What is Amazon A9 Algorithm?
Amazon A9 is the name of Amazon's search engine. The department that builds, maintains, and improves the search engine is known as the A9 department. Officially according to the wiki, they are a company that provides "search engine and search advertising technology". It is called A9 as a play on the word "algorithm" because there are 9 letters after the letter A.
Whatever you call this company, the software development behind it is pretty sharp. With an experienced team, A9 is definitely where Amazon will align with Argo’s needs in the near future.
Deciding what Amazon A9 is and determining how sellers and vendors can benefit from it are two very different things. In this post, we’ll outline how Amazon A9 relates to Google, where it might be going, and most importantly what you need to do to help you sell more on Amazon.
Amazon A9 vs Google
Amazon is a long way from where Google is today. Looking at the relevance of each product search, we can clearly see the difference. The problem with this is that Amazon doesn't really care that much. Their search engine is product based. This means that the products shown are done based on relevance + popularity metrics, very similar to what Google uses to run its products. Anyone remember PageRank?
Earlier this month I did a guest post about how Amazon was exactly like Google back in 2010 (pre-Panda & Penguin updates). In this post, I discuss in detail why Amazon’s A9 algorithm looks exactly like Google’s did a few years ago. – So what does this have to do with selling more stuff on Amazon?
Where is the A9 algorithm?
In my opinion, there are two core choices Amazon can take, and honestly, I don't know which one they will decide on, only if they do.
With Google they are usually pretty clear on what they are going to do (usually it's the exact opposite of what they say it is), but with Amazon's A9 it's a little more complicated but it can only be 1 of 2 clear paths for a number of reasons.
1.) “Enough for now”
Investment in A9 was muted. While investing in search engine improvements will never officially “stop,” Amazon’s first core choice is that they are considering investing in other approaches. Amazon's primary goal in purchasing Whole Foods last year was likely to address local market issues. Meaning a search engine is not that important. They are also looking for AI solutions, even though Google has officially stated that they will not be in their algorithms for the next decade or so.
How long will “now” last?
Why do they say that? Why would they leave it at "good enough for now" and how long will this last?
All super important issues from Amazon’s point of view, but unlike Google, Amazon doesn’t make 95% of its profits from advertising, it’s because of search engine relevance. Amazon needs to focus on some tech-business elements, even from a warehouse-lightweight perspective, as they seek to improve why people buy a product. These investments will be prioritized because they are generally considered to have a better return on investment and therefore receive priority.
I can personally see that if Amazon were to go down the "good enough for now" route, they would wait for a better search technology solution and buy the company that creates it. This is very similar to what they did in 2009 when they purchased Snaptell. ) It makes sense, and now it seems to be Amazon’s approach too!
2.) “Massive improvements and investment in A9”
Of course, other core options are massive investments in technology and A9. This will solve some of Amazon’s problems with low-quality listings and even low-quality sellers from abroad that use dropshipping techniques, leading to a host of VAT and tax issues that Amazon is currently dealing with. This is actually one of the reasons why UK sellers (myself included) started to get frustrated with the Amazon platform. But, of course, Amazon is now too big to fail.
Obviously the investment will be high. But the results could be extremely valuable to Amazon. Leading to (of course) higher average customer sales. This is a common goal at Amazon, to increase the amount of money people spend on our site…. It all comes down to.
Artificial Intelligence: There is a lot of debate right now surrounding AI and how Amazon fits into this. (Although Google has officially stated that they do not use AI in their algorithms, and they seem to be OK with that…) Amazon has made their own statement and said that they want to integrate more and more AI into the platform to improve the overall efficiency of the site. This of course gives you a sense of the astronomical cost. But Amazon is Amazon for a reason.
There are several core areas that Amazon optimization experts know can be improved with simple algorithm changes and updates.
Simply put, A9 has three main functions: product search; visual search; advertising. I plan to make this topic into a series, and let’s put the other two aside for now. Let’s talk about “product search” today.
From A9's perspective, what is the process for a keyword, from being filled in, to being included, to finally being on the homepage? What can we think of from these processes, and how to optimize them, are probably topics that everyone is concerned about.
Based on the above picture and our team's exploration and attempts, the A9 algorithm's functions based on product search mainly include the following four steps:
Step 1: You need to tell Amazon what your product is. How do you explain this? That is to say, if we are selling Apple products, you must write the keyword "APPLE" where it can be seen by Amazon A9. Where are these visible places? That is the officially specified positions including the product title, five-point description, product description, background keywords, attribute words, product brand, etc.
Step 2: It is not enough to just put the keywords where consumers can see them. Even if you put it up, Amazon may not be able to see it. This is the second step of inclusion, also known as Indexing. Friends who have done SEO know that indexing is of utmost importance to any search engine. If the spiders cannot crawl effectively, how can there be any ranking?
Step 3: After the inclusion is completed, you are on the list. But the problem arises again. Although you are already on the roster in the APPLE class, there are so many Apple products. When customers search for "APPLE", how many places will you rank? This is determined by the conversion rate and effective click-through rate of your keyword. So in this step, the search engine needs to perform ranking based on the search weight, that is, sorting and ranking.
Step 4: This step actually occurs almost simultaneously with the above three steps. Amazon and Google are still a little different. Amazon has a category division, which is the category node that everyone is more familiar with. For details, please refer to the browse tree guide or product classifier in the background.
A9 Through categories, all products and keywords can be effectively divided like a city wall. This is why sometimes you don’t put the categories correctly or they are not accurate. The reason why the front words cannot be searched and the advertisements cannot be displayed is also a huge difference between shopping search engines and Google-type consulting engines. Amazon's category function has powerful card control and division.
