Note: the following results were obtained using various tests and interviews with Amazon SEO and search engine experts in General. The following statements should not be considered as final determination of Amazon's ranking algorithm, but correspond to the best estimate at present.
This introduction describes the basic functionality of the Amazon ranking algorithm.
This article does not present any specific concrete solution to optimize the rating on Amazon. However, it is highly recommended for sellers who want to optimize their score Amazon to read this article.
The better the Amazon ranking algorithm factors and their interrelationships, the better the factors that develop their ideas and strategies to optimize the ranking of Amazon.
Individual measures specific to Amazon SEO can be obtained with our step by step instructions for optimizing the ranking, which act as a complement to this knowledge base.
The positioning of the product on Amazon resulted in most relevant keywords, search is the most important factor influencing the sale of a product on Amazon.
As stated, most buyers to Amazon use the classic search to find the desired products (and the rest of the users accessing products via external links, lists of best-selling, filters, etc.).
As with all search engines on the Internet, even Amazon is likely to be a site where many users see only the first page of search results, a few clicks up to page 3, and only a very small part click outside the page 3 of the results of ric ERCA.
Unfortunately, there are still some doubts about it because there are still no studies for Amazon. However, if you examine the results of the study on user behavior on Google, the results are clear.
While Amazon show 16 results per page (instead of 10 as in Google), the user will see 48 products in the first 3 pages. Only in very few cases there will be a suitable product among them 48 products.
Although you should not draw any conclusions from his own experience compared to the general public, anyone may still wonder when it acquired modified a product on page 10.
Amazon is always based on the same recipe for success of Google: whoever is on top!
Improvement of the positioning of the best-seller is therefore one of the central factors of success for Amazon and has a big influence on the sales of a product.
Amazon's goal is to get as many as possible to shop at Amazon.
After a user has searched, click on a product of research results and then buy it for Amazon to reach his goal. It is therefore of interest and challenge of Amazon show the user exactly those products in search results that most likely first clicks and then buy.
To meet this challenge Amazon performs two steps.
In the first phase, products that are basically suitable for search are filtered by the selection of all products.
In the second phase, filtered products are sorted and arranged. This provision is determined by the classification algorithm.
Amazon has millions of products.
A user who wants to buy something at Amazon, however, just want to know a very specific part. Through a search request, the user wants to see exactly the products that interest you.
Amazon should interpret this research question and select the subset of products best suited in terms of content to present it to the customer. To do this it uses two criteria for selecting suitable products in terms of content.
First, all the words in the search query must be included in the product.
Amazon takes into account various information fields in the product: title, product characteristics, description, search terms, the seller's brand name and other product information such as color.
If you are looking for a "black leather backpack with buckle in gold": the product must contain the keywords black, leather backpack, rivets and gold (the words fill with and inside are not considered, no distinction is made between uppercase and tiny as well as singular/plural). It is not necessary that all the words are in the same data field. They just need to be available.
To ensure that your products can be included in search results, retailers must therefore provide their products the fullest possible range of relevant search terms. Nor should we forget the particular characteristics of the product.
In addition to the existing keywords you must also define a master image. If this is missing, the product will not appear in search results.
Once the products corresponding to the content were determined, must be ordered and presented in a certain order-a classification.
This classification is used to make things clearer: the user can easily navigate within an ordered list, which is also divided into several pages.
Second, Amazon can influence what products are the fastest and easy to find for sales order.
Amazon determines which product is listed and where via its ranking algorithm. The ranking algorithm takes into account the various properties of a product, the weights and assigns a ranking to the product, which determines its relative position in the list.
As mentioned above, Amazon's objectives are reflected in its ranking algorithm: Amazon wants to generate the greatest possible number of purchases through a ranking compiled optimally.
Through proper placement, customers would have to find the right product in the shortest time possible, click on it and then buy it.
To determine the probability of buying two sets of factors are taken into account.
The first group of factors consists of the performance factors.
This data represents past performance of the product or the product's success or seller and are registered with the rate metric of click through rate (CTR) and conversion rate on sales.
The second group of factors consists of the relevant factors.
Important factors are used to determine how a product corresponds to the contents of the search query. Relevancy is determined based on keywords used.
Amazon wants to show the products of who sold well in the past and are highly relevant to the content of the search query.
Finally, the probability of purchase increases with a high degree of relevance in terms of content and good sales history. The more likely you are to buy, the better the sales on Amazon and the greater the Commission by Amazon itself.
Amazon ranking optimization is to increase the relevance of content using keywords and improve performance with the help of a good product presentation and product attractive conditions.
It is important to note that the increase in relevance is specific keywords, i.e. the relevance should be enhanced on the basis of individual keywords, which will improve the ranking of the product. The increase of performance factors has a positive effect on the classification of all the keywords in the product.
Individual performance factors and significance are discussed in detail below.
Products that in the past have achieved good results were classified as high in the ranking, as they are more likely to purchase.
