were one hell of a great bus driver, and I really appreciated you, and some day I’d like to take you to that restaurant for some nän.