We often hear from sellers who advertise on Amazon that they judge the success of their campaigns using a metric called advertising cost of sales (ACoS). It’s easy to understand why: ACoS represents ad spend as a percentage of sales. For example, if you spend $5 on ads and generate $25 in sales, your ACoS is 20%–a straightforward measure of your advertising’s profitability. The lower the ACoS, the cheaper it was to drive a sale; the higher the ACoS, the more expensive it was to drive a sale.
While useful, ACoS should be viewed in context. It is a clear signal of how efficiently your campaigns generated sales within a specified time period, and it can help you determine where to stop, start, or continue spending. But used in isolation, ACoS can be a misleading substitute for understanding the overall impact of your advertising.
Here are four reasons why it is worth considering a more holistic approach.
1. ACoS doesn’t measure the entire impact of advertising.
Advertising drives results that matter for your business but aren’t captured by ACoS. The advertising cost of sales metric helps you quickly assess how cost-effective your ad campaigns are in generating immediate sales. But immediate sales and high ROI are not the only indicators of success.
A campaign with higher ACoS, for example, may potentially be better at generating new customers, subscriptions, or detail page views than a campaign with lower ACoS. These outcomes would be a boon for your business and could ultimately help yield more long-term sales. While ACoS is a valuable metric, using it as the only proxy for success will cause you to miss the bigger picture.
Recommendation: Establish clear goals and key performance indicators (KPIs) for different kinds of campaigns, and track results independently against those goals. For example, are you launching a new product and need traction? Consider focusing on overall sales instead. Are you trying to increase customer loyalty? Look at subscriptions. Does your product page have a high conversion rate but not receive enough traffic? Consider using clicks or detail page views as your KPIs.
2. Comparing performance by ACoS alone ignores meaningful campaign differences.
You can’t evaluate ACoS in a vacuum without considering the context of factors that influence it. Comparing ACoS across campaigns assumes that all else is equal; but the reality is often more complex, and there can be a great deal of variance. Particulars such as the advertising goal (customer acquisition vs. retention), product life-cycle stage (new launch vs. growth), keyword competitiveness, product price, and even ad creative can vary between campaigns and impact ACoS differently. Furthermore, the appropriate trade-off between ACoS and sales will change depending on your campaign goals.
For example, ACoS tends to be lower for remarketing campaigns, your best-selling products, or branded keywords, but higher for prospecting campaigns, new products, or highly competitive keywords1. Concluding that one campaign performed better than another simply because it had a lower ACoS ignores the strategic and tactical differences that may have directly impacted ACoS in the first place.
Recommendation: Let your goals help determine when to compromise on ACoS. If you’re setting up awareness, prospecting, or unbranded search campaigns to support more aggressive growth—such as efforts to acquire new customers, launch a new product, or gain market share—consider accepting short-term sacrifices on your advertising cost of sales. New customers, for instance, might have a longer decision path or cost more money to acquire than existing customers (or customers already familiar with your brand). Although these types of campaigns may have higher ACoS, they will likely be better at helping you increase sales year over year.
3. Higher ACoS campaigns can pave the way for lower-acos campaigns (with better results).
Different campaigns can complement each other to deliver better results, and activity from one campaign can affect the ACoS of a different campaign. For example, consider the following scenarios—in each, your brand has two separate campaigns running at the same time:
In scenario 1, you have:
- A display ads campaign that generates few sales but drives new branded keyword searches (high ACoS)
- A sponsored ads campaign (sometimes referred to as search) that targets those keywords and, in turn, succeeds in driving sales on Amazon (low ACoS)
In scenario 2, you have:
- A sponsored ads campaign that drives few sales on Amazon but a high volume of detail page visits (high ACoS)
- A display ads remarketing campaign that successfully brings those customers back to complete a purchase (low ACoS)
Brands such as Seiko, Rhodius, and WaterWipes, for example, have seen positive results in conjunction with similar strategies. After complementing their sponsored ads campaigns with display and broadening their advertising goals beyond ACoS, each saw more than double the volume of earned media and branded keyword searches during the campaign period2–in two cases even lowering their overall ACoS.
Recommendation: Use sponsored ads and display ads in parallel to reach shoppers at different stages of the customer journey. Although different tactics may result in higher or lower ACoS for given campaigns, together they will expand the total advertising levers available to you (such as keywords and audiences) as well as the range of shopping experiences you can address on and off Amazon. Check out this post for more detail about how sponsored ads and display ads work together.
4. ACoS is subject to diminishing returns.
Even if ACoS remains your primary optimization lever, it is misleading to compare the ACoS of new campaigns to the ACoS of existing campaigns. The law of diminishing returns says that as you put more dollars into an investment, the returns will eventually decrease. But this does not necessarily mean that you should stop investing once you reach that point. If your return is still positive and higher than what you could achieve elsewhere (your opportunity cost), it makes sense to keep investing.
In the case of advertising, this means you should not necessarily stop spending if you notice your ACoS dips below a previous benchmark. The question is not What has your ACoS been up until now? but rather What is the best return you can achieve with one more dollar? For example, if you put your entire budget into a sponsored ads campaign, you may eventually hit the ceiling of the audience searching for your products. At that point prospecting—that is, reaching audiences who are not searching–may be a more cost-effective tactic.
Ultimately, ACoS is just one tool to help guide your decision-making. Advertising is a holistic effort that should seek to reach audiences at multiple touchpoints, and using complementary strategies through multiple channels can help achieve a range of goals, short-term and long-term. By looking at ACoS alongside other metrics that support the long-term growth of your business, you can achieve a more complete perspective on the success of your campaigns.
Contact us today to learn more about our measurement solutions and how they can help you understand and optimize your ad spend.
- Remarketing campaigns and branded keyword campaigns tend to have lower ACoS since customers have already expressed interest in your product or brand; best-selling products tend to have lower ACoS because they benefit from past success and market validation; prospecting campaigns and new products tend to have higher ACoS because it takes more investment to gain traction; highly competitive keywords tend to have higher ACoS because increased demand means the cost of advertising is comparatively higher.
- Note there are range of inputs that impact changes in branded search and earned media/organic exposure on Amazon. Branded Search: % change in indexed keyword searches of advertised brand or products during the campaign period (vs. pre-campaign period).
Earned Media (organic retail impressions): % increase in organic impressions for the brand products during the campaign period (vs. pre-campaign period).
Source: Amazon Internal Data