Ashley Furniture beats Walmart 66.5% to 13.4% in AI Overviews deal answers

Semrush on November 9, 2026 published a study showing that Amazon, Walmart, Best Buy and Target do not dominate AI answers to US holiday shopping questions. When Google's AI Overviews answered Black Friday furniture deal prompts, Ashley Furniture appeared in 66.5% of them. Walmart, the best placed of the four, appeared in 13.4%.

In Short

Semrush checked more than 10,000 AI answers to US holiday shopping questions from September 2026 and found that Amazon, Walmart, Best Buy and Target lead fewer than a quarter of holiday topics in Google's AI Overviews, with specialist stores and product brands leading most of the rest. That matters to any retailer or brand heading into Black Friday on November 27, because which store gets named depends on the exact question, the AI platform asked and where that platform takes its sources from. Being named and being cited turned out to be separate results: Gemini mentioned Amazon more than any other site yet cited amazon.com in none of its answers.

What Semrush measured

The 28-page report, titled "Peak Season Positioning: Who AI Search Recommends from Black Friday to Christmas," is a mini-study drawn from the September 2026 US snapshot of the Semrush AI Visibility Index. Semrush, which describes itself on the report's cover as "An Adobe Company," tracked 10,514 answers from four systems: Google AI Overviews, ChatGPT, Gemini and Google AI Mode. For each answer it recorded which sites were named in the text and which were cited as sources.

The topic list began with 291 holiday shopping topics from Semrush's AI topic set, the same set that sits behind its AI Visibility Toolkit, selected by name around Black Friday, Cyber Monday, Christmas, gifts and stocking stuffers. Semrush set aside 22 topics that were not about US holiday shopping, such as deals in other countries or Black Friday marketing advice. That left 269 topics, grouped into 12 shopping aisles such as gaming and Christmas gifts.

Brand names were then stripped out. A prompt like "walmart black friday sale" will always mention Walmart, so Semrush removed every prompt containing a brand name and every prompt naming a branded product, such as PS5 or Kindle. That trimmed 1,871 prompts from AI Overviews and 96 from each of the other three platforms. Another 281 AI Overviews prompts and 110 per other platform were dropped for falling outside US holiday shopping.

What remained, which the report calls "tracked prompts," was 6,767 AI Overviews prompts across 235 topics, out of 8,919 in the data, and 1,249 prompts each for ChatGPT, Gemini and AI Mode, out of 1,455. Those three platforms answered exactly the same prompts, so their results can be compared directly. Added together, the four platforms produce the 10,514 answers in the headline figure, by PPC Land's calculation.

AI Overviews is treated differently from the other three, according to the report. Its prompts are worded more like searches, so Semrush compares it with the chat-style platforms on who leads, not on exact percentages. The chat platforms have at most five prompts per topic, too few to name a leader topic by topic, so their results are reported across all topics combined.

Every figure in the study is a share of tracked prompts: the percentage of answers that mentioned or cited a given site. One answer can name several sites, so the shares do not add up to 100%. Results pool all topics, which gives bigger topics more weight. According to Semrush, weighting every topic equally "changes some figures but not who leads on any platform."

Specialists lead the narrow questions

The study names a leader for a topic only when it has at least 10 tracked prompts in AI Overviews. That covers 89 topics and 92% of AI Overviews prompts. Amazon, Walmart, Best Buy or Target lead 23.6% of them, which works out at about 21 of the 89 topics by PPC Land's calculation. Specialist stores and the brands that make the products lead most of the rest.

The study's definition of a specialist is narrow: a store or brand focused on one kind of product, such as Ashley Furniture, evo or Canon. Department stores, warehouse clubs, marketplaces and forums do not qualify. A leader can be any site an answer mentions, including Reddit, and where two sites tie both count as leaders.

Furniture is the largest of these topics. Black Friday and holiday furniture deals had 239 tracked prompts in AI Overviews, and Ashley Furniture was mentioned in 66.5% of them against 13.4% for Walmart. That is close to five times the rate, by PPC Land's calculation.

Other topics show the same pattern, according to the report:

On ski and snowboard deals, the outdoor retailer evo reached 64.2% and none of the big four appeared at all. Shoe deals were almost as one-sided: adidas and Nike each reached 56.4%, against 3.9% for Amazon. PlayStation led the gaming topic at 72.1% even though prompts naming PlayStation products such as the PS5 had been removed from the sample.

Not every lead was wide. Samsung and T-Mobile jointly led Black Friday phone deals at 65.4%, with Best Buy close behind at 61.7%.