Amazon measure performance in three key figures: turnover, volume of click-through and conversion rate.
It is obvious that Amazon's sales are used as a performance indicator. More often a product is purchased, the higher your purchase has spread among buyers. Popular products generally have a greater likelihood of purchase, as they are likely to be purchased by others. Therefore, products with high sales are shown in the front of the leaderboard and receive ideally the best-seller ranking for their category.
To understand why the click rate and conversion rate are used as performance factors, the process of buying at Amazon should be monitored.
After performing a search query, the user must first click on a product to go to the product page. There must buy the product with another click. These two phases are measured using the click rate and conversion rate.
The click rate measures the frequency with which a user has clicked on the product when it was displayed in the search results.
If an item is clicked when the user sees it, it means the product is very relevant and attractive. If the product always appears but is never clicked, the product is obviously not interesting. A product that is never clicked can never be bought. The probability of buying this product is 0%. Therefore, the higher the click rate, the higher the Amazon sales ranking, the more likely that products are purchased.
You must specify that a product does not have a click rate universal, but that the click rate varies depending on the keyword and ranking.
A backpack will have a higher click through rate with the keyword "backpack" than with the keyword "smartphone". Similarly, the product at the 1 St place usually has a click rate higher than 35° place. For this reason, the click rate is always determined and compared based on the keyword and ranking.
If the keyword "smartphone" to the 1 St place for Samsung Galaxy S5 has a speed click rate lower than the iPhone 6, then the iPhone 6 will probably be above the Samsung Galaxy S5 in ranking (to find out click-throughs of different products in same position Amazon makes many A/B test).
Amazon ranks either produced in the same place for some requests and then compare click rates. In General, the higher the click through rate of a product in a keyword, the higher the degree of selling Amazon.
The conversion rate measures how many people bought the product after having seen the page.
If a product is purchased when the user arrived on the product page, then the product seems to be very good and exactly what I was looking for.
However, if a product is never purchased after users were on his page, seem to have found information on the product page that prevent them from making a purchase or that they did not expect. This can be, for example, for a bad product evaluation or for additional information in the product description.
The higher the conversion rate, the more likely you are to purchase. Therefore, products with a high conversion rate will have a higher ranking the products with a low conversion rate.
Just like the click rate, it must be highlighted that the conversion rate varies with your keywords. The conversion rate is then used to control the click rate.
If the click rate is very high for a keyword, but the conversion rate is clearly below the average, then the product on the search results page appears to be quite attractive at first, but soon users are on the page of the product, lose interest.
Example: If a product is "iPhone 6", the picture shows an iPhone 6 and the price is €50, many users probably click on the product. If the description on the product page says however that it is just a wooden imitation of iPhone 6, only a few users will buy the product, i.e. the conversion rate will be extremely low.
If the click through rate and conversion rate are not the same, this indicates that the information displayed on the search results page is incomplete and can lead to false expectations. These products will then be classified in order of less importance, since the probability of buying these products is low.
Overall, the performance factors serve to allow Amazon to assess whether a product is purchased. Amazon has no interest to display products that don't sell, because Amazon doesn't earn money from them.
Traders are now faced with the question of how they can influence the performance factors.
The bad news at this point is that you cannot directly modify the click rate, conversion rate and sales.
The good news is that retailers have control over almost all components that make sure that you click and purchase a product.
Changing these components, retailers can then make the user click more frequently on product and buy more often. This increases the click rate, conversion rate and sales. Merchants can then indirectly exercise significant influence on performance factors.
We take first into consideration the possibility that retailers have to influence the click rate. To answer this question, you need to put yourself in the perspective of the user and asking yourself, "what information to use to decide whether to click a product?"
- Tag Bestsellers
- Shipping costs
- Product category
- Other offers
Almost all factors that users can use to decide whether to click on a product are under the control of the dealer.
Only the brand, the best-selling label and product classifications are difficult to influence by the dealer.
Photography is the most important decision criterion.
In addition, information on the left side of the page seem to be relevant, namely: title, brand, price, FBA and availability. The ratings seem to be ignored in the search results page. We do not have accurate information, we do not want to express a definitive statement on the matter.
For retailers, this study demonstrates that the choice of image and right product title greatly affects on the click rate. Therefore, it may be useful for retailers rotate images and do several tests to determine how the selection of the main image affects the click rate or the ranking.
As for the price, it is likely that the users are particularly geared toward relative prices, namely comparing the price with that of the product above or below it to see how it is fixed the price of the product.
Other features like a good evaluation, using FBA, the correct categorization of products and bestselling label can have a positive impact on the click rate.
When the user clicks on a product is brought to the product page. For sellers, the goal is to maximize your conversion rate. Even at this point the question is again: what are the factors that affect the user in deciding whether or not to purchase a product?
Is back to help treat the customer viewpoint when answering this question. Compared to the search results page, the product page contains additional images, bullet point, product information, descriptions and ratings as new information that the retailer can influence.
