Sunday, January 28, 2018

Amazon Ranking and Bestseller Lists - What's the Deal?

It's really hard to draw conclusions in the self-pub marketing game. After almost ten years of self-publishing on Amazon, I still don't know why some ebooks sell more than others.

I've always believed the secret to success is luck. You can improve your luck by writing great books, having great cover art and great book descriptions, writing more great books, and...

Advertising? Social media? Newsletters? Amazon visibility?

I know for certain that social media works. When I have a new title, or I put something on sale, I Tweet it and put it on Facebook, and sales go up. Cause and effect.

When I have a new title, preorder, or special sale, and I send out a newsletter, sales go up. Cause and effect.

When I get a BookBub ad, sales go WAY up. Cause and effect.

But how about other kinds of ads, like AMS? Is the key to being successful on Amazon, being visible on Amazon?

It's a seductive idea. After all, Amazon has been designed to entice the reader to click, and to buy. Every pixel on the Amazon page has a purpose.

Surely the bestseller lists, and product ranks, have a purpose as well. They must help sell more books, or else why would they exist?

Does getting on a Top 100 Bestseller List on Amazon give your book enough added visibility to fuel a lot of sales?

Or is book rank more about satisfying authors and their egos, and giving prospective buyers more secondary information to corroborate a prospective sale, rather than being the thing that introduces buyers to a book?

I've been asking myself that question a lot.

My first experiment was observational and highly subjective. I looked at how I bought things on Amazon.

We rarely examine our own buying habits, and volumes have been written about why we buy things, but I buy a lot on Amazon, and here's some of the things I've noticed.

Buying Situation #1 - A Need

This is the main reason I buy things. I have a need for something, so I search for it by typing it into the search box. Say it's a hammer. I type in 'hammer', and maybe narrow it down by category. Then I see what comes up, and compare a few products based on prices, appearance, and customer reviews.

I don't read customer reviews, unless a product has too few, or a lot are bad, or a lot are overwhelmingly good and it seems like they might be fake. (We can all spot fake reviews that were written by shills).

Maybe one product pages will take me to more pages by showing me what customers also bought, or sponsored products, or what else customers looked at.

Then I buy something.

Buying Situation #2 - I Know What I Want

If I know the exact product, I go directly to it. Maybe price is a consideration, but if I'm brand loyal (a specific author, a kind of shampoo) I don't worry about price, reviews, or competing products. I click on it.

Buying Situation #3 - I'm Shopping

I have a vague idea that I want something, like a book, but I'm unsure of what I want, so I'm just going to poke around Amazon until I find something.

Of the three kinds of shopping I do, this third one is the least common for me. And here's why:

I know the authors and types of books I like, and usually buying one of them leads me to instant recommendations about others I'll like based on customers like me. Hell, Amazon sends me emails based on things I've bought and my browsing history. Not a lot of poking around needed by me; it's all on one page.

Do I ever browse the Bestseller lists?

As an author I do, to see who is selling.

As a reader, never. Neither does my wife, an uber-reader who burns through seven books a week. She's always got a full Kindle because Amazon recommends books, she signs up for the newsletters of authors she likes, and she gets BookBub emails daily.

So we have two readers in our house who spend a lot on Amazon, and neither of us buys books because we discovered something new on a bestseller list.

My buying habits may not be yours, however. I know many readers who learn of new books on Goodreads, and I've never gotten much use out of that platform, as a reader or an author. I think Goodreads is terrific, but I just don't use it. I'm certainly not obsessed with my Goodreads Author Standing as I am with my Amazon Rank.

But should I be worried about Amazon Rank at all?

What's the relationship between causation and correlation in regards to Amazon rank? Does a high rank help a book maintain sales through visibility? Or is it only an indicator of book sales and the added visibility does little for sales?

My own efforts inform me that it is promotion that leads to sales, and an incidental side-effect of sales is the rank position and any subsequent appearance on a bestseller list, which doesn’t help the book much.

Some authors seem to believe that appearing on a bestseller list leads to extra visibility and additional sales.

Am I right? Are they right? Maybe a little of both?

It’s difficult, if not impossible, to conclusively prove either of these approaches, because there is no way to conduct a perfect experiment with a control. Other books influence the ranks and positions of our books, and there are always “boosts” that we can’t account for; someone reads a promotional Tweet or Ad a day or two after it runs, a promo site finds the book and promotes it, giving it a boost, word of mouth spreads, Amazon has a delay or lag with their site update...

