A new data study found that when the exact same product shows up in both Google's AI Mode and regular Google Search, AI Mode's price is 21.6% higher on average. The study tracked over 2 million product listings across 23 days and is the first large-scale look at whether AI Mode actually finds you the better deal.
Short answer: Productrise tracked 2M+ product listings over 23 days and found that when the same item appeared in both Google AI Mode and traditional Search, AI Mode's price was 21.6% higher on average. AI Mode also showed far fewer options per query (3.9 vs. 27.8) and matched a different seller 49.6% of the time, suggesting price isn't the deciding factor in what it surfaces.
Last updated: September 5, 2026 · By Vishal Swami, Founder & Lead AI Reviewer, AISagely
I use AI Mode for quick product research most weeks, and I'd assumed it was just regular Search wearing a chat interface — same inventory, same ranking, friendlier summary on top. This study is the first hard evidence that's not how it works. In my own side-by-side checks since this study came out, I've caught AI Mode recommending a listing $20-$30 above the top regular-Search result for the same item more than once, which is exactly the pattern the data describes at scale.
What the study actually found
The study comes from Hugo Huijer, founder of the e-commerce data company Productrise, published September 1, 2026. Per the Productrise report, the team tracked more than 2 million product listings and over 100,000 SERPs and AI Mode responses across the US and UK between August 9 and August 31, 2026, using product-centric shopping queries.
The headline number: when the same product identifier showed up in both AI Mode and traditional Search for the same query on the same day, AI Mode's price was 21.6% higher on average. Widen the lens to every product shown on either side, not just exact matches, and the gap grows — AI Mode's median price was $149 versus $100 in regular Search, a roughly 49% difference.
A few more numbers from the same dataset stood out to me:
- Only 1.28% of the products shown in traditional Search also showed up in AI Mode for the identical query — the two surfaces mostly aren't recommending the same inventory at all.
- When a matched product's price disagreed between the two (38.1% of matches), AI Mode was the pricier one 68.4% of the time.
- AI Mode surfaced a different seller than traditional Search on 49.6% of matched products.
- AI Mode displayed an average of 3.9 products per query, against 27.8 for traditional Search.
Why AI Mode isn't optimizing for the lowest price
Google hasn't published the ranking logic behind AI Mode's shopping results, so nobody outside Google can say for certain why this gap exists. But the pattern in the data points somewhere specific: traditional Search's shopping carousel has long rewarded the lowest price for a given item, while AI Mode appears to weigh other signals instead — feed completeness, review volume, and how well a listing's data matches the shopping conversation, according to PPC Land’s coverage of the study.
SEO consultant Brodie Clark, quoted in that same coverage, summed up the old model bluntly: "the retailer with the lowest price gets the click in the end." AI Mode breaks that assumption. Analyst Katelyn Geary put it more sharply: "open market comparison is replaced by a curated path to higher-priced stock." Not everyone reads that as a problem — Jamie D'Alessandro framed it as "a small win for merchants who don't always want to compete solely on price." Whichever side of that you land on, the practical result for a shopper is the same: AI Mode is picking a handful of listings for you, and price alone doesn't decide which ones make the cut.
Traditional Search vs. Google AI Mode: the pricing data
| Metric | Traditional Google Search | Google AI Mode |
|---|---|---|
| Median price shown (all listings) | $100 | $149 (+49%) |
| Price for the same matched product | Baseline | +21.6% higher on average |
| When matched prices disagree, which side is pricier | 31.6% of disagreements | 68.4% of disagreements |
| Average products shown per query | 27.8 | 3.9 |
| Same product appears on both surfaces | — | 1.28% of traditional results |
| Same seller on matched products | — | 50.4% (different seller 49.6% of the time) |
Example prompts that force a real price check
If you're using AI Mode to shop, don't trust the first price it surfaces — make it show its work. These are close to what I use when I want a number I can actually rely on:
- "Show me the 5 cheapest listings for [product] and name the seller for each one, not just the top pick."
- "Compare the price you just showed me against a plain Google Search for the same product and tell me if they match."
- "List every seller you found for [product], sorted by price, lowest first."
- "Before recommending a product, tell me if a lower-priced identical listing exists anywhere in your results."
None of these are complicated, but per the data above, AI Mode's default behavior is to show you 3.9 options and let you assume that's the full picture. Asking it to sort by price, out loud, is often enough to surface the cheaper listing it already had.
Common mistakes shoppers make with AI Mode
The mistake I see most, and the one this study measures directly, is treating AI Mode's first answer as the market price instead of one seller's price. With only 3.9 listings shown per query on average, you're seeing a curated shortlist, not a real comparison shop.
The second mistake is assuming AI Mode and regular Search are pulling from the same ranked list with a chat wrapper on top. They're not — barely 1 in 80 products overlaps between the two for the same query, so a price check in one doesn't confirm anything about the other.
The third mistake is skipping the seller name. Since AI Mode swaps sellers on roughly half of matched products, the "same" item can come with different shipping costs, return policies, or warranty terms that a price-only comparison misses entirely.
Tools that make price-checking easier
If you want a second AI opinion before trusting a price, run the same product query through Perplexity, which cites its sources and tends to show more of its reasoning than AI Mode's compact shopping cards. Gemini is worth a side-by-side check too, since it's a separate Google product with its own retrieval behavior rather than a mirror of AI Mode. If you run a store and want to know whether your own listings are even showing up in AI Mode's answers, AI Search Console is the closest thing to an official visibility report Google offers. Merchants chasing better placement should also read our guide to AI-focused SEO tools, since feed quality looks like one of the levers behind AI Mode's picks. And if you're evaluating any tool's claims about "AI-verified" pricing or ratings, our piece on reading AI tool ratings without getting fooled applies just as well to shopping recommendations as it does to software reviews.
My take
This is the first study I've seen with the scale to back up something I'd only had a hunch about: AI Mode is not a faster way to comparison shop, it's a recommendation engine that happens to sit on top of a search box. My honest verdict — use AI Mode to narrow down what to buy, then run the specific product through a plain Google Search or a price-tracking extension before you check out. The 21.6% gap won't show up on every purchase, but the study makes clear it's common enough that skipping the second check is a real cost, not a rounding error.
Frequently Asked Questions
Is Google AI Mode's pricing always higher than regular Search?
No. The 21.6% figure is an average across matched products, and AI Mode was cheaper on some items — prices disagreed on 38.1% of matched products, and AI Mode was the pricier option in 68.4% of those disagreements, per the Productrise study.
How big was the study behind this claim?
Productrise tracked more than 2 million product listings and over 100,000 SERPs and AI Mode responses across the US and UK, collected between August 9 and August 31, 2026.
Why does Google AI Mode show higher prices for the same product?
Google hasn't disclosed its AI Mode ranking criteria. The data suggests it isn't optimizing for lowest price the way traditional Search's shopping carousel does, weighing factors like feed completeness and review signals instead, according to coverage from PPC Land.
Does this mean I should stop using AI Mode for shopping?
Not necessarily — it's still useful for narrowing choices quickly. The fix is checking the specific product and seller AI Mode recommends against a plain Google Search before buying, rather than treating its first answer as the best price.
How is AI Mode different from a regular Google Search results page?
AI Mode shows far fewer products per query (3.9 vs. 27.8 on average) and pulls from a largely different set of listings — only 1.28% of traditional Search's matched products also appear in AI Mode for the same query.