A department store outruns the big four

Nordstrom is the case the report flags as a "surprising performer." It does not meet the study's definition of a specialist, yet it was mentioned in 57.6% of Black Friday clothing prompts in AI Overviews. The best of the big four managed 5.0%.

Broader aisles favour the big four

The picture changes when topics are grouped into aisles. Semrush needs at least 200 tracked prompts before naming an aisle leader, and 10 of the 12 aisles cleared that bar. Amazon, Walmart, Best Buy or Target led five of them.

Best Buy led gaming at 58.1% and phones, computers and tech at 49.6%. Target led general deals and sales events at 38.4%. Walmart led toys, kids and baby at 32.3% and home, kitchen and appliances at 26.9%. The other five aisles went to Hulu for streaming and software (37.5%), Macy's for fashion, beauty and personal care (30.0%), Home Depot for tools, auto and garden (20.7%), and Reddit for both sports, fitness and outdoors (15.9%) and Christmas gifts (14.2%).

The aisle view and the topic view can point in opposite directions. Walmart leads the home aisle, but Ashley Furniture leads furniture deals, the largest single topic inside it. A retailer can head a category on average while losing the topic that carries the most prompts.

eBay's seasonal gap

Everyday retail rankings do not carry over to the holiday season, according to the report. Among the 30 sites in the Index's September ranking for E-commerce and Retail in AI Overviews, eBay ranked third, behind Amazon and Walmart and ahead of Target and Etsy. Ranked on the same 30 sites across the holiday topics, eBay fell to ninth.

The holiday order ran Walmart, Amazon, Target, Macy's, Nordstrom, Kohl's, Etsy, JCPenney, then eBay. Semrush notes that the two lists are different measures, so only positions are compared, not figures.

Same question, different shortlists

ChatGPT, Gemini and AI Mode broadly agree on which retailers matter most. All three put Amazon first and Walmart second across the 1,249 shared prompts. The disagreement shows up inside individual answers.

Gemini names the big retailers most often: Amazon in 40.5% of prompts, Walmart in 30.4%, Target in 18.7%, Best Buy in 17.9% and Home Depot in 11.8%. AI Mode named Amazon in 33.1% and Walmart in 21.3%. On ChatGPT the two were close to level, at 17.4% for Amazon and 16.6% for Walmart. AI Overviews, which answers shorter search-style prompts, put Walmart first at 20.2%, ahead of Amazon at 16.4%, Best Buy at 15.2%, Reddit at 11.9% and Target at 11.3%.

Semrush tested how firm those leads are by re-running the rankings on many random mixes of topics. Amazon came first on Gemini and AI Mode every time. Walmart came first in AI Overviews in 92% of runs. On ChatGPT, Amazon came first in 75% of runs, which Semrush treats as close enough to call Amazon and Walmart level.

The overlap between platforms is low. Of the 641 prompts where at least one of ChatGPT, Gemini or AI Mode named Amazon, all three did so in 120, or 18.7%. Two of the three named it in 39.8% and only one in 41.5%. The pattern held for the other big retailers:

• Walmart - named by any of the three in 503 prompts; by all three in 15.9%, by two in 37.8%, by one in 46.3%
• Best Buy - 310 prompts; all three 19.4%, two 36.8%, one 43.9%
• Target - 323 prompts; all three 12.7%, two 33.1%, one 54.2%

Target, in other words, was named by a single platform in more than half the prompts where it appeared at all.

One dehumidifier, three answers

The report illustrates the split with a single tracked prompt: "What are the best dehumidifier deals available this Black Friday?" ChatGPT named five appliance makers - Frigidaire, GE Appliances, Midea, Waykar and hOmeLabs - and no stores. AI Mode named three stores, Amazon, Home Depot and Lowe's, and no makers. Gemini named a mix of both: Amazon, Home Depot and Walmart alongside Frigidaire, Midea and hOmeLabs.

"One question, three very different shortlists," the report states. It also limits how far that can be pushed. "Each platform answered each prompt once, so read these as one shortlist per platform on the day," according to the study, and individual answers can vary from one ask to the next.

Named, never cited: Gemini and amazon.com

The study's third chapter separates two things that AI visibility tools often report together. A mention is a brand appearing in the text of an answer. A citation is a site appearing as a source. The report's executive summary puts the gap bluntly: "Winning citations or mentions offers no guarantee of earning the other."