So I can’t think of any way to definitively state that appearing on a bestseller list is an effective sales tool.

But I do have plenty of evidence to show that a bestseller list appearance doesn’t do much when it comes to additional sales/downloads. I’ve become convinced that rank makes authors feel special, but doesn't boost sales by any significant degree.

I say “significant” because I’m sure being on a list is helpful. Just not very helpful, and not worth pursuing.

That may seem counter-intuitive, but you’ve probably seen examples with your own books.

I recently made my book ENDURANCE free for five days. I didn’t do any ads, but did link to it on social media. First day, 1375 downloads, and it peaked at #70 on the Top 100 freebie list.

Now, this book had nothing pushing it up the charts, just my Tweets and Facebook, and anyone who retweeted, and those websites that promote free books on their own initiative to make Associates dollars.

After that initial social media burst, there was no other promo for ENDURANCE. This was intentional on my part to prove a point.

Being on the biggest bestseller list made ENDURANCE visible to readers who are presumably searching bestseller lists for ebooks, so once I hit #70 I should have had hundreds or thousands of new eyes on it, and corresponding downloads because, hey, it was free.

But you can guess what happened once my promotion stopped. Once I stopped promoting, it stopped climbing. By Day #2, it was #95, and on day four it was #214. Whent he promo ended it bounced to the paid list at #42,844.

According to William Ockham who figured this out with Data Guy, “A sale has a half-life of one day. A sale 24 hours ago is worth 1/2 of a sale made right now. A sale an hour ago is worth 23/24ths of a sale made now. Sum the value of sales for each eligible title and rank order the results.”

So the moment you peak, that value begins to immediately start fading unless you continue to sell at that speed, a higher speed, or a slightly lower speed. But that slightly lower speed needs to pick up as time passes, for a rank to maintain its spot. It has to run faster to stay in the same place.

The added visibility of being on a list, without any promo to back it up, doesn't do much to sustain position.

Promos will help a book sustain its rank, because they are driving more downloads. So as long as the promotion/marketing/ads/retweets/facebook boosts continue, a book can retain its spot, or even climb up.

Without any promos, the only thing to drive a book’s downloads is its own visibility on a bestseller list.

When I hit #70, if appearing on the list and being visible to all of those people browsing the list had a big impact, you’d expect a lot of downloads, and for it to continue to climb.

But being on the list wasn’t enough. By day #2, ENDURANCE only had 729 downloads. Day three, 331, day four 160, day five, 113.

While it isn’t on the Top 100 Bestseller list, it was still on many other lists (Horror, Occult, Whispersync) in high spots.

But those high spots aren’t generating enough new downloads for it to sustain its position. It kept dropping without any outside promo push. And this is a freebie list. Being #1 in Horror—a big category—as a freebie only yielded 300 downloads.

That's 300 downloads if you believe that people only downloaded it because they say it on that bestseller list (or another list). But I know that the reason it peaked on the list was my sale and social media push. That got it on the list.

And being on the list didn't keep it on the list.

ENDURANCE isn’t an anomaly. I did a BookBub with BLOODY MARY recently, hit #1 on the overall free list with 45,000 downloads. But once the promo stopped, the book dropped like a rock.

My experience has been that being on a bestseller list doesn’t have much of a positive effect on a book’s downloads.

I did 1375 downloads on Day #1, while promoting ENDURANCE. Day #2, I had 729. One might attribute those 729 to list visibility, but there is no way to be sure. I could have had people clicking on my Twitter and Facebook push the next day, which meant my promo was still working. Those websites that promote free books could have picked it up the next day, giving it a little push. And there are probably reasons I don’t even know that factor into who downloaded that book on Day #2, and why. During a book launch, for example, it may take some fans a while to discover that a new book is out, or word-of-mouth can build, or social media can build, so a book can continue to climb. But that’s still a push doing it, not a by-product of hitting a list.

This happens every time I do a KU freebie or countdown without a sustained outside promotional push.

I have many other examples. My books are always on the WOMEN SLEUTH bestseller lists, but they are often below #100. Since you can’t browse below #100, they get zero visibility from being #202, but they still consistently sell somehow. My belief is readers find them from actively searching for the next book in my series, or through Amazon's "Customers Also Bought" feature.

I’ve been #1 on both Free and Paid Bestseller List, several times. It feels good, but it doesn’t stick unless massive promotion backs it up.

