Winners and Losers in the Book Promoting Business

I’ve been doing a lot of stuff to promote my books over the last eight months. If you scour this blog, you can read about all of it in excruciating detail. However, I thought it might be useful to do a quick-hits post about what works and what doesn’t.

GoodReads Giveaways

The only reason to run a GoodReads Giveaway is to get your book on a lot of “to-read” shelves, so that it looks legit. This is pretty much guaranteed to happen if you make your giveaway book available internationally, because few authors do. The catch is that your winner is going to be far away, and if you are sending a signed book, you are going to have to cough up shipping (and fill out a bunch of export forms). Including the cost of the book, expect to spend $30-$40. I cannot fathom any reason to run more than one giveaway per book. But I think legitimacy on GoodReads does justify that original outlay.

LibraryThing Giveaways

These are different from GoodReads in that you are allowed to give away up to 100 e-books (GoodReads only lets you give away physical books). These are a complete waste of time. I only got 30 entries (vs. over a thousand on GR) and 10 of them didn’t respond to my emails. My objective was to get a bunch of ARC reviews for my sequel on Amazon. All I got there were a couple one-sentence reviews from people who I’m guessing didn’t even read the book. I did get one actual review on GoodReads, but this was way too much work for one actual review.

Amazon Giveaways

These are a good way to very cheaply build up a following on Amazon. The people who enter Amazon giveaways don’t actually buy books, though. So why bother? I’m not sure, but I suspect that the activity that surrounds being followed (Amazon direct markets your stuff to your followers sometimes) will improve the chances of your book showing up as a suggestion when people are browsing other stuff. And even if it doesn’t work, you can basically get 200 followers for $1, so why not?

Twitter Ads

Twitter ads rule. I did one post already about them, and I’ll do a follow-up about them exclusively once I’ve done more science. But the bottom line is that if you target carefully, you can get people to your book’s page for less than $0.10 each, and a lot of those people (because of the good targeting) will buy your book. Right now I’m spending about $0.80 acquiring each buyer. Since I get $2.04 right away, and another $3.44 if they buy the sequel, that $0.80 acquisition cost is money well spent.

$0.99 Sales

You might think that lowering the price of an e-book from $2.99 to $0.99 would result in more sales. I’ve seen no evidence of this. My conversion rate (the portion of people clicking through an ad who went on to buy the book) with Twitter ads was the same at the lower price. (I’ll be doing some testing soon, to find out what happens to conversion rates at even higher price levels. But for now I know that 0.99 and 2.99 are the same price in my reader’s minds.)

The only reason to run a $0.99 sale is so that you can place an ad in a newsletter. (Most newsletters charge outrageous prices to advertise full-price books, and many refuse to take those ads at all.) Although there are lots of newsletters that claim to run free ads, very few of them actually follow through. And those that do run free ads, don’t generate any sales. So you have to pay one of the big boys to run the ad. And of course you’re only getting a royalty on the reduced price, so you need to sell a ton of books to break even. I haven’t tried doing a paid ad yet, but I’ve read the results from other people, and I honestly don’t think it’s worth it. You are better off spending your money on Twitter ads.

I found these guys when I was researching Twitter ads. For $9, they tweet like crazy about your book for about 36 hours. They have lots of testimonials about how many clicks they get on these tweets, and it looks like 700 is fairly typical. I used them and got 2 sales. I’m not sure how many clicks I got, but I’ll update this post when I find out. It really doesn’t matter though. 2 sales for $9 is an acquisition cost of $4.50 which is 560% higher than Twitter ads. The bottom line is that clicks to your page do not matter if those clicks are not highly targeted. These guys get a lot of clicks by talking about your book on accounts that have nothing to do with your genre. Total waste of money.

UPDATE: BookTweeter sent me the stats. I got 1,405 clicks. If I got 1405 clicks on a Twitter ad, I would expect about 140 book sales (vs the 2 I got from this campaign). So there you go: a well-targeted campaign will results in up to 70x better conversion, I guess.

