Amazon Ranking in 2026-2027: A9, A10, AI Search and Rufus. What Actually Matters

📌 Quick Answer

Amazon ranks products by how likely a shopper is to buy them. In 2026-2027 that works in three layers: keyword relevance (what sellers call A9), performance signals such as conversion, sales, reviews and price (what sellers informally call A10), and AI intent matching through Alexa for Shopping, which replaced Rufus on May 13, 2026. Win relevance first, conversion second, and intent-ready content third.

Most sellers spend their time arguing about whether an Amazon A10 algorithm exists. Meanwhile, the way shoppers find products has moved on. They type a keyword, but they also ask a question, tap a suggested prompt, or let an AI assistant narrow the shortlist before they ever scroll a results page.

This guide cuts through the noise. You will learn what A9 and A10 really are, what changed when Rufus became Alexa for Shopping, which seven ranking signals deserve your time, and a 30-minute audit you can run on your top ASIN today.

A note on evidence: Amazon has never published its ranking formula or factor weights. Everything below is labelled as either Amazon's own published research and guidance, or industry observation. If an article gives you exact percentages for each ranking factor, treat it as opinion, not fact.

Table of Contents

• ●What Are the Amazon A9 and A10 Algorithms?

• ●The 2026-2027 Amazon Ranking Stack: Three Layers

• ●Rufus Is Now Alexa for Shopping: What Actually Changed

• ●The 7 Amazon Ranking Factors That Actually Matter in 2026-2027

• ●Myths vs Reality in 2026-2027

• ●Do AI Tools Outside Amazon Matter?

• ●Your 30-Minute Listing Audit

• ●A Realistic 90-Day Plan

• ●Common Mistakes to Avoid

• ●Why Trust This Guide

• ●Frequently Asked Questions

What Are the Amazon A9 and A10 Algorithms?

A9 is the long-standing nickname for Amazon's product search ranking system. It comes from A9.com, the search subsidiary Amazon created in 2003 to own product discovery.

A10 is a community label. Sellers and agencies began using it around 2020 to describe a shift toward customer experience, conversion and seller quality, and Amazon has not announced a successor by that name. Jungle Scout notes that the name originated with sellers, not with Amazon, and SellerLabs makes the same point in its explainer. Some 2026 articles claim A10 launched in 2025, but none of those claims trace back to an Amazon announcement.

The practical takeaway: stop asking which version is live. Ask what the system is trying to do. It is trying to show the product a shopper is most likely to buy, and that has been true for every version of the name.

The 2026-2027 Amazon Ranking Stack: Three Layers

It helps to think of ranking as a stack rather than a single formula.

Layer 1 gets you into the pool of eligible results. Layer 2 decides your position in it. Layer 3 decides whether an AI assistant recommends you when the shopper describes a need instead of typing a product name.

Why Layer 3 Is Not a Gimmick: Amazon's COSMO Research

Amazon's own scientists published a system called COSMO in 2024. In plain terms, it mines shopping behavior, such as what people searched for and then bought, and builds a knowledge graph about why people buy things. Amazon's research blog gives the example of a shopper whose behavior implies they need slip-resistant shoes, a need they never typed. The paper reports the knowledge was expanded across 18 major product categories and deployed in search applications such as search navigation.

What Amazon has not said is how much weight this knowledge carries in organic ranking versus navigation and recommendations. Some agencies argue it sits underneath Alexa for Shopping, and that is plausible, but it is an inference, not an Amazon statement. The safe reading is simple: Amazon is investing in understanding shopper intent, so listings that explain who the product is for and when to use it are better positioned than listings that only repeat keywords.

Rufus Is Now Alexa for Shopping: What Actually Changed

If your playbook still says Rufus, update it. On May 13, 2026, Amazon retired the standalone Rufus chatbot and replaced it with Alexa for Shopping, which combines Rufus's product knowledge with Alexa+ personalization, according to CNBC and Modern Retail reporting. Agencies tracking the rollout report it now appears in the search bar, search results and product detail pages for signed-in US customers, without needing a Prime membership or an Echo device.

What changed: the name, the placement, and how visible the assistant is.

What did not change: the assistant still works from your detail page, reviews and Q&A, and Amazon still has not published a ranking formula or an official optimization playbook for it. Canopy Management makes the same point: the listing principles built for Rufus carry over.

Two cautions worth stating plainly.

Availability Differs by Country

Alexa for Shopping is a US launch, so if you sell in the UK, UAE or Canada, check what shoppers in your marketplace actually see before changing strategy.

Seller Central Has No Confirmed Assistant-Traffic Report

Seller Central does not give you a confirmed report of organic traffic from the assistant. Anyone selling you a dashboard of assistant rankings is selling inference.

The 7 Amazon Ranking Factors That Actually Matter in 2026-2027

These are the signals that consistently show up in Amazon's own seller guidance and in industry testing. The order is practical priority, not an official weighting.

1. Keyword Relevance and Indexing

If Amazon cannot match your listing to the query, nothing else matters. Put the most important keyword and key benefit early in the title, where shoppers on mobile will actually read it, and use your bullets and backend search terms for supporting phrases. Do not stuff. Repetition adds no relevance and hurts readability.

Do this: build a keyword map with one primary term per ASIN and 5 to 10 supporting terms, then check that each appears naturally in the title, bullets or backend terms. Our Amazon SEO services team starts every engagement with exactly this map.

2. Click-Through Rate

Ranking only helps if shoppers click. The main image, title, price and review snapshot all compete in a crowded results page, and a listing that wins impressions but loses clicks sends a weak signal back to Amazon.