Amazon is the starkest example. Gemini mentioned Amazon more than any other site, in 40.5% of tracked prompts. Across every Gemini answer Semrush tracked, amazon.com was never cited as a source. Gemini did cite the other three big retailers: Walmart in 5.1% of prompts, Best Buy in 3.4% and Target in 3.1%.

What the robots.txt files show

Semrush points to Amazon's robots.txt file as the likely reason. Amazon's file blocks Google-Extended, the product token site owners use to keep their pages out of Gemini's training and out of the sources Gemini uses to answer. Walmart, Best Buy and Target do not block it.

Google states that Google-Extended does not affect Google Search, according to the report, and the data fits that. AI Mode and AI Overviews, both part of Search, still cite amazon.com. The gap appears only in Gemini.

The robots.txt evidence comes from archived copies of each retailer's file from September 2026, with Best Buy's archived copy dating from May 2026, and from the live files on October 28, 2026. Each retailer's Google-Extended position was the same in both, with one exception: Best Buy's October file names Google-Extended and explicitly allows it.

Semrush is careful about how much weight the explanation can bear. It presents Amazon's block as "the likely explanation for the Gemini pattern," and adds a complicating factor: Gemini cites fewer sites per answer overall, about 2.3, against about 5.5 on ChatGPT and AI Mode. Gemini still manages to cite Walmart, Best Buy and Target within that smaller set. The study does not show what Gemini would do if Amazon removed the block.

The report frames the choice as a strategic one. "Amazon is a true retail giant and this suggests they're comfortable in their own ecosystem - a call that smaller brands can't make," it states.

That reading fits a pattern PPC Land has followed for more than a year. Amazon blocked AI crawlers from Meta, Google, Huawei, Mistral and others in its robots.txt in August 2025, after earlier restrictions on agents from Anthropic, Perplexity and Google's Project Mariner. In November 2025 it sent Perplexity a legal demand to stop Comet browser users from shopping on Amazon, and in September 2026 it cut off Meta's Muse agent from buying on Amazon.com.

Who explains Amazon's deals instead

When Gemini mentions Amazon, someone else's page does the explaining. Across the 506 Gemini answers that named Amazon, the most cited domain was blackfriday.com, a deals site, at 26.3%. Walmart's own site came next at 8.9%, then Best Buy at 7.1%, and Target, Reddit and Mashable at 5.9% each.

Three of the six most cited sources in Amazon answers, then, belong to Amazon's direct competitors. The report's note on this point is about accuracy as much as traffic: for time-sensitive offers like Black Friday, the third-party pages an AI cites carry whatever prices and dates they happen to show.

Reddit, YouTube and Google's own pages

AI Overviews runs the other way. Its most cited source across the holiday topics was Reddit, at 35.2% of tracked prompts, followed by Google's own pages at 35.0% and YouTube at 29.3%. The most cited retailer, Walmart, sat at 16.1%.

Reddit is cited about three times as often as it is named, according to Semrush. The mention and citation shares for the leading sites in AI Overviews were:

The google.com figure reflects Google linking to its own pages, according to the report. Every one of the big four is mentioned more often than it is cited in AI Overviews, while the three non-retail sources are cited far more often than they are mentioned.

Gift questions are where Reddit's citations peak. In Christmas gift prompts, AI Overviews mentioned Reddit in 14.2% of prompts, more than any store, and cited it in 53.2%. Only Google's own pages were cited more often. Semrush offers an interpretation rather than a finding: "One read is that Reddit is full of shoppers sharing what worked, all citable as real recommendations, while retail pages are curated around prices and product details."

Citation and traffic are not the same thing, and Reddit's own numbers show the difference. On its third-quarter 2026 earnings call on October 29, chief executive Steve Huffman said Google's AI Mode does not drive traffic, and Reddit warned that daily users could fall from 130 million in the fourth quarter as search referrals weakened. A source that Google's AI leans on heavily can still see few visitors from it.

Best Buy: mentioned and cited

One retailer bucks the split. Best Buy leads seven holiday topics in AI Overviews, more than any other site, and every one of them is centred on tech, gaming or appliances. The report describes it as a big retailer with "the profile of a specialist."

The Cyber Monday Deals and Sales topic, with 497 tracked prompts, shows the difference citations can make. Best Buy and Amazon were mentioned at almost the same rate, 48.7% and 48.9%. Best Buy's site was cited in 54.5% of prompts, against 27.0% for Amazon's, roughly twice the rate.

