For each sponsored keyword, Amazon calculates the Ad Rank with this formula:
The probability of click is the Expected CTR %.
The seller chooses the bid per click, while Amazon estimates the probability of that click. We talk of probability because Amazon does not know how many people will click on the ad.
If the campaign has been running for some time, you can calculate the Click-through rate (CTR) which is the clicks over impressions. So, when Amazon estimates the probability of clicks, it is equivalent to estimating how much the CTR will be in the future.
Let’s say you have 3 Amazon Sellers who want to sponsor their ad on the same keyword.
The first bid is 0.76$.
The second bid is 0.75$.
The third bid is 0.60$.
For each of them, Amazon calculates the probability of clicks or expected CTR. In this case, we have found out that for the first two is 2% and for the third one is 5%.
At this point, the ad rank gets calculated for each of them:
Then the products have been ordered from the highest to the lowest Ad Rank. And, from the data calculated we see that the first in the search result is actually the seller that offered 0.60$ because it has a high Click-through rate (CTR).
This is very important because it shows that to obtain a "first" position on the search results page, it’s not only necessary to bid more but it’s a mix between the bid per click and the probability of clicks.
A seller who wants to promote a product that is not very relevant to the final customer will have a low probability of clicks and therefore, to compensate for it, it will have to bid higher. On the contrary, sellers who sell relevant products (e.g., the best sellers in the category) will have a high probability of clicks and will be able to afford to spend less on clicks.
The seller has two levels of action: