Amazon Rufus is now Alexa for Shopping: what sellers need to know

What changed (updated October 22, 2026)

We first published this guide in March 2026, when Amazon Rufus was the assistant every seller was trying to understand. Two months later Amazon retired the name. This version replaces the original and is current as of October 22, 2026.

• May 13, 2026. Amazon brought Rufus and Alexa+ together into one assistant, Alexa for Shopping. Amazon's own Rufus page now carries a one-line note: "On May 13, 2026, Rufus was renamed Alexa for Shopping" (Amazon).
• Where it runs. The Amazon Shopping app, Amazon.com and Echo Show devices, rolled out to US customers over the following week, free for signed-in customers, no Prime or Echo needed (Amazon, TechCrunch).
• June 11, 2026. Amazon Ads said existing sponsored ads campaigns are automatically eligible to serve inside Alexa for Shopping conversations (Amazon Ads).
• July 27, 2026. A new 75-character limit on product titles in non-media categories took effect, with a new 125-character Item Highlights field (PPC Land).
• July 30, 2026. Amazon reported that more than 350 million customers used Alexa for Shopping over the previous 12 months and that its active users nearly doubled in the second quarter (Amazon Q2 2026 release, Retail Dive).
• September 29, 2026. Amazon Ads announced Branded Conversations, a sponsored format inside Alexa for Shopping, launching in the US in January 2027 (Amazon Ads).

If you only have a minute: the assistant is bigger than Rufus was, it now sits in the main Amazon search bar, and it can buy on the shopper's behalf. Amazon still has not published how it chooses which products to recommend.

Is Amazon Rufus discontinued?

The name is. The assistant is not.

The Amazon Rufus AI assistant launched in beta in February 2024 as Amazon's generative shopping assistant. On May 13, 2026, Amazon folded it into Alexa for Shopping, which it describes as bringing Rufus's product expertise and Amazon shopping history together with the personal context of Alexa+ (Amazon). The Rufus chat bubble was replaced by an Alexa icon, and shoppers can now type questions straight into the search bar (Modern Retail).

This was a consolidation, not a retreat. Two weeks before the change, Amazon said Rufus monthly active users were up 115% year over year and engagement was up 400% (Modern Retail). In February it had credited Rufus with nearly $12 billion in incremental annualized sales in 2025, a figure that is Amazon's own attribution estimate. The reason Amazon gave for merging was practical: Rufus and Alexa kept separate memory, so a shopper who started on one had to start again on the other (EcomCrew, citing an interview with Amazon's Rajiv Mehta).

Two things are still unclear as of October 2026. First, Amazon's launch announcement covered US customers only, and we have not found an Amazon announcement that dates the rename in other marketplaces where Rufus launched, such as the UK, Germany or India. If you sell outside the US, check what your marketplace shows today. Second, Amazon has not said how much of the original Rufus model still runs underneath; GeekWire reported that it keeps powering parts of the experience, but Amazon's own posts do not spell that out.

Amazon Rufus did not disappear. It moved into the search bar, picked up a memory and a wallet, and kept reading the same listings.

Kris Krokos

What is Alexa for Shopping, in seller terms

Strip away the launch language and Alexa for Shopping does five things that matter to a seller. All five come from Amazon's own feature list (Amazon).

1. It answers questions in the main search bar. The shopper no longer has to open a separate chat. A question and a keyword now go into the same box.
2. It writes AI overviews at the top of search results and on product detail pages. Before the shopper sees your title, they may read Amazon's summary of the category and of your product.
3. It builds side-by-side comparisons. A shopper can select products from search results and get a table of features, prices and reviews.
4. It shows up to a year of price history. A "was" price that only existed for a week is now easy to check.
5. It can buy without the shopper being there. Price alerts, Auto-Buy at a target price, Scheduled Actions for recurring orders, and Buy for Me for some products sold on other retailers' sites.

The last point is the one most seller guides skip. Once a shopper sets "buy this when it drops to $40", the purchase happens in the background. The listing that wins is the one the assistant suggested when the shopper set the rule, not the one with the best ad at the moment of purchase.

What it reads

Amazon has been more open about inputs than about weighting. When it launched Rufus, Amazon said the model was trained on its product catalog, customer reviews, community Q&A and information from across the web (Amazon). Amazon's science team later described a retrieval step that pulls from reviews, the catalog and community Q&A, and calls internal store APIs before writing an answer (Amazon Science). Alexa for Shopping adds the shopper's own history: past orders, lists, browsing and preferences they have told Alexa.

For a seller, that means three of the four main inputs are things you influence directly: your catalog data, the reviews your product earns, and the questions your listing answers. The fourth, personal history, you cannot touch.

What Amazon has confirmed, and what is commentary

Since May, dozens of agencies and tool vendors have published "how Alexa for Shopping ranks products" guides. Most of them describe the same plausible model, and almost none of it is confirmed by Amazon. It helps to keep the two lists apart.

What Amazon has not published, as of October 2026: a ranking formula for organic recommendations, a Seller Central report that breaks out traffic or sales from Alexa for Shopping, or any statement that a specific listing field raises a product's chance of being recommended.

The commentary is not worthless. Some of it rests on real measurement. AMALYTIX, a listing analytics vendor, ran 500 top US search terms through Rufus in November 2024 and analyzed more than 1,300 recommended products: the median rating was 4.5, the median review count was 2,991, and 92.1% were Prime eligible (AMALYTIX). The authors say plainly that this is correlation, not causation, and the data predates Alexa for Shopping. Treat numbers like these as a description of what tended to win, not as rules.