Free Days

The theory is that you can give away your first book for free, and then people will love it so much that they buy the sequel. I’m very skeptical about that logic, because although I have a bunch of free Kindle books, I’ve never read any of them. And I’ve asked a bunch of my friends and they say the same thing. You grab it because it’s free, and then you never get around to reading it. Plus, if you are the sort of person who looks for free books, you probably have more of them than you could ever get through. So if you do read my book, it seems really unlikely you’re going to go buy my sequel with all those other books in your to-read queue.

To make matters worse, giving away books for free is expensive. It doesn’t cost anything to actually give them away. Amazon eats that cost. But it costs a fortune to acquire the “buyers.” As with the $0.99 sales, you have to pay big bucks to get a placement in a newsletter so people will know you are running a giveaway. Once you are a few hundred out of pocket, the chances you’ll ever make that up on sequel sales strike me as slim-to-none. It’s a sucker’s bet, and until I find a blog where someone says it actually worked, I’m not playing.

Participating in GoodReads Groups

I had the good fortune to stumble upon an open call to participate in a “male authors in romance and erotica” week-long chat in a very big, popular GoodReads group. The people in this group are almost all top-1% reviewers on GoodReads. They quote your book constantly in status updates as they read it, and they write elaborate reviews. They also have tons of friends who see all this stuff, and will become interested in your book. There is a strong network effect, and it has resulted in a bunch of sales. The down side is that it’s a big time sink.

Interacting on Twitter

I didn’t get on Twitter as an author. I became an author after finding my voice on Twitter. So my situation might be somewhat unusual. By the time I wrote my first novel, I had hundreds of friends and acquaintances who were my built-in audience. It’s like how most people sell their first bunch of books to friends and family, except I sold mine to my Twitter friends instead. Personal relationships account for probably 100 of my initial sales, and of course many of my subsequent sales are a knock-on effect of those people recommending it to their friends and writing glowing reviews on Amazon.

When I released my sequel, I had a direct line to about 60 people who I am absolutely sure will buy it right away. And a dozen of those people RT’d my launch announcements and are posting testimonials about the sequel without my even asking them to. It probably isn’t something another author would be able to reproduce, because you need the social network in advance of publication. But it’s absolutely been key to what success I’ve had.


I’ve done everything people recommend to help sell books (except the stuff which just seems idiotic to me, like giving it away), and I can tell you that almost none of it works. None of it except Twitter ads and leveraging social networks (first on Twitter, and now on GoodReads). But even without the social networks, if you can get a good conversion rate on your book page (which you get with a great cover, blurb, and reviews), you should be able to sell hundreds of books a year using Twitter ads. I promise to do a follow-up soon, with more details on my adventures optimizing the Twitter ad campaign.

Using Promoted Tweets to Sell Books

Coming on April 20, 2016!

Coming on April 20, 2016!

I’m doing a bit of a marketing frenzy this month, in the lead-up to the launch of Duality on April 20. Since Duality is a sequel, I’m advertising my first book Entropy. The theory is that if people read the first book, they will have no choice but to buy the sequel. To that end, I’ve lowered the price of Entropy to $2.99 (and I’ll be running a $0.99 promotion on the four days before the launch).

I’ve tried a lot of things in my quest to raise awareness about my novels. Nothing has been all that successful, to be quite honest. The bread and butter of what success I’ve had has been Twitter: talking to my followers and getting them interested enough to buy. Sometimes that leads to subsequent word-of-mouth sales. Everything else (advertising $0.99 sales on free sites; giveaways on LibraryThing, GoodReads, and Amazon; creating character accounts on Twitter; participating in “Author Chat” on GoodReads; this blog) has led to only a handful of sales at best. But I’m keeping at it, and next up in the hit parade is Twitter Ads.

It’s still early, but it looks like I may have found my silver bullet.

If you are on Twitter, you’ve seen promoted tweets. They are everywhere. It’s super easy to give Twitter your credit card number, and start promoting your tweets. I did that with my character account @entropy_sir. Then I started running little $5 campaigns, to see how they worked. I’ll spare you the gory details and jump to the conclusions.

Ignore the real-time data

It’s really fun to watch your ongoing campaign and see what’s working and what’s not. Except it’s all bullshit. The numbers in the columns usually don’t even add up to the totals they show. And any conclusions you draw will be wrong. Wait until 12 hours after the campaign is over, then go look at the numbers.

