October 20, 2026

Black Friday 2026: How AI decides what Australians buy

Celia Harding Founder and Director, LEOPRD October 2026

For over two decades, Celia has advised brands on earned media, reputation and now AI visibility.

When Australians ask AI what to buy this Black Friday, the answer will name a handful of products and point the shopper to a specific place to buy them. LEOPRD has tracked thousands of AI shopping responses across Black Friday 2025 and the lead-up to 2026, and the sources behind those answers have moved in twelve months. Editorial still leads, but retailer pages, affiliate sites and YouTube are gaining ground fast. Brands and retailers that publish verifiable Australian deal evidence early will shape the shortlist well before the sales start.

AI shelf visibility is how often, and how favourably, a brand or retailer appears when shoppers ask AI platforms such as ChatGPT, Google Gemini, Perplexity, Microsoft Copilot and Google AI Mode what to buy.

TL;DR (Key findings)

Australians spent $1.5 billion online over Black Friday and Cyber Monday 2025, and 47% of Australians used AI for gift ideas, research or planning that year.
In LEOPRD's Black Friday tracking, editorial's share of AI citations fell from 55% to 40% while retailer first-party pages rose from 16% to 29%.
41% of Black Friday AI responses referenced early access or early-sale timing, so shoppers' shortlists form weeks before the day itself.
On commercial questions such as "What are the best Black Friday TV deals?", AI suggested a next step 90% of the time, and third-party retailers received 43% of that routing.
LEOPRD's Reputation to Revenue 2026 report found brands were AI's primary recommendation in only 29% of answers where the shopper named no brand.

Why does AI matter so much for Black Friday?

Black Friday is a billion-dollar decision weekend, and AI is now part of the decision. Australia Post recorded $1.5 billion in online spending across 3.1 million households over Black Friday and Cyber Monday 2025.

In the same season, Visa found 47% of Australians had used AI for gift ideas, research or planning. That puts the AI answer in the middle of the country's biggest online sales event.

Search behaviour tells the same story. AI is increasingly the first answer, rather than the first click:

Ahrefs found AI Overviews reduce clicks to the top-ranking result by 58%.
Semrush research found 93% of Google AI Mode sessions end without a click to another website.
Gartner predicts organic search traffic will fall by 50% by 2028 as answer engines take over.

Google demand has not gone away. Searches for "Black Friday deals" in Australia reached 110,000 in November 2025, up 40% year on year, and every November has been higher than the last. The difference is that Google's own results now open with an AI-generated answer.

The shopper's question has changed too. OpenAI's study of how people use ChatGPT found 49% of messages were people asking for information or advice to help them act. They ask what to buy, who to trust and what to do next.

"We used to outsource memory to Google. Now we outsource judgement to AI." Celia Harding, Founder, LEOPRD

What does LEOPRD's research show about AI recommendations?

Appearing in an AI answer and being recommended by it are two different results. LEOPRD's Reputation to Revenue 2026 report analysed 420,903 anonymised prompts and 31,200 AI responses covering 27 brands, seven categories, eight AI platforms and two markets: Australia and Great Britain.

When people asked AI for a recommendation without naming a brand, the monitored brands were:

Missing 36% of the time
Mentioned but passed over 35% of the time
The primary recommendation just 29% of the time

44% of prompts asked AI for research, advice or judgement, and AI was used around 20 times more before a purchase than after it. That is exactly the window Black Friday shoppers sit in.

The report also found 86% of the evidence AI cited about a brand came from outside the brand's own website, and only 40% of repeated tests returned the same recommendation outcome every time. LEOPRD's commercial modelling associates brand exclusion with roughly A$2.9 million of exposure for every A$100 million of Australian online retail spending. That figure is a modelled exposure, not a measured loss.

What evidence does AI look for before it recommends a Black Friday deal?

AI needs evidence it can verify across more than one source before it will back a product or a merchant. In LEOPRD's Black Friday tracking, AI answers looked for proof on five questions: does the product perform well, is the discount good value, can the merchant be trusted, does it suit this shopper, and can an independent source back the recommendation up.

Price gets a product considered. Independent proof is what lets AI justify recommending it.

The verification sources AI pointed to most often were Reddit, CHOICE, YouTube, Consumer Reports, Which? and Canstar Blue. Customer-review platforms such as ProductReview and Trustpilot were not major visible citation drivers for Black Friday.

Reviews still count. They feed publisher articles, product summaries and merchant reputation, and can shape model training without appearing as a citation. For Black Friday queries, expert testing, comparison and consumer-authority content carried more visible weight.

The details AI wants to find, and repeat, are specific:

Exact Australian prices, plus previous and sale prices
Sale dates and early-access dates
Model numbers and product variants
Stock availability, delivery and click-and-collect options
Returns and warranty terms
Stable links to product or category pages
Clear "best for" recommendations

*Source: LEOPRD Black Friday AI response tracking, 2025 and 2026.*

How have AI's Black Friday sources changed between 2025 and 2026?

