If you want success on Amazon, you need to understand how Amazon’s Search Algorithm works – right?
Sounds obvious, but you’d be surprised…
Most sellers have no idea how Amazon ranks and delivers search results; let alone how (easily) exploitable it can be!
Are you ready for a shocking fact?
THREE TIMES as many buyers search for products to buy on Amazon, rather than Google.
Think about it…
Where do you go when you need to know if a product is worth buying?
What about when you want the best deal on anything from a book to a refrigerator?
Yet, you probably don’t pay attention to it’s search engine – much less consider it as a marketing channel worth optimizing for. Even most ‘Amazon Marketers’ are still spending their days trying to optimize their Amazon Listings for Google…
But, what if you knew how to rank in Amazon instead?
You’d have THREE TIMES more ready-to-buy customers than you’d EVER get in Google – and you’d do it in a fraction of the time!
You’re about to read The Ultimate Guide to Ranking Your Products on Amazon…
But before we get into the meat of the matter, here are some basics you should know…
Introducing A9: Amazon’s Product Search Algorithm
A9 is the name of Amazon’s product search algorithm. Since this is a guide about ranking products in Amazon, it makes sense to start at the source. So, this is A9’s official statement for how they calculate search results.
[quote name=”Amazon”]Our work starts long before a customer types a query. We’ve been analyzing data, observing past traffic patterns, and indexing the text describing every product in our catalog before the customer has even decided to search.[/quote]
As we can see here, much of the work is done before the customer even touches the keyboard. Once the customer actually hits “Enter” to perform a search, the A9 algorithm delivers results through a two-step process:
[quote name=”Amazon”]Once we determine which items are good matches to the customer’s query, our ranking algorithms score them to present the most relevant results to the user.[/quote]
It’s a pretty simple process at its core:
- First, they pull the relevant results from their massive “catalog” of product listings.
- Then, they sort those results into an order that is “most relevant” to the user.
Now, some of you SEOs out there might be thinking, “Wait a second… Isn’t relevancy Google’s turf? I thought Amazon only cared about conversions! What’s all this focus on relevance doing here?”
The answer is simple: Relevance doesn’t mean the same thing to Amazon that it does to Google. Read this statement from A9 carefully to see if you can catch the difference:
[quote name=”Amazon”]One of A9’s tenets is that relevance is in the eye of the customer and we strive to get the best results for our users. […] We continuously evaluate [our algorithms] using human judgments, programmatic analysis, key business metrics and performance metrics.[/quote]
- Google says, “What results most accurately answer the searcher’s query?”
- Amazon says, “What products is the searcher most likely to buy?”
The difference between those two questions is the difference between how Amazon measures relevancy compared to Google.
On the whole, ranking in Amazon is more straightforward than Google because you’re essentially cutting the work in half. This is because there’s no such thing as off-page SEO for Amazon; they only use internal factors to determine how a product ranks. Backlinks, social media, domain authority… These are all things you don’t need to worry about on Amazon.
That being said, there are a few simple rules you must always remember about Amazon. These 3 rules are critically important to making the most of this guide, so make sure you read them twice:
- Amazon’s top goal in everything they do is always maximize Revenue Per Customer (RPC)
- Amazon tracks every action that a customer takes on Amazon, right down to where their mouse hovers on the page
- The A9 algorithm exists to connect the data tracked in #2 to the goal stated in #1
So far, so good?
Core Pillars of the A9 Algorithm
From A9’s website and from the information that Amazon makes available to us through their Seller Central (login required), we can group Amazon’s ranking factors into three equally important categories:
Conversion Rate* – These are factors that Amazon has found have a statistically relevant effect on conversion rates. Examples of conversion rate factors include customer reviews, quality of images and pricing.
Relevancy – Remember the first step in the A9 algorithm? They gather the results, and then they decide how to list them. Relevancy factors tell A9 when to consider your product page for a given search term. Relevancy factors include your title and product description.
Customer Satisfaction & Retention – How do you make the most money from a single customer? Make them so happy that they keep coming back. Amazon knows that the secret to max RPC lies in customer retention. It’s a lot harder to get someone to spend $100 once than $10 ten times. Customer Retention factors include seller feedback and Order Defect Rate.
*Note that Amazon uses both predicted and real conversion rates for product rankings. For example, if your product is priced higher than other similar products, Amazon will predict a lower conversion rate for your listing and use that rate until real data corrects it.
Okay! We’re finally ready to start talking about how to rank product listings in Amazon. What you’ll find below are 25 Amazon ranking factors that either Amazon themselves or independent marketers have confirmed the A9 algorithm to use.