Twitter has no idea what your gender is

I actually knew this already. I once looked at the Twitter analytics for an account I have that is followed by only women, and Twitter thought it was followed by a majority of men. This was confirmed by campaigns I targeted at people who follow romance writers. That audience is going to contain almost no men, yet Twitter was sure about a quarter the people viewing were male. Lesson: Don’t choose “Women” as a way of narrowing your audience because you’ll lose all those women that Twitter erroneously thinks are men.

Include a link to your book, no picture, no hashtags you have two similar tweets but one has a picture (like the one I included here), you will get a lot more clicks on the one with the picture. But you will get a lot more book link clicks on the one with no picture. It seems people have one click in them, and if they spend it on looking at your picture, they are not going to click on your book link. Remember that when you include a link to Amazon, it puts a picture of your cover and a little of your blurb and your ⭐️ rating. So even if you only have a link to your book, it’s going to have a picture anyway. But clicking on that picture will take them to your book.

Don’t use the word “Erotic” in your tweets

BannedThis is really silly. Twitter doesn’t want you advertising “adult” things, whatever the fuck that means. And when I tried promoting a tweet in which a reviewer called my book erotic, Twitter refused to let me use that in my campaign. But yet Twitter is okay with tweets that don’t mention that about my book. It’s the same book! So I guess it’s okay to promote adult things as long as you don’t say they are adult things. Or something. I have no idea.

Big using the “Maximum Bid” option

There are a bunch of different ways to bid on ads, but only one that makes sense for promoting books. Buckle up, because I’m about to walk you through it.

First, make sure you are targeting a huge audience. I found that the best bet was to put in account names of popular authors in my genre (EL James, Anne Rice, Alessandra Torre, etc.) and big presses like Harlequin.  Most authors have almost no followers, so it took some digging to come up with a list that would add up. Whenever you add an account, the setup page will suggest similar ones. I looked at those suggestions, and a lot of them were right on target. Make sure you uncheck the box that says “Also target your followers.” Those people already know you wrote a book. When the page says your audience is a couple million people, you’re good. With a huge audience, you can make a low bid, and still get seen. You will be paying per click, so every person who looks at your page on Amazon is going to cost you money. If you lose money, there’s no point in this exercise. Let’s take my books as an example.

Assuming they buy the Kindle book, and further assuming that someone who reads my first book will read my second, I’m going to make about $5 in royalties from each buyer. Now, suppose I know that 1 in 100 people who look at my book page will buy it. That means I need to pay no more than $5/100 = $0.05 per click. Here’s the catch: I pulled that “100” number out of the air to make the math obvious. That’s not the right number. That number depends on the quality of your audience targeting (picking the right account followers to show ads to), and the quality of your cover and title and blurb, and the quality of your reviews. That number is different for everyone, and the only way to find it is to advertise and watch your KDP dashboard. Take the number of “link clicks” Twitter tells you, divide by the number of sales KDP tells you. That’s your conversion rate. Make your bid per click your total expected royalty divided by that rate, rounding down to the nearest penny.

Here’s a little worksheet to use to choose your bid:

  1. Number of link clicks from your first campaign: _____
  2. Number of sales from that campaign: _____
  3. Line 1 divided by line 2 (this is your conversion rate): _____
  4. Expected lifetime royalty earnings from a reader: _____
  5. Line 4 divide by line 3 (round down): _____

Line 5 is your maximum bid. Use $0.10 for your first campaign, and then put the results in the worksheet to get the bid for your next one. And keep tuning. Note that you can put in a lower number than the number on line 5. If your number is a lot higher than $0.10, go ahead and cap it at $0.10. Using a higher bid is like paying for expedited shipping on your proofs. It gets you the results faster, but in the end it’s just a waste of money. Patience, grasshopper. We’re in this for the long haul.

Note that if you find that your bid per click is too low to get any impressions, then you need to increase your target audience. Don’t increase your bid higher than line 5 because you will lose money. And as I’ve argued before, you should never lose money when advertising.