Why this matters more than an algorithm update

Classic Amazon search shows the shopper a page of results and lets them compare. Every product on that page gets an impression, and position decides how many clicks it gets. Alexa for Shopping changes the step before the click. The shopper asks a question, reads an overview, maybe asks for a comparison, and only then opens a listing. Products that never make it into the answer never get the impression.

We cover the mechanics of that shift in Alexa for Shopping vs Amazon search. The short version for sellers:

• The question replaces the keyword. "Quiet blender for a small kitchen under $100" carries constraints a keyword like "blender" never did. Your listing has to state the facts that answer it.
• Comparisons expose gaps. If your listing does not state noise level and a competitor's does, the comparison table shows a blank next to your name.
• Price history punishes fake discounts. A year of history sits one tap away.
• Personalization removes the single rank. Two shoppers asking the same question can get different picks, so there is no one position to track.
• Ads are moving into the answer. Sponsored ads and sponsored prompts already serve inside conversations; Branded Conversations arrive in January 2027.

None of this makes keyword search irrelevant. Search results still sit under the overview, and most shoppers still scroll. But the share of decisions made inside the assistant is growing, and Amazon reported in July that US customers who use Alexa for Shopping spend over 40% more per order than those who do not (Amazon Q2 2026 release). That is Amazon's figure and it does not prove the assistant causes the difference, but it explains why Amazon is pushing it.

What Amazon sellers should do now

The good news is that very little of the March advice was wrong. Listings that are clear, complete and honest were easier for Rufus to summarize, and they are easier for Alexa for Shopping too. What changed is the urgency and a few specific rules.

1. Rewrite titles for the 75-character limit

This is the one hard deadline. Since July 27, 2026, titles over 75 characters in non-media categories can be rewritten by Amazon's AI; Brand Registry owners get 14 days to review the suggested change (PPC Land, M2E Pro). Write the title yourself: brand, product type, the one or two specs a shopper filters on. Move materials, use cases and compatibility into Item Highlights. If you leave it, a model writes your title for you.

2. Fill the attributes a shopper would ask about

Comparison tables and overviews are built from structured data. Go through the category's attribute list and fill every field a shopper might put in a question: size, capacity, material, compatibility, power, care. Consistent values across variations matter as much as completeness. Our guide to product attributes explains why missing values cost more than wrong ones.

3. Answer the real questions in the listing

Read your reviews and the questions shoppers ask on your detail page. The recurring ones ("does it fit a 15-inch laptop?", "is it loud?") are exactly what shoppers now type into the search bar. Answer them in bullets, Item Highlights and images, in plain words, so the assistant has a sentence to quote.

4. Fix the complaint, not the review

The assistant reads reviews as evidence. A recurring complaint about sizing or a flimsy part shows up in AI summaries and comparisons. The fix is upstream: a clearer size chart, a better part, a more accurate claim. Amazon's new Review Requests ad product, launching in November 2026, can speed up reviews on new products, but it will not hide a pattern (Amazon Ads).

5. Price for a shopper who can see a year back

Inflated reference prices are now visible. Price alerts and Auto-Buy also mean a price drop can trigger orders you did not plan for, so make sure stock is ready before you discount.

6. Check where your ads are serving

Existing sponsored ads campaigns are eligible for Alexa for Shopping by default (Amazon Ads). Look at your sponsored prompt reporting, decide whether you want those placements, and plan whether Branded Conversations are worth testing when they launch.

7. If you also sell on your own site, check Buy for Me

Buy for Me lets Amazon customers buy some products from other retailers' websites, using details Amazon pulls from those sites. In January 2026 several small brands complained they had been included without asking; Amazon said businesses can opt out by emailing branddirect@amazon.com (Modern Retail). Decide on purpose whether you want to be in it.

For the field-by-field version, see how Alexa for Shopping evaluates a listing and our full guide to Amazon listing optimization for AI shoppers. If you want to see how Amazon's assistant compares with ChatGPT, Gemini and the rest, read AI shopping assistants compared. More on catalog work lives under Product Data & Catalogs.

The bottom line

Amazon Rufus taught sellers that the listing is no longer read only by people. Alexa for Shopping makes that permanent: it sits where search used to start, it remembers the shopper, and it can complete the purchase. Amazon has told sellers what the assistant reads and very little about how it decides. Until that changes, the practical response is the same one that worked for Rufus, done more carefully: state the facts a shopper asks about, keep them consistent, and stop doing the things a model can now see through.

FAQ

Is Amazon Rufus discontinued?

The Rufus name was retired on May 13, 2026, when Amazon renamed the assistant Alexa for Shopping. The shopping assistant itself continues inside Alexa for Shopping, available to US customers in the Amazon app, on Amazon.com and on Echo Show.

What is Amazon Rufus?

Rufus was Amazon's generative AI shopping assistant, launched in beta in February 2024. It answered product questions using Amazon's catalog, customer reviews, community Q&A and web information. Since May 2026 it is part of Alexa for Shopping.

Is Alexa for Shopping the same as Rufus?

It is Rufus plus Alexa+. Amazon combined Rufus's product knowledge with Alexa+'s memory and personal context, moved it into the search bar, and added price history, Auto-Buy, Scheduled Actions and Buy for Me.

Can sellers optimize for Alexa for Shopping?

Not directly, because Amazon has not published how it picks products. Sellers can control what it reads: titles, Item Highlights, attributes, images, answered questions and review quality. Claims about exact ranking factors are third-party commentary.

Is Alexa for Shopping available outside the US?

Amazon's May 2026 launch covered US customers. As of October 2026 we have not found an Amazon announcement dating the change in other marketplaces, so sellers outside the US should check what their own marketplace shows.