Author Services Guide: Using Amazon’s AMS Platform Effectively

This week, I've talked a lot about Amazon's AMS platform, first announcing the system and my hopes for it, as well as ranting about some of the more idiotic elements of their creative guidelines. Today, I'm going to go into some nitty gritty detail about how to take advantage of the AMS platform.

Marketing sucks. Many authors break out into a cold sweat at the thought of trying to promote their work. Spending money on advertising is scary business. Many authors stubbornly decide organic advertising (things not paid for, which grow naturally) is the only way to go about it. There is a lot of potential in pursuing paid advertisements, if you understand how to mitigate risk and maximize performance.

I know those pretty catch phrases annoy a lot of people, but that's the name of the advertising game. The idea is to mitigate your risks–or control your expenditure to ensure you're not throwing away your hard-earned cash–while also ensuring you're getting the best bang for your buck. It's not easy. It's also absolutely impossible to guarantee the performance of an advertising campaign.

To begin, I'm going to explain the differences between the two different services AMS currently offers advertisers: Product Targeting and Interest Targeting.

Interest targeting is the easiest advertising option for authors. You select a category you want and Amazon will serve ads to those who are browsing products in that category. It's very broad, giving authors a higher chance of getting impressions. (Impressions are ad views.)

However, it's a broad category. Science Fiction and Fantasy can cover a lot of territory, and your specific book won't appeal to every Sci Fi and Fantasy fan.

But, this form of advertising is the easiest. All you need to do is set a cost per click value and run the campaign. Set up is easy, as is basic management. In an interest-based advertising campaign, there are only a few figures you will need to know to be able to manage your campaign.

Please bear with me, as there is some math involved–don't worry, though. It's very basic math. If you can add, subtract, divide, and work with percentages, you have all of the knowledge you need to make sense of the numbers.

Impressions and number of clicks meld to become your click through rate–or the percentage of clicks versus impressions. Your mileage will vary, but don't be surprised if your click through rate is below one percent. This is normal. In a cost per click campaign, you're less worried about your click through rate as you are your return on investment. This is how many people buy your product after clicking on your advertisement. This buy rate is your major concern–and what we're gunning for with this ad campaign.

So, how can you tell if you're getting a good return on investment? It's a number game. There are three figures you need for this: Your total number of sales, your total number of clicks, and your cost per click. Here's the basic math:

Note: These are hypothetical figures, as I do not currently have sales figures. (The AMS system is simply too new.)

Step One: Calculating Number of Clicks per Sale (Click to Sale Ratio, AKA Conversion Rate)

Clicks divided by Sales = Number of Clicks per Sale.

Example: 115 users clicked on my ad; 10 users bought the book.

115 / 10 = 11.5.

Step Two: Calculating Investment

Click to Sale Ratio multipled by Cost Per Click.

Example: 115 users clicked on my ad; 10 users bought the book. I paid $0.15 per click.

11.5 * $0.15 = $1.725

Step Three: Calculating Return on Investment

For the purposes of this example, my royalty per sale is $2.15. I paid $1.725 per sale in my ad campaign.

$2.15 – $1.725 = $0.425 (Actual Return on Investment)

$0.425 / $2.15 = .19767 (ROI Percentage, AKA 20%)

Step Four: Understanding the Results

In this scenario, my return on investment is $0.425, which is a 20% profit margin. 20% is not a high margin. That said, it is a profit. Any time you come out with a positive return on investment, you're a winner. You've gained more sales than you've spent money. This is the name of the advertising game.

While a lot of analysts are likely rolling in their graves at this simplified version of data management, this post isn't designed to make you a statistics expert.

It's here to teach you the basics. In an interest-based ad campaign, such as the one Amazon AMS allows you to do, this is the information you really need to focus on.

Now that you've had a crash course in Marketing 101 for cost per click campaigns, we're going to go kamikaze and take a look at product-based advertisement. Product-based advertising, in theory, should have a lower impression count but a higher click through rate and conversion rate. Conversion rate, by the way, is the rate clicks become buys.