Top 25 Amazon Ranking Factors
Amazon isn’t like Google where they go to great lengths to hide the factors that they use in their algorithm. Inside Amazon’s Seller Central, they’ll blatantly tell you several of their top ranking factors. You can also visit the official Amazon Seller Support Blog for some great insights. And here’s the UK Seller Support Blog if you’re interested.
Conversion Rate Factors
After just a couple searches on Amazon, it should be pretty obvious that number of sales compared to other similar products – otherwise known as Sales Rank – is one of the most important rankings factors.
Even now Amazon is testing a new feature in their search results where they automatically append a #1 Best-Seller banner (see below) to the best-selling product in category-specific searches, like this one for “Strollers”:
It’s simple really…
More sales mean higher rankings – and higher rankings mean more sales!
It sounds like a vicious cycle, but luckily there are still many ways for new sellers to compete.
It probably doesn’t need to be said that the number and positive-ness of your customer reviews is one of the most important ranking factors in Amazon’s A9 algorithm.
This example product search for the keyword “vacuum” illustrates some interesting points about how Amazon weights review volume vs. review quality:
Let’s dissect this search results page:
- The BISSEL vacuum (green) has the most reviews AND the highest review rating. It’s also the best-seller in its category, so it ranks at the top.
- The second-ranked Dirt Devil (red) has more customer reviews, but a lower review rating. It’s also a best-seller, so it ranks second.
- The third-ranked Shark Navigator (blue) has less customer reviews, but a higher rating than #2, and it’s also a best-seller, so it ranks #3.
- The Hoover WindTunnel at #4 has substantially more customer reviews than any of the top three listings, but it’s not as highly rated as #1 and #3, and it’s not a best-seller, so it ranks #4.
This is one of those metrics that Amazon doesn’t specifically state they track. But, it’s data they have access to and Q&A’s are listed close to the top of the product page, which typically means it’s important for conversions.
Furthermore, there products like this (me-approved) Philips Sonicare electric toothbrush, which ranks #1 for the keyword “electric toothbrush” over other equally rated best-sellers because it has almost twice as many customer Q&As than any other listing in the category:
Image Size & Quality
Amazon continues to tighten their image size and quality policies for product listings. Right now, some categories won’t even display results that don’t have at least one image that is 1000×1000 pixels or larger. These are called “suppressed listings”.
The 1000×1000 pixel image size allows Amazon to offer customers their Hover-to-Zoom feature, shown below, which they’ve found has a dramatic effect on conversion rates.
Awful artistry aside, you can see that as my cursor hovers over the image, Amazon automatically displays a zoomed-in version in the product information pane.
Notice that image quantity is not what’s important here. This Tippmann paintball gun is the #1 product for the keyword “paintball guns”, but it only has one image. Since the image is big enough and informative enough to give the customer all the info they need, that’s all it takes to make Amazon happy.
That means it’s better to have one large, high quality image than to have multiple normal-sized images. Not to say that multiple images won’t convert better than one image, just that the benefits quickly taper off after the first.
Remember earlier when we talked about how Amazon’s A9 product search algorithm uses both predicted and real conversion rates to determine which products to show in their search results?
One of the biggest factors Amazon uses to determine predicted conversion rate is pricing – they know that customers tend to seek the best deals. More importantly, Amazon uses pricing as a major factor in picking which product to show in the buy box, which is the part of the page containing the Add to Cart button (we’ll talk more about that later).
Notice here that the top-ranking product for the search term “juicer” has less customer reviews, lower customer reviews and lower Sales Rank than every other listing in the top 4. It still shows #1 because it’s got decent ratings and is priced waaaaay below the category average.
Note that customer reviews are still vital here. And pricing isn’t the only reason that the Black & Decker Juicer ranks #1…
Many sellers create multiple listings for variations of the same product. This is suboptimal. It’s much better to use Amazon’s built-in parent-child product functionality to direct all customers to a single product page.
This has several benefits:
- It maximizes your customer reviews, since Amazon will combine your similar products into a single primary product page
- It makes the most sense from a UX standpoint; keeping customers on the same page makes it more likely they’ll buy your product
- Amazon has shown a preference for ranking products with multiple options in their listing
Let’s look at that top-ranking Black & Decker Juicer again:
If you scroll back up the page, you’ll see that this juicer is the only one in the top 4 results to utilize parent-child product connections. When you enable the parent-child relationship, it shows as an extra option in Amazon’s search results…
This not only increases click-through rates, we can see here that it also helps you rank above the competition!
Time on Page & Bounce Rate
Remember, Amazon can measure every way a customer interacts with their website, so it’s easy for them to track detailed time on page and bounce rate stats.
Here’s exactly what these similar-but-different metrics mean on Amazon:
Time on Page: Amazon believes that the amount of time a customer spends on your listing page is a good measure of how interested they are in your product. A customer who reads your full product description, looks through reviews and investigates the Q&A’s is much more likely to buy than the one that spends a couple seconds skimming the features.
