The 5W ChatGPT Review Study examined the 50 products ChatGPT shows most often in U.S. cosmetics shopping results and observed no consistent minimum review count. Between August 3 and November 1, 2026, three of those products had fewer than 100 reviews, 13 had fewer than 500, and the lowest star rating was 3.8. Review counts only weakly tracked visibility. This is 5W’s analysis using Profound’s tools.
By Ronn Torossian, Founder and Chairman of 5W. Last updated: November 2, 2026. Data window: August 3 to November 1, 2026. This is Part 2 of the 5W ChatGPT Shopping Index, and Ronn Torossian discusses both studies in an op-ed on Everything-PR.
In this report
- What did the 5W ChatGPT Review Study find?
- Which beauty products rank in ChatGPT with very few reviews?
- What are the 10 most visible cosmetics products in ChatGPT?
- How fragmented are ChatGPT’s top products?
- How many reviews do ChatGPT’s top beauty picks have?
- Do reviews and ratings predict ChatGPT visibility?
- Does ChatGPT shopping data include reviews in every category?
- What do Part 1 and Part 2 show together?
- What should brands do with these findings?
- How did 5W build the ChatGPT Review Study?
- Conclusion
- Frequently asked questions
What did the 5W ChatGPT Review Study find?
The 5W ChatGPT Review Study found that ChatGPT’s most visible cosmetics products carry strong ratings but show no consistent minimum review count, and that review counts track visibility only weakly.
| Measure | Result |
|---|---|
| Products analyzed | The 50 most visible cosmetics products, out of 3,685 that ChatGPT surfaced |
| Median star rating | 4.5, with a range of 3.8 to 5.0 |
| Rating of 4.0 or higher | 49 of 50 products (98%) |
| Rating of 4.5 or higher | 36 of 50 products (72%) |
| Median review count | About 2,000, with a range of 19 to 24,681 |
| Fewer than 1,000 reviews | 18 of 50 products (36%) |
| Fewer than 500 reviews | 13 of 50 products (26%) |
| Fewer than 100 reviews | 3 of 50 products (6%) |
| Rank correlation, reviews and visibility | 0.13, which is very weak |
| Rank correlation, reviews and average position | -0.32, which is modest |
These are associations among the products ChatGPT already shows. The study cannot say how many reviews a product needs to appear at all, because products outside the top 50 were not analyzed, and it does not show that reviews have no effect on ranking.
Which beauty products rank in ChatGPT with very few reviews?
Three of the 50 most visible cosmetics products have fewer than 100 reviews: Juvia’s Place I Am My Skin Hydra-Tint Foundation Stick with 19, Essence Radiant Light Face Palette with 21, and e.l.f. Cosmetics Soft Glam Cream Bronzer with 27.
| Product | Reviews | Star rating | Visibility | Rank among the 50 |
|---|---|---|---|---|
| Juvia’s Place I Am My Skin Hydra-Tint Foundation Stick | 19 | 4.5 | 1.4% | 35th |
| Essence Radiant Light Face Palette | 21 | 4.2 | 1.9% | 19th |
| e.l.f. Cosmetics Soft Glam Cream Bronzer | 27 | 4.4 | 1.8% | 21st |
| Essence Foundation Stick (lowest star rating) | 147 | 3.8 | 2.5% | 8th |
The lowest star rating in the group, 3.8, belongs to Essence Foundation Stick, which has 147 reviews and ranks eighth by visibility. At the other end, e.l.f. Cosmetics Power Grip Primer has 24,681 reviews and ranks first, so very high and very low review counts both appear among the most visible products.
What are the 10 most visible cosmetics products in ChatGPT?
e.l.f. holds four of the 10 most visible cosmetics products in ChatGPT shopping results, led by e.l.f. Cosmetics Power Grip Primer, which appears in 4.3% of shopping answers.