You want a very, very high conversion rate–ideally 100%, though that's not going to happen.

While interest-based campaigns take out a lot of the work, product-based advertisement offers a lot more in terms of interested users. Here's an example of my active product-based campaign. This shows some of the products I'm specific targeting for my Storm Without End campaign.

Storm Without End Product Image Campaign AMS Amazon

If you click the image, you can get a closer look at the types of products I chose. In short, I picked titles I thought were similar to my own–I also picked similar merchandise. This is taking interest-based advertising to the next level. By picking and choosing products that I feel are in line with what my novel is about, I have a higher chance of gaining interest from shoppers. I want these shoppers seeing my title and becoming familiar with my name as a brand.

It's important to point out at this is less than a full day into the Storm Without End campaign. These numbers are, for all intents and purposes, useless. Once I have viable numbers–and enough of them to be worth looking at–I will continue this author services series discussing the pros and cons of Amazon's AMS system.

For now, I want to bring your attention back to the branding comment I made above.

Branding, or the process of creating security and familiarity in a customer, is critical; as someone sees a product often, a sense of trust is developed–regardless of whether or not they've used/read the product. That's why branding advertising is so important. It brings a sense of trust and familiarity, whether it is deserved or not. It's a human mentality thing–if you see something frequently, it must be good or safe. It's a security in numbers, and unfortunately, a sheeple thing.

In a way, it's sleasy, and a lot of marketers take advantage of that. It's an unfortunate part of the game–a part I don't like. But, it works. If it didn't work, advertisers wouldn't spend so much money on these types of campaigns.

In case you hadn't noticed, I do not have stats for individual products. This is enough to drive me up a freaking wall, three times over. I can't tell if a specific product is performing better than another.

So, like interest-based promotion on AMS, I have to rely on averages to know if this form of campaign is working. That's a terrible way to do it, but that's all I have to work with at this point. Amazon's AMS system has embraced ‘Keep it Simple, Stupid'–which works, to a point. Ultimately, it means that I will have losing products and winning products, but I will target my costs per click to the overall picture.

Easier for authors to manage in the large picture, and it has the benefit of removing the time sink that is adjusting a cost per click campaign per product or keyword.

The math is the exact same for the interest-based campaigns. Ratio of Clicks multiplied by the cost per click is your average investment, with your royalty figures serving as the basis for your profit margin, or return on investment.

That's AMS in a nutshell–a very simplistic nutshell. Can it prove a good tool for authors?

Only time will tell.

But no matter what your decision is, when you're advertising your book, your return on investment is king. Your goal is to always have a positive return on investment. That means you're making money instead of losing money.

With paid advertisements, you are always at risk of losing money. That's the name of the game, so author beware–and be aware. You can save yourself a lot of heartache and headache if you watch your advertising campaigns careful. Always take the time to do the math. If you're bleeding money while advertising, you're not doing it right. Close the wound and find something that works.

At the same time, I think it's important that all authors realize that taking the risk of advertisements might be well worth it. Paid advertising can open doors that organic growth simply can never reach. The trick is knowing when to hold 'em and when to fold 'em.

Leave a Comment:

Brian Meeks says February 12, 2015

I’m not sure I agree with your math. You said you had 1115 people who clicked on your ad. That means for each click you spent fifteen cents.

$167.25 for the ad campaign. You sold 10 books. Your revenue from sales was $21.50. You lost $145.75 on the campaign.

I think the problem lies with the “11.5” you used in your ROI calculation. That would have been correct if you had sold 100 books. As it was, you should have used 115.

115 x .15 = $17.5

$2.15 – $17.5 = -$15.10 per sale.

Now, it may have been that you meant you had 100 sales in which case, your math is correct.

    RJBlain says February 12, 2015

    I fatfingered the extra 1, it should be 115 (like the other examples.) Corrected!

Brian Meeks says February 12, 2015

It should be noted that I made a math error, too.

The 11.5 I converted to 115, but the 11.5 should have been 11.15 and as such my 115 should have been 111.5, this means the loss per book sold was only $14.575.

