5 Myths About the Amazon Algorithm


We’ve found working with so many sellers and brands over the years that there are still a lot of persistent myths about selling on Amazon in general. Most of those are myths about the Amazon algorithm, called A9. Here, we bust five of the most common ones.

MYTH 1:

Nobody knows how the Amazon algorithm works.


TRUTH:

Lots of people do, you just have to look it up.

A9 is a company in Palo Alto that creates product algorithms, code that tells Amazon’s website how to sort and load product lists for each customer’s experience. Anyone who wants to read about how this algorithm works has to do nothing more than search for information online and read the manuals, forums, science articles, and a myriad of other documents that tell you EXACTLY how it works. We’ve got posts here in our own archive about it, on Amazon SEO in general, and on strategic use of keywords and bullet points, plus more tips for ranking on Amazon.

This sort of algorithm is an item to item collaborative algorithm. This means it works on a node system. It’s like a tree of products, or a catalogue, put in order of hierarchy. That means the information Amazon uses to suggest and deliver products to you when you search for them is based on the finite terms used to describe products entered into its catalogue. It bases results on factors pertaining exclusively to the signed-in customer, factors surrounding that customer’s behavior on Amazon and online, and what is popular that day.

Here’s a little illustration of its structure:

Diagram of how the Amazon A9 algorithm works

MYTH 2:

You can figure out keywords that people will use to find you by typing into the search bar and seeing what is autosuggested.

REALITY:

The search bar is personalized to YOU and YOU ALONE.

This is a useless and possibly damaging practice being bandied about by bloggers who think they have uncovered a “secret.” A little knowledge is a dangerous thing. In this case, the lack of knowledge of how an algorithm works is leaving holes in understanding globally on this subject.

Amazon has a department called The Personalization Platform Team. Look ’em up. They spend their time working on coding the search bar to serve absolutely personalized product lists to you. By testing this carefully, it’s easy to see exactly how this works. I spent quite a significant amount of time talking on Moz and other sites to engineers and programmers to understand and clarify this information, some of whom have worked at Amazon.

MYTH 3:

The Algorithm prefers Amazon-branded products and ones from Vendor Central over Third Party Sellers.

REALITY:

The source of a listing does not impact the search results and there’s no distinction between Vendor Central and Seller Central products as long as the listing is meeting customer needs.

Products populate search results, not sellers, so it’s important to take into consideration if there is more than one seller in that product listing – then the most applicable seller will have the opportunity to make the sale.

As a merchant on Amazon, you can have two different roles. One is as a Vendor working with Amazon in a retail partnership. This is when products are sold to Amazon to be commercialized on the platform and, it does all the marketing and fulfillment for you.

Or, you can be a Third-Party Seller on the marketplace, using it as a platform for marketing the product, but handling all other aspects of the transaction independently. There is also a hybrid model to be a seller but benefit from Amazon’s fulfillment solution. For more about the differences between Sellers and Vendors, see this article.

FUN FACT: On Amazon’s marketplace, nearly half of the top sellers are China-direct manufacturers.

MYTH 4:

Amazon’s “consumer-first” focus means that the lowest priced items will surface higher in search results.

REALITY:

The lowest priced item does not correlate to a higher search placement.

A product could be priced the lowest but have poor listing performance metrics which will always hinder search population. Remember that Amazon’s main priority is happy customers, not selling them the lowest priced item that will lead to poor customer satisfaction.

This is not the same, of course, as winning the Buy Box because most consumers are price-conscious, and many won’t take time to read reviews. If you’re otherwise hitting all the other metrics used in the algorithm at a reasonably high mark, you can deploy repricing software to make sure you’re offering the most competitive prices at any given moment in the day.

MYTH 5:

When you’ve got a good keyword for your product, use it frequently, everywhere in your product listing.

REALITY:

Repeating keywords over and overdoes not increase indexing power.

Keyword stuffing is a dated and ineffective strategy on Amazon. Aside from the fact that the algorithm indexes frequency caps, it’s not attractive to the customer– it sounds completely unnatural. This in turn can lead to a decrease in listing conversions and therefore placement on Amazon.

Amazon has shared little specifics on exactly what, when or how content is indexed by the search algorithm, however, based on the data supported through case studies, conversion-focused content optimization should be achieved in the following areas:

  • Product Title
  • Product Bullet Points
  • Product Description
  • Backend Product Listing Keywords

New to Seller Central Amazon Analytics: Amazon recently issued a new release of Amazon Analytics in Seller Central, currently available for third-party sellers who are brand-registered. Previously, only Vendors had access. In addition to search term data, the Analytics dashboard contains valuable data like click data associated to ASINs and search terms.

Shopping Feed syndicates search-optimized product listings on Amazon and all of the world’s most powerful marketplaces, syncs and refines inventory data, perfects product listings, and automates fulfillment workflow.

Contact us for more info: contact@shopping-feed.com  (920) 333 3488
Follow us on Twitter: @shoppingflux