Bounce Rate: A “bounce” is when a customer performs a search, visits your page, and then either goes back to the search results or clicks on a Related Product offer. Keep in mind that Amazon has a much more exact measurement of bounce rate than Google, again, because all user activity happens within their platform.
Product Listing Completeness
Finally, the last conversion metric to optimize for is listing completeness. The individual sections of the product listing mostly have to do with relevancy, as you’ll learn below, but the actual completeness of the listing has an effect on conversion rate.
As a general rule, the more complete you make your listing, the better. Do your best to fill in every single field in the listing setup page to maximize your chances of appearing at the top of product search results.
Optimizing your product title for Amazon is an excellent example of the way that optimizing for Amazon differs from optimizing for Google.
In Google, you want a concise, engaging title with your keyword close to the beginning.
In Amazon, all you care about is keywords. You want to cram as many keywords into about 80 characters as you possibly can.
In fact, you can actually go beyond 80 characters if you want, and it’s better to have too many keywords than too few. I’ve seen top-listed products with titles that make no sense and have over 200 characters, like this top-rated “Nexus charger”:
It should be noted that Amazon is starting to crack down and standardize Product Titles – keep an eye out for this moving forward…
Features / Bullet Points
The other big reason that particular Nexus charger ranks so highly is because it has lots of keyword rich, informative features. Features, which are displayed as bullet points right below the pricing and product options, are an absolute must.
Just like with images, Features are so important that Amazon no longer allows products without bullet points to be featured in the buy box, and not having them is a serious road-block to good Amazon rankings.
Another good example of proper Feature usage is this Asus computer monitor, which ranks #1 for “computer screen”:
Notice how the bullet-points are both extremely detailed and include a ton of keywords? At the same time, they’re easily readable, which means they won’t confuse customers and risk hurting conversions.
Your product description is basically where you expand on your Features. It’s also the part of the page you have the most control over. If there’s anywhereto really put a lot of effort into engagement, it’s in the product description.
That being said, keep in mind that unlike with Google there is no benefit to having a keyword appear multiple times on the product page; if it’s anywhere in your product listing at least once, you will be relevant to rank for it.
If you want to see a truly appetizing product description, check out the one for this DeLhongi Espresso Maker – the #1 ranked listing for the term “espresso maker”.
There’s nothing advanced about this product listing – they just covered all the bases. It’s thorough, inviting, easy to skim, includes plenty of images, captions, and they even included extra tech. specs that aren’t listed in the normal Specifications section (which we’ll talk more about below).
Brand & Manufacturer Part #
Remember earlier when we looked at the top results for the keyword “Juicer”? You can refresh your memory below:
Something that every single one of the top listings do right in that category is list the brand and manufacturer number first in the product title. In fact, if you do the search yourself it’s not until the 15th result that Amazon shows us a product listing without the brand and manufacturer number included in the title.
You always, always, always want to include a brand in your title because it enables your product for search filters AND allows you to capture customers searching for a specific brand. And if you’re in a niche where customers are using the manufacturer number to search for products, you definitely want to include that keyword in your title.
These are different than Features – this is the part of the page where you actually list the technical and physical details of your product. This includes size, shipping weight, color, publication date (if you’re doing books), tech. specs and more. You can see this top-ranked product for the “home theater system” search term using their product specifications to the max:
Category & Sub-Category
You probably didn’t realize this, but once a customer has entered into a category – every other search they perform on Amazon will, by default, be limited to that category.
Take a look at the example below:
You can see here that a simple search for “dog food” actually takes us three categories deep into Amazon’s product catalog, indicated by the red lines in the image above. The blue box shows that we’ll stay in the Dog Food category until we either return to the home page or manually tell Amazon to show us All Departments.
When setting up your product listing, make sure you put your product in the most relevant, narrow category possible.
In addition to categories, you can also specify search terms that you want associated with your product.
Even though Amazon lists five different 50-character search term fields, you’re better off thinking about it as one big 250 character text box in which you can enter every possible search term you can think of for your product.
This is somewhat complicated to explain, and I can’t do a better job than Nathan Grimm has already done over at Moz (it’s about 1/3 of the way through this article), so just head over there if you want to learn more about this specific factor.
This is one of the biggest hidden ways that Amazon determines a listing’s relevance to a given product search. This is also yet another example of how Amazon tracks every single minutia of a customer’s activity on their website. Take a look at this URL that links to a listing for a Black & Decker electric drill, and see if you can tell me what search term I used to find it:
You can see the source keyword right at the end of the URL – &keywords=electric+drill – that tells Amazon that the source keyword was “electric drill”.
