Why calculating TACoS below account level is problematic (and what you can do instead)
Adspert offers TACoS analysis at the account level. This article explains why calculations below the account level (e.g., at the campaign, keyword, or product level) are inaccurate and why Adspert deliberately does not provide them.
TACoS – total Advertising Cost of Sales – is a key metric in Amazon Advertising that considers total revenue: organic revenue plus revenue from paid ads.
The calculation at the account level is simple: TACoS = Ad Spend / Total Revenue (organic + paid)
But what happens if you don’t want to set a single target value for the whole account, but instead want to define individual TACoS goals for certain products, campaigns, or keywords?
The Problem: Attribution – Assigning value to individual keywords
To calculate TACoS for a single keyword or campaign, you need two things:
The ad spend (Amazon provides this directly).
The total revenue influenced by this keyword or campaign – including organic sales.
And this is where it gets tricky. Because organic sales cannot be cleanly attributed to a keyword – especially when:
Multiple keywords advertise the same product
The same keyword is used in multiple campaigns
A product is advertised in several campaigns at once
A product is also sold organically without ads
Advertising for product A triggers sales for product B (“halo sales”)
If you want to calculate TACoS at the product level, a different attribution problem arises: in this case, you know the revenue for the product, but the costs are unclear and must be distributed. Which keyword contributed which share of the sales for a product?
Conclusion: No matter which attribution model you choose – it may be wrong. And the more granular you optimize (e.g., at the keyword level), the greater the inaccuracy.
The solution? A pragmatic one.
Many experts therefore recommend calculating TACoS at a higher level – e.g., at the category level or at least at the parent ASIN level. This doesn’t solve the problem, but it alleviates it and makes your optimization decisions more robust.
Remember: Granularity is important in performance marketing – but it shouldn’t create an illusion of accuracy.
Problems with managing by TACoS
Even if you can (approximately) determine TACoS, you’ll usually want to take action based on it – such as raising or lowering bids. Here’s another problem:
The costs and revenue from paid ads react very quickly to bid changes. If you lower bids, ad costs decrease (so do ACoS and, in the short term, TACoS), but at the same time, visibility, sales, and thus Best Seller Rank (BSR) and potentially the number of new reviews also drop. All this can lead to a long-term decline in organic sales – and thus to a higher TACoS.
In short: An adjustment of bids that lowers TACoS in the short term can actually increase it in the long run. You therefore need to carefully consider what you truly want to achieve and what effects your changes might have.