Do this: test your main image against the top three competitors on a phone screen. If the product is not obvious at thumbnail size, fix that first. Professional Amazon product image editing is often the cheapest ranking lever available.

3. Conversion Rate

Industry testing keeps pointing to conversion as one of the strongest performance signals: if shoppers click but do not buy, Amazon learns the listing is a poor fit for that query. Conversion comes from clear images, honest benefits, a competitive price, delivery speed and social proof.

Do this: add infographic images that answer the top three buyer objections, and upgrade the detail page with A+ Content and comparison charts. Amazon's own seller guidance on boosting listings recommends rich content such as A+ Content and video.

4. Sales Velocity and Consistency

Sales history helps ranking, and steady sales beat spikes. Advertising can feed this loop by earning early orders on target keywords, which is why we treat Amazon PPC management and listing work as one system, not two. Be aware that this is industry observation: Amazon does not publish how ad-driven sales are weighted in organic rank.

Do this: protect your best-converting keywords with controlled ad spend, and avoid pausing and restarting campaigns that are driving consistent orders.

5. Reviews and Ratings

Reviews affect both conversion and relevance. Recent, detailed reviews tell Amazon and the shopper what the product is really used for. Review language is also the raw material AI assistants draw on.

Do this: export your last 50 reviews, list the phrases shoppers repeat (use cases, praise, complaints), and mirror the real ones in your bullets and Q&A. Never offer incentives for reviews. It violates Amazon policy and puts your account at risk.

6. Price, Availability and Seller Health

A great listing that is out of stock or overpriced loses the Buy Box and the ranking that comes with it. Fulfillment method, stock depth and account health feed the same trust picture. Our account management work is mostly this: preventing the quiet problems that erase organic rank.

Do this: set stock-out alerts at least 45 days ahead of your lead time, and check account health weekly.

7. Intent-Ready Content and Complete Attributes

This is the new frontier. Fill every attribute Seller Central asks for, since the Listing Quality tools in Seller Central flag missing attributes and content gaps that can suppress search visibility. Then write for the question behind the keyword: who is it for, when would they use it, how does it compare, what size or material or compatibility does it have?

Do this: rewrite one bullet per ASIN as a use-case statement, for example by naming the buyer and situation instead of listing a generic feature. Our AI-powered listing optimization service builds this layer on top of your existing keyword work, so you keep the rankings you already have.

Myths vs Reality in 2026-2027

Do AI Tools Outside Amazon Matter?

Yes, as a discovery layer. Shoppers increasingly research in AI tools before they land on Amazon, and OpenAI launched ChatGPT shopping research in November 2025. SellerLabs reports that Amazon blocked OpenAI's crawlers around that time, which means your Amazon listing may not feed those answers directly. That is a single industry source, so verify it for your own category.

The practical response is to make your brand easy to find and describe outside Amazon: a clear brand website, consistent product naming, and credible third-party mentions. Think of it as the same product story told in more places. Our guide on finding winning products before they go viral shows how external signals such as TikTok and Google demand show up before Amazon search volume does.

Your 30-Minute Listing Audit

Run this on your top-revenue ASIN today.

☐ Title: main keyword and top benefit are in the first 60 to 80 characters ☐ Main image: product is clear at thumbnail size on a phone ☐ Bullets: each one leads with a benefit and names a buyer or use case ☐ Attributes: no empty recommended fields in Seller Central's listing quality tools ☐ A+ Content: includes a comparison chart and at least one use-case image ☐ Reviews: top 5 repeated phrases from reviews appear in your copy ☐ Q&A: every real shopper question has a clear answer ☐ Price and stock: you hold the Buy Box and have more than 45 days of inventory ☐ Backend terms: no repeated words, no competitor brand names

Want a second pair of eyes? Our free ASIN audit covers this checklist and more, and our free PPC audit tool shows where ad spend is leaking.

A Realistic 90-Day Plan

Days 1 to 30: Fix the Foundation

Keyword map, title and bullet rewrite, attribute completion, image upgrade on your top 5 ASINs.

Days 31 to 60: Lift Conversion

A+ Content, comparison charts, review-theme mining, price and stock review, ad structure cleanup.

Days 61 to 90: Build the Intent Layer

Use-case copy, answered Q&A, brand presence outside Amazon, and a monthly rank and conversion report.

Measure what you can actually measure: keyword rank for your primary terms, unit session percentage (conversion), organic versus ad-attributed sales, and review themes. Do not chase an assistant ranking report that Amazon does not provide.

For context on the economics behind these choices, see our honest breakdown of whether Amazon FBA is still worth it in 2026 and 2027, and follow our marketplace updates page for changes as they land.

Common Mistakes to Avoid

Chasing the Label

Spending weeks debating A9 versus A10 instead of fixing conversion.

Keyword Stuffing

It hurts readability, conversion and AI interpretation at once.

Rewriting Everything at Once

Change one variable at a time on your top ASINs so you can see what worked. Protect rankings you already hold.

Ignoring Stock

A stock-out can undo months of ranking work.

Buying Shortcuts

Review manipulation and fake Q&A put the whole account at risk.

Copying One-Market Advice

Alexa for Shopping is a US launch, so confirm what applies in your marketplace.

Why Trust This Guide

GrowithAmazon is an Amazon Ads certified agency that manages 35,000+ ASINs across 8+ Amazon marketplaces, and that day-to-day work with Seller Central, Vendor Central, PPC and listings shapes what we prioritize here. For an example of conversion work in practice, read how we helped Home Harbour cut return rates from above 40% to below 20% in this case study, or browse all client results. This guide cites Amazon's own research and seller guidance where it exists and labels industry observation as such. GrowithAmazon is an independent agency and is not affiliated with Amazon.