Rank may indeed be a measure of success and sales, but rank isn’t the reason for success and sales.

A simpler example of what I’m saying is watching box office results for movies. STAR WARS hitting the #1 box office spot probably didn’t lead to many more ticket sales, people who said “Hey, look at Star Wars at #1—I’d never heard of it before but now that it’s visible I’ll go see it”.

When STAR WARS hit #1, do you know what #10 was? No. And neither does anyone else. Because people aren’t looking at box office lists to determine what to see.

What draws attention to movies? Ads. Marquees. Reviews. Theater websites. But the movie theater websites aren’t ranking movies. AMCtheaters.com doesn't have a list of which movie is the most popular so you can decide what to see by rank.

Where is the evidence that hordes of readers—or even handfuls of readers—are discovering new books by browsing the bestseller lists?

It may be helpful—to what degree, I don’t know. But I do know, with 100% certainty, that hitting a list will not keep you on that list unless there are other factors involved.

The above examples were for free books. Here is a current promo I'm running with a paid book.

I currently have four box sets all on sale:


I set all four at 99 cents on the 26th, and Tweeted three of them, listing the book titles, and one where I didn’t list the titles (Origin)

One the 26th, just via Twitter, I sold:

Whiskey Box - 270
Dirty Box - 185
Afraid Box -147
Origin Box - 32

That was the order I tweeted in, one after the other, and pretty much worked as I expected. I’ve never had these box sets at 99 cents, so a lot of thrifty shoppers were waiting for this deal.

My Jack Daniels books are more popular than my horror, which is expected again. But the sales directly aligned with the order I tweeted.

I tweeted four times in a row. The first Tweet, with Whiskey, got the most downloads.

Then some of my followers began to go Tweetblind. Or they didn’t read closely and thought I was repeating. Or they were overwhelmed. Dirty was next, then Afraid, and finally Origin, and with Origin I didn’t list the titles, just tweeted I had another box set.

Lessons learned: When promoting more than one thing, don’t do it one after another. And when doing a box set, list the titles.

The next day, I did two promotions. For Origin, I did a BookBub ad in the horror genre. For Afraid, I did a newsletter swap with four other well-known horror authors.


The above is a jpg. Here's the link to the page:


The five of us all sent out a newsletter yesterday, promoting all five box sets. So yesterday I had a BookBub for one box set, this promo for the other horror box set, and did no promos at all for my two Jack Daniels sets.

On the 27th I sold:

Whiskey Box - 64
Dirty Box - 63
Afraid Box w/ Group Newsletter - 431
Origin Box w/ Bookbub - 1519

Afraid squeaked into the #300s with rank. Origin climbed to #80.

All of us became Top 50 Horror Authors. We all climbed up on various bestseller lists. I made enough in sales to make up what it cost to send out my newsletter, and of course BookBub always pays for itself.

But our ranks aren’t sticking. They’re dropping.

In my case, neither of my horror box sets have even been on a bestseller list, because I've kept them at $9.99.

So my anecdotal evidence, that I don't shop using the bestseller lists, seems to be supported by my experiments. Not many others seem to shop using the lists, or else we'd all be sticking or climbing, rather than dropping.

Stop worrying about your Amazon ranking. It is an indicator of sales, not a driver of sales.

What does cause me worry, however, is how to maintain sales. While the vast majority of titles appear on a bestseller list, then drop off pretty fast, some of them do tend to stick around for a while.

Why? Could the added visibility of high ranking be helping some books and not others? Or is something else at work here?

I have an idea, but I don't think anyone is going to like it.

Every book, and every backlist, has a ceiling. The ceiling is based on:

1. How many people would like your books if they encountered them.

2. How many people you can help to encounter them.

3. How many people will encounter them because of reasons that have nothing to do with your efforts.

You can write great books in a popular genre, so there is an established base of fans out there. And you can (and must) continue to promote those books. Here's some of the promotions I'm constantly using:
  • I always have a book on sale and a book that's free.
  • I send out newsletters.
  • I do ads (Bookbub, AMS, Twitter, Facebook).
  • I use social media (Twitter and Facebook).
  • I update my website.
  • I update my book backmatter to include new titles.
  • I update my book descriptions and Amazon author page to include new titles.
  • I experiment a lot.
After that, it's out of my hands. Remember your Serenity Prayer.