Brian Meeks says February 12, 2015

The other thing that might be worth looking at is did your book have any additional borrows? If so, then that would need to be factored into your ROI calculation. If you had 100 borrows, too, then the campaign would have been run at a profit.

    RJBlain says February 12, 2015

    These figures would use what amazon is reporting as direct sales as a result of the campaign, not factoring in any ‘unexpected’ or gravy sales due to branding. Since they can’t be proven figures, and these figures are advanced marketing concepts, I didn’t want to throw them into the mix–and I don’t intend to. If I cover advanced marketing concepts later, these residual sales will be addressed. For the purposes of this post, I’m sticking with the basics and quantifiable, proven figures. (All of the numbers used, of course, are theoretical.)

Brian Meeks says February 12, 2015

Thanks for clearing up the mistake.

So, the one area I’m not clear on. The PPC campaign, was that real or theoretical?”

Did you actually run a campaign where you had 115 clicks and 10 sales? If you did, I’m going to give it a try. If it was a hypothetical then I won’t.

    RJBlain says February 12, 2015

    Hypothetical! I *will* be writing a future post with actual campaign figures for both types of campaigns. 🙂

Brian Meeks says February 12, 2015

Thanks. Might I suggest that you put a note in the post about it being hypothetical.

I was ready to drop $1000.00 on a campaign based on your numbers. I would have been miffed to find out they weren’t real.

I’ve not seen any posts of people who ran campaigns at a profit or even close to it, but all of those books were NOT of the same quality as yours. When I read your numbers I thought, “Wow, quality does matter, this makes sense.”

    RJBlain says February 12, 2015

    I have added a note, while I had offhandedly mentioned my current figures (on my campaigns) were useless, it wasn’t up front and center where it should have been–fixed!

    I’m doing two test campaigns, both at $100 each. One is product targeted, one is interest targeted, both in the fantasy genre–I’ll definitely be keeping everyone in the loop on how they do.

    I’ll be really interested to see the impression figures. So far, my counts have been low, but I haven’t figured out a competitive (and viable) CPC bid goal yet.

    And thank you very much for the compliment–so very appreciated!

      Brian Meeks says February 12, 2015

      I eagerly await your results.

      The point where I’ve seen people getting reasonable numbers of impressions has been in the .35 – .50 range. This makes it hard to turn a profit on even a $9.99 title as one would need to get a conversion every 14 – 20 clicks. The conversion rates I’ve seen have been closer to 1 in every 100 clicks.

      Still, there is value in knowing how expensive it is to convert regardless of the profit or loss per book.

      It does give the Traditional publishers a scalable method to drive a book onto the NY Times list, though. The question is, if I have a conversion rate of 1%, and I can get 5,000 sales in a week through a Bookbub ad (on multiple venues) is it worth the $250,000 to take a shot at the other 5,000 sales needed to make the NY Times list?

      I don’t know the answer, nor do I have the money to try. But there is definitely value in making the NY Times list so it’s worth knowing.

      Also, if one is able to increase the conversion rate to 2%, then the cost drops to $125,000.

      I love math. 🙂

        RJBlain says February 12, 2015

        It’s an interesting problem. Should authors go for the higher impressions, hoping to brand, or should they take fewer impressions but get the ROI? I’m going to be approaching it with caution. There’s no way I’m going to be tossing down $0.30 for a click, not without a 50% conversion rate, which frankly put, isn’t happening. 2% conversion rate is pretty damned rare. But, once again, it depends on the books.

        The price will come down on CPC values once people get over how shiny Amazon’s AMS system is… and once the people who have no concept of how marketing really works runs out of money. It’s a pretty classic issue with new markets. People throw down a lot of money out of the gate trying to get an edge instead of playing for the long term game.

        My goal is to find a CPC value where I get a decent ROI—and by decent, I’ll take marginally profitable for the moment.

        And honestly, I’ve gotten 5,000+ sales in a day during a promotion without Bookbub. There are loads of free options and venues. Bookbub’s losing its foothold, I think… because there are so many other equally good solutions that don’t repeat the same basic type of book. Bookbub, unfortunately, doesn’t like leaving their comfortable formulas when it comes to titles they feature.