Promote tweets with snippets of your reviews

It really doesn’t matter what tweets you use, because you are only paying per click. So if you tweet something that doesn’t garner many clicks, it just means you’ll have to wait longer to spend your money. Nonetheless, it certainly couldn’t hurt to promote tweets that actually draw more clicks, right? I tried all sorts of things, and the ones that got the most clicks were the ones where I included a snippet of a review. This tweet is the champ:

My theory as to why: it establishes credibility, it makes a bold statement, it doesn’t actually say the book is good. So it makes the viewer curious. There is an effective sales technique where you tell the prospect, “My product probably isn’t for you.” The prospect then instinctively tries to convince the sales person that the product really is for them. I have no idea why it works. But I think this tweet kind of does that when you put it in the face of romance readers, which is exactly what my targeting did. (That is from an actual review by the way. I didn’t just make it up.)

Another tweet that did really well in getting clicks followed the same pattern:

Again, it didn’t say the book was good. I had a bunch of other review tweets that talked about how great the writing, and the character development, and stuff like that was. Those didn’t get anywhere near as many clicks.

So while you should use reviews (and only five-star ones, obviously), my advice is to grab words that are intriguing, rather than complementary.

Is this the silver bullet?

Maybe? I just started this, but the results have been outstanding. My conversion rate (line 3) is 8. One out of every eight people who look at my page by the book. That’s insane. No book should have that rate. When I ran my first couple tests, where I didn’t target romance readers explicitly, I had no sales at all. So picking the right accounts to chase appears to be important.

My line 3 number will probably climb to something more reasonable over time, and I’ll do an update when I have more experience with these. But so far, I’m loving Twitter ads.

Demystifying Amazon Giveaways

The real one bounces

The real one bounces

Amazon got into the “giveaway” business about a year ago, and they have been tweaking and tuning and changing them ever since. As a result, any research you do on whether they are a good promotional tool is going to turn up a lot of outdated information. If it isn’t Spring 2016 as you read this, then consider this post outdated too. Go look for a better source.

One of the recent changes was allowing you to give away Kindle books. That seemed like an interesting new approach to raising awareness. If you follow this blog, you know awareness is the only problem you have as an author promoting your books. But there are so many questions! So for the past couple of weeks, I’ve been sciencing the fuck out of these giveaways, to try to get some answers.

How an Amazon Giveaway Works

You start on your book page, and go down to the bottom, and then choose to create a giveaway. You will be buying the book that Amazon will be giving away. But you get your royalty. So, for example, if your book is $2.99 and you are in KDP Select, you’re going to get $2.04 back, so the actual cost of the giveaway is only going to be $0.95. It’s going to cost you that even if nobody wins, because they don’t refund Kindle books (unlike physical prizes). If nobody wins you can either get a gift code for the book, or you can run another giveaway.

Amazon has three different ways to run the giveaway. But only one of them is useful. There is one where you pick odds and it randomly decides whether each entrant wins based on those odds. This is not good, because even with odds of 1:1000, there is a 9.5% chance there will be a winner after only 100 entries. (1-(1-0.001)100) What you want to use is the one that awards the book to the nth entrant. That lets you control with certainty how many eyeballs there are before someone wins.

There are two strategies you can use to maximize the number of people who enter to win a book. I’ll call these endless giveaway and maximum excitement.

The Endless Giveaway

In this strategy, you set the giveaway duration to 1 day (which really means two days), and the odds to 1:500. You will get 100-150 entries before it’s over, so nobody wins. Let’s say you get 100. Then you take that book and use it in a new giveaway at no charge. Again, about 100 will enter and 59% of those people were also in the first, so you’re up to about 141 total eyes. Now you run it again. I’ll spare you the math, but you’re going to get 17 new eyeballs for a total of 158. Each time you’ll get fewer brand new people. You can keep doing this forever. Nobody is ever going to win that book. Eventually you will stop reaching new people. Everyone who enters giveaways will have entered yours.

Maximum Excitement

In this strategy, you still set the giveaway duration to 1 day, but you set the odds to 1:200 or better. I tested this, and 1:200 is a magic threshold. At 1:250 or worse, and you will get about 100-150 entries over the course of two days. But at 1:200 or better, you will get all 200 entries within a couple of hours. You will give away the book. This is probably the better strategy because it’s more honest, and because you’ll reach those 200 people with a hell of a lot less effort than running winner-less giveaways day after day with the same book.