There are also some fields on which the retailer has no influence (for example, "customers who bought this item also bought …").
The arrangement of information fields can be used to derive some prioritization of information fields.
New pictures and bullet points appear in a very visible to the user. To access the product information and description, the user has to scroll and skip some information fields. The ratings are at the bottom of the page.
According to the arrangement of images and bullet point, one can deduce that they have an important role in the decision to purchase.
If the user likes what he sees, most likely will buy it immediately. If you have not yet 100% convinced, will scroll down to get more information. If he doesn't like the pictures or bullet point, probably will leave the site and searches for other products in the search results page.
The pictures and bullet points are crucial and must be given adequate attention by the dealer.
For customers not yet decided, retailers must enter details in the product description.
In the reviews of the products, retailers can get positive feedback from users by leaving a comment, especially in case of negative reviews. This enables retailers to prove to take seriously customer problems and to explain what measures they have adopted to prevent this from happening again.
This also provides vendors with the opportunity to use existing information fields in order to exert a positive influence on the conversion rate. The information fields have some formal aspects that merchants should follow (for example, ideally five Bullet points and images with at least 1,000 x 1,000 pixels to enable zoom function).
However, these are mainly those aspects relating to the content, namely what information should be included in which informational field. Since neither we nor the retailer we know exactly what will cause the customer to purchase, we recommend that you test various changes and check the effect on the conversion rate.
Adding a single note to bullet point may therefore significantly increase your conversion rate.
For example, it should be noted that the expensive shoes are hand-made in fine leather, this could prove to be the last element that pushes some customers to make the purchase.
Performance factors have a great influence on the leader board. As is clear from the explanations provided, sellers can exert considerable influence on relevant performance factors indirectly through the presentation and conditions of the product.
In our instructions for optimization of the ranking we present the formal requirements for individual information fields. The content must be determined by the seller. You should make some changes in testing and to test the effect of the changes on the leader board in the following weeks.
The second important part of A9, Amazon ranking algorithm, is formed by factors of relevance.
While the performance factors measure the sales of a product, the factors of relevance measure the level of correspondence between a product and the contents of the search query.
The relevance is mainly determined by the keywords.
This is good news for sellers, who have full control over the keywords and can therefore directly influence the relevancy factor.
As mentioned earlier, Amazon filters all products that do not contain all of the words in the search query.
Amazon will consider the contents of the following fields: title, bullet point, product information, description, search terms (on the seller), logo and salesperson.
For the ranking, however, is important not only that the words of the search query are all available, but also where and in what order they are present.
Since each information field is weighted differently for the ranking, it makes a difference in what field is the key word. The keywords in the title of the product they promote the ranking much more, followed by bullet points, by search terms, product specifications and description of the product.
If the search query words are in your title, the product has a better chance of getting a good ranking than when all search terms are only in the product description.
The ranking is also included the degree of correspondence between the search query and keywords of the product. Ideally, all search terms are in the same order in an informational field. If the individual search terms are distributed on different fields, relevance decreases.
If you are looking for "black leather bag", a product that has the title "black leather bag" has greater relevance than a product with the title "black leather bag". A product that has "black" in bullet point and "leather handbag" in the description will have little relevance.
Finally, it is important to remember that shorter are the title and bullet point, more rilevantie will be. If a product is only the word "ball" and another product has the title "nice rubber ball which can also be used as a fitness ball for yoga", then the first product will have a greater importance than the latter product when trying "fitness ball".
The optimization of the ranking, the operators must therefore face the challenge of selecting the correct keywords first and then decide which information field and in what order are inserted keywords.
Detailed advice on keyword strategy are available between our services.
The factors of relevance and performance factors are described, together, determine the classification of a product in search result through the Amazon. It should be pointed out that the two factors do not affect the ranking independently of one another, but partly affect each other.
In some cases, the two factors act in the same direction.
Keywords well selected and positioned not only increase the relevance of a product, but also its rate of click-rate, because users tend to click more often on products that contain their keywords in the title of the product.
In some cases, performance and relevance factors affect in different directions.
From the point of view of the keywords, it would be useful to include many keywords in the title or in the bullet point of the product.
From a performance standpoint, this is probably not the best approach, because a long title and complicated loses its meaning and is therefore less clicked.
The interaction of the individual factors must always be taken into account during optimization, so that an excessive optimization on one side does not trigger a counter-negative reaction on the other.
The classification of a product based on keywords is a central factor of success for selling a product on Amazon.
In a search query, the results are determined in two stages.
In the first phase, Amazon will select products among all products on Amazon itself that have a reference to your search query. In the second phase, the selected products are sorted based on the probability of buying and displayed to the user.
Amazon uses performance factors and relevance to determine the probability of purchase.
Sellers can influence these factors directly or indirectly by modifying the presentation of the product, the product condition and keywords.