What the study does not show

The report is explicit about its scope. It measures visibility in AI answers; sales, traffic and sentiment sit outside it. Results are reported by domain, as a single snapshot of how the four systems answered in September 2026, two months before Black Friday. About 2.0% of AI Overviews prompts mention a past year. Semrush kept them in because every platform's leader is the same with or without them.

The single-answer design matters. Each platform answered each prompt once, so a retailer counted as missing from an answer might appear on the next ask. PPC Land reported in November 2025 that personalization and non-determinism make AI answers vary by user, which means a tracker's clean-account results do not always match what real shoppers see.

The chat-platform sample is also thin at topic level. ChatGPT, Gemini and AI Mode have up to five prompts per topic, which is why topic leaders are only named for AI Overviews. Microsoft Copilot is not part of the study; Semrush stopped tracking Copilot in its enterprise AI visibility product on November 9, the same day the report came out, citing changes to Copilot's access requirements.

Semrush also sells the tools the study recommends. The report closes with a five-point pre-holiday checklist that covers checking whether AI bots can reach a site, including robots.txt and bot-blocking services such as Cloudflare, and whether key pages load without JavaScript; mapping narrow holiday questions and their leaders; testing more than one platform; publishing deal pages with dates, prices and product details; and monitoring prompts "across platforms with Semrush." It also introduces Topical Gravity, described as "Semrush's way of measuring visibility topic by topic."

Why this matters for marketers

The study arrives as AI visibility measurement becomes a product category of its own, with Semrush among the companies selling it. Adobe agreed to buy Semrush for $1.9 billion in cash on November 19, 2025, with the deal completing in April 2026. In June 2026, Semrush expanded its AI Visibility Index to 126 million US prompts and found 36 brands that held a top-100 mention position on every platform in every month. The second 2026 edition, published in October with a UK benchmark, found that fewer than half of the top 100 mentioned brands also appeared in the top 100 cited list, with Tesco ranked 25th and Asda 606th in ChatGPT citations.

The holiday study narrows that mention-versus-citation gap to a single season and a single category, and adds a mechanism. If the Amazon pattern holds, a robots.txt decision taken for crawler or training reasons can remove a site from the sources one AI system cites while leaving its brand name in the answers. For retailers that block AI crawlers to protect data or traffic, the trade-off is that third parties, competitors included, may end up explaining their deals.

The stakes are seasonal. Adobe forecast US online holiday spending of $275.1 billion for November and December 2026, up 6.7%, with AI-referred clicks to US retail sites forecast to rise 130% year on year. Google's AI Overviews are also more present in commercial searches than a year ago: Semrush found in October that they appeared on 88% of tracked commercial keywords in August 2026, up from 45.75% in September 2025, with review queries rising from 30.14% to 89.04%.

Blocking is not always deliberate, either. Earlier this month, an audit found that bot protection stopped Blend's automated shopping agent at 29 of 200 UK and US retailers, even though the stores' robots.txt files allowed the major AI search crawlers. Semrush's checklist points at the same layer.

The report ends by looking past recommendations to purchases. It names Google's Universal Commerce Protocol and WebMCP as emerging standards that let AI agents browse, compare and buy for shoppers. Both are early. Chrome's WebMCP entered an early preview in February 2026 as a way for sites to expose structured tools to agents, and in May 2026 a scan of more than 3 million sites found that only 26 had implemented UCP, none of them the large companies named in Google's launch. UCP adoption at that level means the agent-shopping layer the report anticipates is not yet what decides who gets named in November.

What the data does establish is narrower and more immediate. In September 2026, the answer to "who wins the holiday season in AI search" depended on the topic, the platform and the source, and the largest retailers led only a minority of the narrow questions shoppers ask.

