Why you can’t trust ACoS metrics in Amazon Advertising (and two alternatives)

If you use Amazon Advertising (formerly Amazon Marketing Services, or AMS) advertisements to sell media (such as books or video games) or other products (electronics, toys, clothing, office goods, etc.) you are no doubt familiar with the Amazon Advertising dashboard. It presents many metrics, but one of the most interesting is ACoS, or Average Cost of Sales. Here’s an example from my own dashboard, showing the ACoS for Amazon Advertising campaigns for Lean Media and other titles that I publish:

ACoS is supposed to reflect the success of a campaign, showing advertising expenses vs. sales over a given time period. Consider the following campaign:

  • 300 ad clicks
  • 10 sales
  • $20 product
  • Gross sales = $200
  • AMS costs = $90

ACoS is ad spend divided by gross sales, so the ACoS would be 45%. It looks OK when put into a spreadsheet and generate a comparison chart:

The problem with this number, is it reflects Amazon’s gross sales against your expenditures. In other words, it not only includes your revenue, it also includes Amazon’s revenue, which might be 30% or more of the gross!

ACoS vs. ACoN

I created two other metrics to evaluate the success of my campaigns. The first is ACoN, or Average Cost of Net. “Net” is the net revenue from Amazon, or the amount remitted from Amazon to your company’s account for the sales of products.

For example, if you sell something on Amazon for $10 and Amazon takes a 30% cut, your company will net $7 (assuming you don’t have a distributor or some other middleman handling revenue and taking their own cut).

Let’s break down ACoN, using the same example from earlier.

  • Gross sales = $200
  • Amazon’s cut = 33%
  • Company gets = $134
  • AMS costs = $90

ACoN equals ad spend/revenue. This is the result:

So you’re spending $90 to get $134 from Amazon. Not bad, but …

Introducing ACoP

There’s still a problem. What about your company’s costs to produce the goods – COGS, operational expenses, and other costs?

My company produces books and other printed media, so my costs are wrapped up in printing, packaging, design, and royalties. Marketing is also a big cost, but let’s assume that Amazon Advertising costs are separate. ACoN is a useful number, but it fails to account for the company’s costs of production, which may be significant.

For this reason, I use a third metric: ACoP, or Average Cost of Profit. Here’s how it works, using the same example:

  • Revenue = $134
  • Cost of goods sold = $59
  • Profit = $75

ACoP = Ad spend/Profit, so this is the result:

In other words, Amazon Advertising costs are greater than the profit derived from the sales of those 10 items. An ACoP of more than 100% indicates a money-losing Amazon Advertising campaign. This is a markedly different conclusion than what ACoS might show. Based on the high ACoP, it’s time to recalibrate the campaign by lowering costs or using better ad copy, or consider ending the campaign.

A few other notes:

  • Via a comment left by Ricardo from Reedsy (see below) ACoP is less useful for ebook sales.
  • ACoP requires calculating costs. I use an estimate based on production and other costs, which I plug into a Google Sheets spreadsheet to calculate both ACoN and ACoP on a monthly basis. A sample is shown in the video below.
  • For ACoP, accountants might quibble over the inclusion of sunk costs or expenses outside of COGS, while others might want to factor in time or the intangible benefits of brand exposure. You can alter the criteria used to derive profit, or create a new metric, as you please.
  • The ACoN and ACoP metrics can be used for other online ad campaigns (Google AdWords, Facebook ads, etc.) or traditional advertising, but some work will be required to get the right data into a spreadsheet.
  • ACoP and ACoN can also be used as Lean Media feedback to evaluate (and later improve) things like Amazon Advertising ad copy, cover design, or other creative and design elements that are displayed in an AMS ad.

Video presentation – Amazon Advertising: ACoS vs. ACoN (and ACoP):

2 thoughts on “Why you can’t trust ACoS metrics in Amazon Advertising (and two alternatives)”

  1. That’s an excellent analysis, Ian (I was pointed to it by The Hot Sheet). I’ve done a lot of thinking around ACoS and I agree that it’s not a metric to look at directly, but to use in order to calculate a better one.
    However, while I’d understand your approach as a print publisher, I think the ACoP doesn’t make sense for authors or publishers advertising ebooks. There is no variable cost attached to producing and distributing the ebook, so effectively every sale with an ACoN under 100% yields a profit.

    Then, there is also the positive side of things to consider: AMS fail at accounting for your costs, but they also fail at accounting for a lot of sales that can be directly attributed to the ad. They don’t factor in audio sales, for example, and are rumored to miss a lot of print sales. More importantly, authors often fail to incorporate the notion of lifetime value of a customer. Sure, your campaign might have an ACoS of 120%, but if 50% of readers who purchase go on to purchase another item, you’re probably already in the profit zone.

    1. Thanks, Ricardo, for the note about ACoP. I agree, for ebooks it may not be that useful a number.

      In addition to the LVoC concept, there’s also the issue of brand exposure, a side benefit of AMS campaigns that may not lead to sales but may help familiarize customers with your brand. The “impressions” metric may help quantify this, particularly for AMS Headline ads where it’s impossible to miss the ad.

Leave a Comment

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload the CAPTCHA.

Scroll to Top