Here are some things beyond your control that could happen, and this is (probably) why some books seem to stick on the bestseller lists.
  • You could have a publisher promote you.
  • You could have IP tie-ins (TV, movies, comics).
  • You could get media exposure.
  • Word of mouth could explode/go viral.
Basically, I believe we're limited by our own ceilings, unless extraordinary luck steps in. 

You can gradually raise your ceiling by writing more books, and doing all the things I mention above. And the more you do, the easier it will be for luck to find you. 

But you need to stop worrying about your Amazon rank, and stop judging your success by appearances on the bestseller lists. 

Your time is best spent tending your garden. Keep writing, keep promoting, keep growing that readership. 

There's no secret here. There's only hard work and luck.

11 comments:

Mike Hall said...

As a book buyer I'm pretty much like your wife in the way I find books (though my reading rate's a bit lower.) I never look at rankings or best seller lists.

However, when I recently commented to this effect on a blog posting about "rank stripping" and queried whether it really mattered, I was assured that rank had a significant impact on Amazon's recommendations and on what appeared as "also bought". I have no idea if this is true but if so, it could in theory carry through into my purchases; your results suggest otherwise for fiction (and some of the recommendations I've received are very clearly driven by my interests rather than sales rates, which for some of the recommended books are very low, though this may be because they involve smallish non fiction niches.)

Joe Konrath said...

It doesn't make sense that also-boughts are impacted by rank. Also-boughts are triggered by sales, and sales do effect rank. The more sales you have, the likelier you'll be listed as an "also bought" for customers who also bought it, probably in the same way that a higher bid per click will get you an ad impression. Amazon likely weighs more sales over fewer sales when it does "also-boughts"

But you can't have a rank without the sale in the first place. That's confusing causation and correlation.

When it rains, more people carry umbrellas. Cause and effect. When it rains, more people wear jackets. Also cause and effect. But rain effects both. Saying there is a connection between those who carry umbrellas and those who wear jackets is spurious. One doesn't effect the other. They are both caused by the same thing.

When you have an Amazon sale, it effects rank. When you have an Amazon sale, it effects also-boughts. But rank doesn't effect also-boughts. They're just caused by the same thing.

I'm pretty convinced that rank stripping doesn't matter. Amazon learned that authors care, authors made a big stink about scammers who were using click farms, so Amazon responded by stripping the rank in order to placate the authors who care about rank. That backfired by stripping ranks of authors who supposedly weren't scammers, but I'm betting Amazon knows that doesn't matter much--sales will still happen regardless of having a rank or not. Books are still searchable. There are still also-boughts when rank is stripped.

BTW--scamming authors don't hurt the rest of us. It's not a "communal pot".

KENP is decided after the pay period ends. It isn't zero sum. It has never been zero sum. Some scammer isn't taking money out of your pocket.

These scammers are taking a bunch of money from Amazon. A click farm with 2000 kindles, would earn an author of a page stuffed 500 page book $4800. And this can be done without the "skip ahead" loophole. If a click farm has a whole line of Kindles, they can pay someone to manually click each page and still make money for the author who bought the service, even charging $500 for 2000 downloads.

I just wasted a few hours of my life looking at the romance lists, and I spotted at least two dozen authors with ebooks that were page stuffed, with descriptions and covers that were ridiculously similar.

Maybe it's one company, or a few people, doing this. Maybe it's a few dozen people doing this.

But they aren't taking your piece of the pie. The rank they have isn't what's making them money; I'd guess it's the click farms. Rank by itself doesn't help much, and I humbly ask anyone reading this to show me data to prove me wrong.

The only thing these scam books are taking is All Star Bonuses... which are bonuses. That sucks, but it's not stealing money directly from you.

Simple math shows that Amazon could be losing a hella lot of money on this. Several hundred grans a month, at minimum.

But that's Amazon's money. Not yours.

Joe Konrath said...


BTW, the rank stripping saga has obvious parallels to the Dr. Seuss story, The Sneetches.

In the story, there is an intangible thing that the people care about--in this case, a star on their belly.

Like Amazon authors care about rank.

But this thing (the star and the rank) has no inherent value or meaning. It's only important for those who assign it some value.

In the Sneetches, a guy comes in and manipulates the people who care about stars by offering to create and erase stars for them, depending on their whim. He learned that they cared about stars, and was able to profit off their obsessive behavior.

Amazon knows rank is worthless, just as the guy knew stars were worthless. But Amazon sees how authors react to rank, how they value it, and can thusly punish authors who break the rules by stripping rank, at the same time appeasing though who believe rank is being abused.