Editorial is still the largest influence on Black Friday answers, but its share has fallen by 15 percentage points. Retailer first-party pages almost doubled their share, and affiliate and reference sites both grew.

Most of this year's Black Friday editorial coverage has not been published yet, so 2026 answers currently lean on international guides, evergreen articles, deal portals and live merchant pages. The publishers that go live first with useful Australian detail have a window to shape the shortlist before it settles.

Brand-owned websites stayed essentially flat while retailer sites surged. Retailer pages naturally carry live price, promotion, stock and delivery information. Many brand sites explain the product well but never state the Australian price, the local stockist or the current deal.

What about Reddit, YouTube and TikTok?

Reddit's share of Black Friday citations fell from 4.8% to 1.8%, even though it remains a frequently named verification source. It still shapes answers, just with a much smaller share of the visible citations.

YouTube has strengthened, particularly in categories where a demonstration helps: consumer electronics, appliances, beauty, gaming, fitness equipment, and furniture and homewares. AI can use video to understand how a product performs, beyond whether it is discounted.

Facebook and Instagram are rising from a small base. TikTok remains negligible, at 0.007% of citations.

HOW ARE BRANDS AND RETAILERS RESPONDING?

One of the challenges we're seeing is that AI visibility still sits across too many different teams. Marketing, ecommerce, PR, SEO, partnerships and product teams all influence the information AI draws on. But they're not always working from the same strategy.

And unlike paid advertising, you can't simply buy your way into an organic AI recommendation. That makes coordination more important than ever. The information brands put into the world needs to be consistent, credible and useful, whether it's coming from a product page, a retailer, an earned media story or a YouTube video.

Some retailers are already thinking this way.

John Lewis recently launched a YouTube chatshow to improve AI search results, after the proportion of its customers using AI to search for products reportedly rose from 0.3% to 2.5% in a year.

It's an interesting example of how content and communications are taking on a new role in influencing product discovery. And a reminder that AI visibility isn't something one team can solve on its own.

Where does AI send Black Friday shoppers?

When a shopper asks AI to compare deals or help them choose, AI usually pushes them towards a purchase. The share of responses that included a call to action varied sharply by intent:

Commercial questions ("What are the best Black Friday TV deals?"): 90%
Transactional requests ("Buy me headphones with 25% off"): 86%
Informational questions ("When is Black Friday?"): 24%

Of the routing AI provided, third-party retailers received 43%, direct-to-consumer brand sites 38%, media and affiliates 9%, and marketplaces 8%.

Direct brand sites are close behind retailers. When AI finds a clear offer, bundle or warranty on a brand's own page, it can skip the multi-brand retailer entirely. For retailers, carrying the product is no longer enough. Being cited as evidence does not guarantee being the recommended destination.

How fast is AI shopping discovery growing?

AI referral traffic is still smaller than search, but each visitor is worth far more. The numbers brands should know:

Semrush found the average AI search visitor is worth 4.4 times an organic search visitor, and Ahrefs found AI visitors converted at up to 23 times the rate of organic visitors on its own site.
ChatGPT leads AI chatbot referrals with roughly 69% share, followed by Microsoft Copilot (9%), Google Gemini (9%), Perplexity (6%) and Claude (5%), according to Statcounter (July 2026).
Gartner expects organic search traffic to halve by 2028.

ChatGPT dominates usage, but a single-platform view of AI visibility will mislead you. More on that in the platform insights below.

How can brands be part of the AI answer this Black Friday?

Black Friday is a six-to-eight-week information campaign for AI, rather than a single sales day. 41% of the Black Friday responses LEOPRD tracked referenced early access or early-sale timing, which encourages shoppers to plan, compare and buy as soon as a credible offer appears.

1. Publish your evidence early: If Black Friday pages, deal feeds, creator content and publisher partnerships only go live in the final week, AI may have already helped shoppers form their shortlist. Get sale dates, early-access details and Australian pricing live and crawlable now.

2. Help the publishers go live first: Last year's earned placements were among the most influential sources. This year, give priority publishers the machine-readable detail they need (exact prices, model numbers, stock, delivery, returns and "best for" calls) so their guides are live before AI starts answering large volumes of planning questions.

3. Fix the digital hygiene on owned and partner pages: Owned pages, for brands and retailers alike, should state live price, availability, fulfilment and sale dates. Label images, publish full video transcripts, keep product descriptions current and make sure retail partners say the same thing you do.

4. Give AI a reason to send shoppers to you: Each destination needs a distinct proposition AI can see and repeat. If your page offers nothing a retailer or marketplace does not, AI may default to Amazon.

5. Show the product working: In categories where demonstration matters, YouTube reviews, comparisons and how-to videos give AI performance evidence. Old, overseas or negative videos can outrank your local campaign, so audit what already exists.

Can smaller brands compete with big names in AI shopping answers?

Yes. AI does not reward a brand for being big, famous or widely stocked. It rewards the business that is the clearest answer to a specific query.