Note that in one case, Amazon screwed up and didn’t award the book to entrant #200, and the giveaway just kept on going with no winner. If something like that happens to you, just let the timer run out, and then use that book in a new identical giveaway.

Now Set Up Everything Else

You have the option of making the entrant follow you on Twitter, or on Amazon, or watch a video, or you can just let anyone enter. So what should you choose? Have them follow you on Amazon. I’ll explain below.

You assign the giveaway a title, and type a message the people entering will see when they get to your giveaway page. They won’t read this message, because there is a bouncing box on the page that they are supposed to click on. So it doesn’t really matter what you say in the welcome message. But you have to write something, so put in something enticing about your book, like this:

Entropy is a whole different take on the romantic novel. It’s a smart and sexy look at an online affair. 4.6 Stars! Watch for the sequel Duality, coming on April 20, 2016!

You then need to upload a picture. Use the cover of your book. The one you used when you did the KDP setup.

Next comes the winner message. This doesn’t really matter, because at most one person is going to see it. Give them the address of your mailing list, or your Twitter handle, or something like that.

And now the loser message. This is the most important thing. Almost everyone who enters your giveaway is going to see this message. This is the one chance you have to convert that person from an entrant to a buyer. What should you say? Hang in there—after we get all these mechanical details out of the way, I’ll go through some different options.

Amazon will then approve your giveaway (in about 5 minutes during business hours, in a couple hours otherwise). And then you’ll get a link. Go to Twitter and tweet a message including the hashtag #AmazonGiveaway and that link. Don’t attach a picture. I tested it, and having a picture did not improve participation. The link will cause your tweet to include a nice Twitter Card, including your cover. And actually, this tweet doesn’t matter because almost nobody is going to see it except a robot. But you have to do the tweet to get the party started.

Your followers already know you wrote a book. So this exercise is not for them. You are trying to build awareness with a new audience, so the sole purpose of that tweet was for a robot to see it. The hashtag is critical: if you don’t include it, then the robot won’t see it. What robot? The one that drives a site called A few seconds after you do your tweet, your giveaway will show up on that site.

Then if you’re lucky a mommy-blogger on Facebook will see it, and do a status update about your giveaway. She’s like a curator, deciding what giveaways are worth entering. A few seconds later, between 50 and 100 women, aged 20-60, often mothers of young children, will enter your giveaway. And then over the remainder of your giveaway, you will get the rest of your entrants, pretty much all in that same demographic. You should only consider doing a giveaway if that’s your target demographic. These people will come from other Facebook pages like that first one, or from the listing site. Only 4.8% of entrants will start by clicking your link on Twitter.

So Many Questions!

There are so many variables, leading to so many questions about these giveaways. Since they are cheap, I ran a lot of giveaways over the past few weeks, playing with different variables, to find answers to my questions. Here are the questions I had, and my best guesses as to the answers.

What is the point of running an Amazon giveaway?

Your objective is to build awareness of your book, and hopefully convert a few entrants. If you sell a single book, you’ve more than paid for the giveaway. And even if you don’t, that’s a few hundred more people who have seen your book, and your name. This is what is called brand marketing—making people aware of you as a brand, and increasing the likelihood they will buy eventually. You should only do it if you have multiple books. If you only have one book on the market, you aren’t a brand yet.

When should I run it?

2pm Eastern on a weekday. I tried all sorts of different times, and trust me—that’s the best one.

Should I require a Twitter follow to enter?

Absolutely not. The people who enter your giveaway will do so using spam accounts that they only use for entering giveaways. They will never see your tweets because they don’t use these accounts. Requiring a Twitter follow probably reduces participation a little, but it was not by a significant amount.

Should I require an Amazon follow to enter?

Absolutely. Requiring that entrants follow you as an author does not lower participation at all, compared to running it with no follow requirement. Amazon says it will send a message to your followers when you release your next book, and sometimes they will even ask you to compose a message yourself that they will send to your followers on your behalf. (I’ve read; this hasn’t happened for me yet.) Oddly, there is no way to find out how many Amazon followers you have.