It's like telling your kid to eat their veggies or you'll tell the monster under the bed.

I'd love to see some evidence that proves I'm wrong. But I'm convinced that if you ran a BookBub ad you'd sell practically the same amount on Amazon whether your rank was listed or stripped.

I'd also love to see some evidence that authors, without using click farms, can maintain huge sales. I had a discussion with an anonymous writer on K-Boards about this, and he insisted that a combination of low prices, bundling, newsletters, and advertising, can make a book a huge hit.

I just tried doing that, with my recent newsletter swap, and had a moderate, temporary hit.

I'm betting he's using click farms. And, again, that doesn't matter to the rest of us, because he's not taking any of our business or money.



Ed Ryan said...

Great insight, Joe. Thanks.

Out of curiosity, are you seeing poor returns with AMS - i may have misinterpreted you, but you seemed to indicate you were not convinced you were getting the same return with those ads as you were with Facebook or Twitter


take care

Joe Konrath said...

Ed, with the exception of BookBub, I'm not impressed with any ads I've done. I can make back my ROI, but I'm not seeing big results.

Susan May said...

Hey Joe, great post and I spent some time writing a reply but turns out it was too long to put as a comment. Your program wouldn't let me.

But I had some valid thoughts I believe on your post and felt other authors may benefit from them, so I created my own blog post to back yours. It's called "You Can't Bank Rank."

I hope it's okay to post here.

https://www.susanmaywriter.net/single-post/2018/01/29/You-cant-bank-rank

Let me know if posting this here is not okay.

Joe Konrath said...

Thanks for the link, Susan. I encourage my readers to check it out.

J.R. Pearse Nelson said...

“What does cause me worry, however, is how to maintain sales.”

Yes! Amazon is trying to sell more products, not focused on selling more of YOUR books. Focus on having the best products you can. Keep producing more content. Be available on other retailers to maintain more consistent income. Build your brand, and not just through Amazon.

Mark Asher said...

So you said something about not having to search bestseller lists for books all that often because Amazon was recommending books to you?

Maybe Amazon's internal lists they may not publish might be pushing their own form of bestsellers at you. Say they think you like books with tall men who wear dark suits who are serial killers. Maybe they have their own weird list of those and the recommends they send you are randomized from the top 25 in their internal list?

What Amazon chooses to present and make visible seems to me to be very powerful.

Joe Konrath said...

Maybe they have their own weird list of those and the recommends they send you are randomized from the top 25 in their internal list?

Amazon certainly recommends books based on your browsing and buying history. I don't know if that is "internal lists" as much as it is their programs understanding your buying habits.

I don't think the "also-bought" recommendations are predictive. Right now, I'm in the "also-boughts" of the above horror authors I did the newsletter with. And out sales are still slowing down. "Also-boughts" is certainly helpful, but that alone isn't able to sustain the surge that our newsletter campaign caused.

And by "surge", we're talking less than 4000 sales among the five of us. On newsletter day, I sold 431 copies of my 99 cent box set. Yesterday, still on sale, I sold 43 copies.

It got a high rank. It got on bestseller lists. It's cheap. And it's in also-boughts with four other authors who sell well.

Why aren't sales sustaining? Why are they dropping before the Countdown Deal is even over?

I dunno. But it's interesting to me.

Alexis said...

I'm a new author (6 months) with a nonfiction book that's selling far better than I would have thought. Amazon makes it almost impossible to know where sales are coming from despite using custom bitly links everywhere in a vain attempt to track sales. Book sales are up 25% in January although I've done no advertising. I have no idea why this is and thus cannot replicate the bump. My pat response when people ask about how I drive sales on Amazon is ¯\_(ツ)_/¯

It's hard to triangulate the value of being on a bestseller list. But I do believe that in the nonfiction space, people browse the top 100 and thus there is value to being there. But I can't prove it so ¯\_(ツ)_/¯

The also boughts in my category are huge (13-20 pages of also boughts on most of the top books in my category). How many times will people click through to scan also boughts? For Google search results the answer is nobody - almost nobody clicks past page 1 of the search results. If the same pattern of behavior applies to Amazon then being in the also boughts is meaningless if you aren't at the very top. So ¯\_(ツ)_/¯

I like author rank as a quick gauge of how sales are going generally but I agree with your assessment - it's a data point but doesn't really impact anything in a material way. But what do I know ¯\_(ツ)_/¯