That means building category and use-case authority around the narrow questions shoppers actually ask:

"Best Black Friday robot vacuum for pet hair"
"Best-value OLED TV deal in Australia"
"Black Friday skincare sets for sensitive skin"
"Best gaming headset under a set budget"
"Best Black Friday deal with delivery before Christmas"

The most valuable AI position is the brand AI can confidently select for a defined shopper need. A challenger with sharp evidence for one use case can beat a household name with vague claims.

If shoppers trust AI to decide, AI has to trust something it can check. This Black Friday, the brands that hand it that evidence first will be the ones it recommends.

Frequently asked questions

Do Australians use AI to choose what to buy on Black Friday?

Yes. Visa found 47% of Australians used AI for gift ideas, research or planning in 2025, and Australians spent $1.5 billion online over Black Friday and Cyber Monday that year, according to Australia Post. LEOPRD's Reputation to Revenue 2026 report found AI is used around 20 times more before a purchase than after it.

Which sources does AI trust for Black Friday recommendations?

In LEOPRD's Black Friday tracking, editorial coverage was the largest source of AI citations at 40% in 2026, followed by retailer first-party pages at 29%. The verification sources AI pointed to most often were Reddit, CHOICE, YouTube, Consumer Reports, Which? and Canstar Blue. Customer-review platforms such as ProductReview and Trustpilot were not major visible citation drivers.

Do customer reviews still matter for AI visibility on Black Friday?

Yes, though less visibly. Reviews shape publisher articles, product summaries and merchant reputation, and can influence model training without appearing as a citation. For Black Friday queries, expert testing and consumer-authority content such as CHOICE carried more visible weight.

When should brands start preparing for Black Friday AI search?

Six to eight weeks before the sales. LEOPRD found 41% of Black Friday AI responses referenced early access or early-sale timing, so shoppers form shortlists well before the day. Pages, deal feeds and publisher partnerships that only go live in the final week risk missing that shortlist.

Does AI send Black Friday shoppers to retailers or brand websites?

Both. In LEOPRD's tracking, third-party retailers received 43% of AI routing, direct brand sites 38%, media and affiliates 9% and marketplaces 8%. AI tends to favour the destination with the clearest visible reason to buy there, such as an exclusive bundle, price matching or faster delivery.

Is ChatGPT the only AI platform that matters for Black Friday?

No. ChatGPT holds around 69% of AI chatbot referrals, but Gemini, Copilot, Perplexity and Google AI Mode each trust different sources. In LEOPRD's research, one brand appeared 2.3 times more often on Gemini than on ChatGPT for the same prompts.

What is LEOPRD's Reputation to Revenue 2026 report?

LEOPRD's Reputation to Revenue 2026 report examines how AI influences brand discovery, recommendation and consumer decisions. It combines 420,903 anonymised prompts with 31,200 AI responses covering 27 brands, seven categories, eight AI platforms and two markets: Australia and Great Britain. It found monitored brands were the primary recommendation in only 29% of unbranded answers.

The bottom line

AI is now part of how Australians decide what to buy on Black Friday, and it recommends what it can verify. The sources it relies on are shifting towards retailer pages, affiliates, reference sites and video, while this year's editorial coverage is still to be written. Brands and retailers that publish clear Australian prices, dates, stock and proof in the next few weeks can shape the shortlist. Those that wait for the final week will find AI has already chosen.

Find out how AI is talking about your brand this Black Friday

Download LEOPRD's Reputation to Revenue 2026 report to see what makes AI recommend one brand over another. Or book an AI Visibility Audit to find out what ChatGPT, Gemini, Perplexity and Copilot are saying about your products right now, and what it would take to change the answer before Black Friday.

About the author

Celia Harding is the founder of LEOPRD, a PR and AI visibility agency specialising in Language Engine Optimisation (LEO). LEOPRD helps brands understand and improve how they are represented and recommended by AI systems including ChatGPT, Google Gemini and Perplexity. Celia is the winner of CommsCon PR Professional of the Year and B&T Women Leading Tech PR 2026, and LEOPRD was named Best New Agency in Asia-Pacific at the 2026 SABRE Awards.

About the research

LEOPRD's Black Friday findings come from thousands of AI shopping responses tracked in the lead-up to and during Black Friday 2025, compared with responses tracked again in the lead-up to Black Friday 2026. Each response was analysed for the sources cited, the timing referenced and where AI directed the shopper. Platform comparison figures come from LEOPRD's wider brand monitoring.

The Reputation to Revenue 2026 report draws on a de-identified sample of 420,903 prompts from Prompt Cowboy (November 2025 to August 2026) and 31,200 responses from a wider set of 214,000 monitored-brand AI responses, coded across 27 brands in Australia and Great Britain. Commercial figures are modelled exposure, not measured revenue loss.

Sources

Download the 2026 Report

📥 Download: REPUTATION TO REVENUE REPORT 2026

Understand how LLMs recommend your brand - and how to get ahead while everyone else is playing catch-up.