Therefore, if I were to buy this drill, Amazon would know that this listing is highly relevant for the term “electric drill”. The next time a customer searches for that term, this listing would be more likely to show at the top.
Here’s a neat little Amazon ranking hack you can do to take advantage of this factor:
- Construct a URL for your product listing using the [&keyword=your+keyword] query (append the code inside the brackets to your product URL).
- Use a link shortening service like bit.ly to create a shareable link to that URL.
- Drive traffic to the shortened link.
Now anytime you make a sale from one of these shortened keyword links, you’re basically tricking Amazon into thinking that these visitors performed a product search for your target keyword.
Customer Satisfaction & Retention Factors
Negative Seller Feedback
Why do I list negative seller feedback specifically, as opposed to just seller feedback in general?
Interestingly, Amazon actually claims not to track positive seller feedback; at least, not for the sake of their product search algorithm.
Instead they track negative seller feedback rates, or frequency. It doesn’t matter how bad the feedback is – all negative feedback is the same, and it all counts against you equally in terms of search result rankings.
To be clear – as a third-party seller attempting to win the buy box (shown below) you want your seller feedback as high as possible. However, negative feedback rate is the only metric with a known effect on product search results.
Order Processing Speed
Amazon knows that one of the best ways to make customers happy is with fast and accurate shipping. Therefore, a vendor or seller who has shown consistent and efficient order processing is more likely to rank higher than a vendor who’s had complaints of inaccurate or slow shipping.
Customers hate it when they want a product but can’t have it. One of the most common ways this problem occurs is when an item is out of stock, or when a seller doesn’t keep proper track of their inventory.
Whether you’re a first-party vendor or a third-party seller, keeping up your inventory is vital to maintain top rankings, both in A9’s product search results and in your product’s buy box.
Two of the big customer satisfaction metrics are Percentage of Orders Refunded and Pre-Fulfillment Cancellation. In both cases, Amazon has found that vendors/sellers with low in-stock rates tend have higher refunds and cancellations, which of course is bad for customer retention.
Perfect Order Percentage (POP)
POP is a measurement of how many orders go perfectly smoothly from the time that a customer clicks “Add to Cart” to the product arriving at their home.
If you have a high Perfect Order Percentage, that means you have a high in-stock rate, accurate product listings and prompt shipping. That’s exactly what Amazon wants for each and every one of their customers, so they’ll naturally rank high-POP sellers above lower-POP ones.
Order Defect Rate (ODR)
ODR is basically the opposite metric of POP.
Every time a customer makes a claim with an order, that’s considered an order defect. Here are some of the most ways an order can defect:
- Negative buyer feedback
- A-to-Z Guarantee claim
- Any kind of shipment problem
- Credit card chargeback
Each of those examples by itself would count towards your Order Defect Rate, which is the number of order defects compared to the total number of orders fulfilled over a given period of time. Amazon says that all sellers should aim for an ODR under 1%.
Important! Buyer-removed negative feedback does not count towards your ODR. So, it really pays to address each and every one of your customers’ issues.
How often does a customer view your listing and then exit Amazon.com? That’s your exit rate.
If your page has an above average exit rate, Amazon takes that as a sign that you have a low-quality listing. Usually a high exit rate is because your product has a low in-stock rate, or because your listing isn’t fully complete.
This is a metric that I didn’t used to think Amazon measured, but recently I’ve been seeing stuff like this in product search results:
Clearly packaging options are something that Amazon has found their customers care about. But, even if it weren’t, it’s a great way to separate your listing from other similar products (and rank higher through an increased conversion rate).
An easy way to do this – seen in the example above – is to use Fulfillment by Amazon to offer Frustration Free Packaging. This is where Amazon uses less packaging and fully recyclable materials without sacrificing product protection. You can read more about it here.
Too Long; Didn’t Read
This post is getting dangerously close to 4,000 words, so I know that a lot of you probably won’t read it all.
That’s okay – I’ll forgive you with time 🙂
Key takeaways from The Ultimate Guide to Ranking Products on Amazon:
- Maximum RPC (Revenue Per Customer) is Amazon’s top goal.
- Amazon’s A9 algorithm uses conversion rate, relevance and customer satisfaction to rank products.
- Fill out as much of your product listing page as possible, using as many keywords as possible.
- Use FBA (Fulfillment by Amazon) to automate customer satisfaction.
- Find ways to encourage customer reviews, and do everything you can to keep your customers happy.
- Above all: More sales = higher rankings = more sales
There you have it! You now know exactly what metrics Amazon is looking at to rank listings in their industry-leading product search engine. All that’s left is for you to go out there and capitalize on your superior knowledge!
In the meantime, do you have any additional questions about ranking products on Amazon? Do you need me to clarify something I didn’t explain well enough?
Let me know in the comments below, I’ll be responding to every last one 🙂