Timeline

• August 21, 2025 - Amazon adds AI crawlers from Meta, Google, Huawei, Mistral and others to its robots.txt blocks
• November 4, 2025 - Amazon sends Perplexity a legal demand over Comet browser shopping
• November 9, 2025 - PPC Land reports that personalization makes AI answers vary by user, undermining tracker accuracy
• November 19, 2025 - Adobe agrees to acquire Semrush for $1.9 billion in cash
• February 10, 2026 - Google announces WebMCP for Chrome in early preview
• April 28, 2026 - Adobe completes the Semrush acquisition
• May 21, 2026 - Only 26 sites found with a public UCP implementation
• June 30, 2026 - Semrush expands its AI Visibility Index to 126 million US prompts and identifies 36 brands visible on every platform
• September 2026 - Semrush captures the US AI Visibility Index snapshot used for the holiday study: 10,514 answers across 269 topics
• September 20, 2026 - Amazon cuts off Meta's Muse agent from buying on Amazon.com
• September 28, 2026 - Adobe forecasts $275.1 billion in US online holiday spending, with AI-referred retail clicks up 130%
• October 13, 2026 - Semrush's UK and US AI Visibility Index edition finds fewer than half of top mentioned brands are also top cited
• October 21, 2026 - Semrush finds AI Overviews on 88% of tracked commercial keywords
• October 28, 2026 - Semrush checks the live robots.txt files of Amazon, Walmart, Best Buy and Target; Best Buy's file now names Google-Extended and allows it
• October 29, 2026 - Reddit's Steve Huffman says Google's AI Mode does not drive traffic
• November 5, 2026 - Bot protection blocks Blend's AI shopping agent at 29 of 200 retailers
• November 9, 2026 - Semrush publishes "Peak Season Positioning: Who AI Search Recommends from Black Friday to Christmas"; the same day, Semrush stops tracking Microsoft Copilot
• November 27, 2026 - Black Friday

Related PPC Land coverage

• Asda ranks 606th in ChatGPT citations as Tesco sits 25th, Semrush finds - the October 2026 edition of the AI Visibility Index, which measured the gap between being mentioned and being cited across the UK and US.
• Semrush: 36 brands win AI visibility everywhere, 1,200 vanish on one - the June 2026 expansion of the Index to 126 million US prompts across four AI platforms.
• Google AI Overviews now hit 88% of tracked commercial keywords, Semrush finds - Semrush's October 2026 study of how often AI Overviews appear on commercial and review searches.
• Semrush drops Copilot from AI visibility tracking on November 9 - why Microsoft Copilot left Semrush's enterprise AI tracking on the day this study was released.
• Adobe acquires Semrush for $1.9 billion to expand brand visibility tools - the November 2025 deal that made Semrush part of Adobe.
• Amazon blocks AI bots from major tech companies amid commerce battle - Amazon's August 2025 robots.txt changes against AI crawlers.
• Amazon blocks Meta's Muse AI agent from shopping on its site - the September 2026 cutoff of Meta's shopping agent.
• Amazon demands Perplexity stop AI shopping in Comet browser dispute - Amazon's November 2025 legal demand against an AI shopping browser.
• Reddit faces Q4 drop from 130m daily users, blaming Google AI Mode - Reddit's warning that Google's AI search features do not send it traffic.
• Bot protection blocks Blend's AI shopping agent at 29 of 200 retailers - an audit of how bot protection stops automated shopping agents at retail sites.
• Adobe puts US online holiday spend at $275.1 billion, up 6.7% - Adobe's 2026 holiday forecast, including a 130% rise in AI-referred retail clicks.
• LLM tracking tools face accuracy crisis from personalization features - why AI answers recorded by tracking tools can differ from what real users see.
• Only 26 sites have implemented UCP - and the big names aren't among them - early adoption data for Google's Universal Commerce Protocol.
• Chrome's WebMCP could end AI agents' pixel-parsing nightmare - how WebMCP lets websites expose structured actions to AI agents.

Summary

Who: Semrush, the search data company owned by Adobe, which produced the study from its AI Visibility Index. The findings concern Amazon, Walmart, Best Buy and Target, specialist stores and brands such as Ashley Furniture, evo, All-Clad and Canon, and sources such as Reddit, YouTube and Google's own pages.

What: A 28-page mini-study of 10,514 AI answers across 269 US holiday shopping topics. It found that the big four retailers lead only 23.6% of the 89 topics with enough prompts in Google AI Overviews, that ChatGPT, Gemini and AI Mode all named Amazon in only 18.7% of the prompts where any of them did, that Gemini never cited amazon.com despite naming Amazon in 40.5% of prompts, and that AI Overviews cited Reddit in 53.2% of Christmas gift prompts.

When: Semrush published the study on November 9, 2026. The data is a September 2026 snapshot, robots.txt files were checked live on October 28, 2026, and Black Friday falls on November 27, 2026.

Where: The United States, across four AI systems: Google AI Overviews, Google AI Mode, Gemini and ChatGPT.

Why: Shoppers increasingly ask AI systems for holiday deals and gift ideas, and the study shows that the stores they hear about depend on the exact question, the platform and its sources, not only on retailer size. The Gemini and amazon.com result suggests that a site's robots.txt choices may shape whether its own pages are cited, while the Reddit result shows how much AI Overviews draws on community content for gift questions.