        If you can get into a bookbub, the boost is great–but I have a suspicion that their 20% acceptance rate is highly exaggerated. I know many, many people who have tried, and maybe 1% of them has actually gotten into bookbub’s system. (None of my books are good enough for it, so say them, but I don’t mind… I put the $ I would have spent on them into doing the AMS testings.)

Brian Meeks says February 12, 2015

When you say 5,000 sales, do you mean paid or are you talking free downloads? (If you don’t mind me asking) Because if you’ve managed to get 5,000 paid sales in a day I’d like to hear more about the process you used. As I said, it only take 10,000 paid sales (across at least two platforms) to sneak into the NY Times list.

I’m running my 8th Bookbub ad on the 23rd (a free promotion). It’s possible to get the ads, but you’re right, it has become more difficult.

As for ROI, if I could run ads at a break even or loss of 10%, I’d pour thousands of dollars into campaigns. I’d be thrilled to spend $10,000 and only get $9,000 in return because of the future throughput.

I love talking data. It makes me happy.

    RJBlain says February 12, 2015

    It was during my free promotion, so while ‘sales’ they were free ones. I think if I got 5,000 paid sales in a day, I’d need a fainting couch.

    I’m going to have to go get a tally of my total paid sales; those are figures I don’t typically poke at, mainly because they can be quite depressing at times!

    Yeah, if I were in a position to take a 10% loss on a campaign for CPC, especially on the first book of a series, I’d probably do it with little hesitation, simply because those investments will balance out over time–and potentially make a huge difference down the road.

    Talking about data is great. There’s always something new to learn about it.

    I’ll be very interested to see how AMS settles once people blow out their budgets on high cost per click campaigns.

    .30 to .50 per click is insane. At current, my cap is around $0.15 per click, and I’m very doubtful that I’ll raise it more than that. I’m more likely to lower it. It’s just not viable to require a 20% click to conversation rate for a campaign, when most conversation rates are below a percent.

      Brian Meeks says February 12, 2015

      I agree. The pay per click business is viable but not with products that have such a low price point. Of course, as you point out, if enough people take their shot and fail miserably, then the number of people bidding for the ads will plummet.

      Once this happens it is possible that a 1 – 2 cent per click bid might be possible. At that rate I think breaking even is within the realm of possibilities.

      The other question is are there areas where few are bidding?

      I have a satire. Perhaps there aren’t as may people going after those readers and so the costs would be lower? I don’t know.

      Ultimately, though, I believe the system will only be used by the Big 5 because they don’t care how much they spend on promoting a book since it doesn’t cost them anything (Their authors are paid on net revenue meaning the money the company throws at advertising is an expense that comes out of the gross sales. In a sense, they are spending the author’s money.)
      As such, they don’t have much incentive to advertise efficiently.

        RJBlain says February 12, 2015

        Even Big Five may not pursue; there are often better options available to them. I will be curious and interested in seeing how it plays out.

        If prices come down to less than five cents a click, I could potentially see this working.

        But the rates you have mentioned, I am astonished anyone would pay that much to advertise their book. It is like a merchant seeking out a robber and handing them their money before being robbed.

          Brian Meeks says February 12, 2015

          I’m not sure the people who chose those rates were especially strong in their math and analytical skills.

          I’m guessing they were imagining a 50% conversion rate.

          Most authors, especially new indie authors, many not have any experience with advertising auctions (or marketing in general).

          I feel fortunate that I got to be an analyst in the GEICO marketing department for four years. The education was as valuable as the salary.

          RJBlain says February 12, 2015

          I was a jack of all trades at an adult entertainment website, which converted to an adult dating website. It was also really, really valuable in terms of figuring out how money works on the internet. My boss was a CPC/adwords whiz, and I got to play with quite a bit of reporting, so I got a small taste of the analyst gig. Very, very educational.

Robert L. Slater says February 14, 2015

Thanks for the conversation sharing. Your questions really helped clarify things. I, too, will be interested to hear about the results.

Add Your Reply