It’s also likely that if someone is following you, then your stuff will show up more often in Amazon’s recommendations to that person. I don’t have evidence of that, but it seems like something Amazon would do. And the more someone sees your cover, the more likely they are to eventually buy your book. (I assume. I don’t know that for sure either.)

What odds should I use?

Making winning a long-shot does not lower participation once you exceed that magic 1:200 threshold. So if  you are using the endless giveaway strategy, you can use 1:500 and be absolutely sure nobody will win. If you want a winner fast I found that 1:200 and 1:150 were equally enticing, so go with 1:200 to maximize eyeballs.

Who enters these giveaways?

I ran a giveaway in which I required a Twitter follow to enter. I discovered that almost everyone uses spammy accounts that have nothing but contest entry tweets. And based on the names and the few that had bios, I can tell you for certain that men do not enter giveaways. Only women. And most of those women had “mom” in their handle. Several mentioned in their bios that they are disabled. I DMd a few, and that’s where I learned that many of them get the links from Facebook mommy blogger / giveaway curators. A few also go to that giveaway listing site directly. There might be other robot-driven giveaway sites, but I didn’t find them.

What should I say in the loser message?

When I realized that pretty much everybody sees the “lose” message, I hoped that I could craft something that would convert some portion of those people to buyers. I couldn’t. I converted maybe one person. I tried a sales pitch. I tried being funny. I tried being mysterious. Nothing worked. So while I still think that what you put in here is really important, I don’t have a definitive answer of what that should be. (If you have a great idea, put it in the comments, and I’ll try it!)

On the lose page, right below your message, there is a link to your book, and below that there is a button to deliver a sample of your book to their Kindle. I tried encouraging people to do each of those, and neither ended up converting to a sale. However, my gut says that if you could get them to download a sample, that’s your best bet. Every time they look at the stuff on their device, your sample will be looking back at them. And of course if they eventually read your sample, they’ll buy your book, because you’re awesome!

Here is one of the messages I used:

You didn’t win, but you can still get something free! See the button that says “Download Free Sample”? Click that! In a few minutes you can start reading a sexy novel that I bet you’ll love.

I have no idea whether it worked, because Amazon won’t tell me how many samples have been downloaded. But my advice is to do something like that.

Let me just take a moment to rant about how horrible Amazon is at giving you insight into whether your marketing works. Among the things you should be able to find out, but cannot:

  • How many followers you have
  • How many sample downloads you have
  • How many people looked at your book’s page
  • The sources of people visiting your book’s page
  • Correlations between that source and conversion (actually buying it)
  • Correlations between reading a sample and conversion

It is in Amazon’s best interest to let authors/publishers know this stuff. Okay, rant over.

Should I put a link in the lose message?

It just so happened that while I was running these Amazon giveaway experiments, I was also running a giveaway on GoodReads. So I set up a link to the GR giveaway and tested this lose message:

You didn’t win the e-book, but maybe you can win the print edition! Copy this into your browser’s address bar:

I figured if people were willing to go to the trouble of following a Twitter account to enter, they would probably go to the trouble of copy/pasting a link (links are not “live” in the loser message unless they link to Amazon itself, unfortunately). The cool thing about is that it tells you how many people clicked the link. One person. A single person copied that link. So don’t bother with links in your lose message.

Should I run multiple giveaways?

I wondered whether I was just getting the same people over and over. So I ran two identical giveaways (same day of the week, same odds, same start time, etc.) but associated my Amazon account with my “Sir” character account for one, and my “Kitten” character account for the other, and in each I required a Twitter follow. I found that there was a 59% overlap between the two giveaways. That is, three-fifths of the people entering the second one, had entered the first.

That means 40% of the people on the second giveaway were new. So running multiples will pick up a lot of new eyeballs each time. And since this is primarily a branding exercise (since I wasn’t able to get conversions to actual sales), having some of the people see the same book over and over probably isn’t so bad.

Wrapping up

So to summarize, I think running an Amazon giveaway for your e-book is a good idea. It gets a lot of brand new eyeballs on your cover, title, and name for almost no money.

Questions? Ask them in the comments below. I might know the answer, or perhaps I can run some more tests to find them for you.