| Rank | Product | Reviews | Star rating | Visibility |
|---|---|---|---|---|
| 1 | e.l.f. Cosmetics Power Grip Primer | 24,681 | 4.7 | 4.3% |
| 2 | e.l.f. Cosmetics Camo Liquid Blush | 8,647 | 4.7 | 3.4% |
| 3 | e.l.f. Cosmetics Glow Reviver Lip Oil | 11,772 | 4.7 | 3.4% |
| 4 | NYX Professional Makeup Butter Gloss Non-Sticky Lip Gloss | 7,464 | 4.5 | 3.1% |
| 5 | NYX Professional Makeup Bare With Me Hydrating Face & Body Concealer Serum | 1,594 | 4.5 | 3.0% |
| 6 | Essence Lash Princess False Lash Effect Mascara | 7,277 | 4.2 | 2.8% |
| 7 | NYX Professional Makeup Fat Oil Lip Drip Lip Oil | 3,425 | 4.5 | 2.6% |
| 8 | Essence Foundation Stick | 147 | 3.8 | 2.5% |
| 9 | Maybelline Fit Me Loose Finishing Powder | 2,123 | 4.3 | 2.5% |
| 10 | e.l.f. Cosmetics Power Grip Matte Primer | 24,678 | 4.7 | 2.2% |
NYX Professional Makeup holds three of the 10 spots and Essence holds two. Review counts in the top 10 range from 147 to 24,681, and star ratings range from 3.8 to 4.7.
How fragmented are ChatGPT’s top products?
Product visibility in ChatGPT is fragmented: no single product appears in more than 12.1% of shopping answers in any category, and ChatGPT surfaced 10,267 distinct products across the four categories.
| Category | Distinct products | Most visible product | Visibility |
|---|---|---|---|
| Cosmetics | 3,685 | e.l.f. Cosmetics Power Grip Primer | 4.3% |
| Baby gear and furniture | 3,432 | Stokke Tripp Trapp High Chair 2 with Adjustable Baby Set, White | 5.8% |
| Skincare and makeup | 1,550 | DRMTLGY Brightening Eye Masks with Peptides and Hyaluronic Acid | 12.1% |
| Golf equipment | 1,600 | Callaway Elyte X Driver | 7.9% |
Brands fill much of each top 50. In cosmetics, e.l.f. accounts for 15 of the 50 products. In baby gear, Stokke (11), Bugaboo (9) and UPPAbaby (6) account for 26 of the 50. Eighteen of the 50 baby gear products are listings with “at Nordstrom” in the product name, which helps explain Nordstrom’s second-place merchant share in Part 1 of the index. In skincare and makeup, the top 50 are mostly eye patches, lip balms and face masks.
Retailer results are more concentrated. In Part 1, the top three retailers hold 59.3% to 79.5% of shopping results in each category, while product results are spread across thousands of listings.
How many reviews do ChatGPT’s top beauty picks have?
The 50 most visible cosmetics products have a median of about 2,000 reviews, and 18 of them have fewer than 1,000.
| Threshold | Products at or below | Share of 50 |
|---|---|---|
| Fewer than 100 reviews | 3 | 6% |
| Fewer than 500 reviews | 13 | 26% |
| Fewer than 1,000 reviews | 18 | 36% |
| Fewer than 5,000 reviews | 34 | 68% |
| Star rating below 4.5 | 14 | 28% |
| Star rating below 4.0 | 1 | 2% |
Thirty-two of the 50 products have 1,000 or more reviews, but 18 do not. More than one in four products (28%) has a rating below 4.5, and only one product has a rating below 4.0.
Do reviews and ratings predict ChatGPT visibility?
Review counts and star ratings track visibility only weakly among the products ChatGPT already shows. The study measured rank correlation, a score from -1 to 1 where 0 means no relationship.
| Relationship | Rank correlation | Reading |
|---|---|---|
| Reviews and visibility | 0.13 | Very weak |
| Star rating and visibility | 0.06 | None |
| Reviews and share of first-place listings | 0.25 | Weak |
| Star rating and share of first-place listings | 0.26 | Weak |
| Reviews and average position | -0.32 | Modest, because a lower position number is better |
Reviews show the clearest link to placement. Products with more reviews tend to appear slightly higher in the list. Twelve of the 50 products appear first in at least 40% of their appearances, and their review counts range from 321 to 24,681. e.l.f. Poreless Putty Primer appears first in 85.7% of its appearances with 8,383 reviews.
The top 25 products by visibility have a median of 2,123 reviews, and the next 25 have a median of 1,639. Both groups have a median rating of 4.5. Because every product in this study already ranks near the top, the correlations can understate any effect reviews have across all 3,685 products. These are associations and do not show that reviews cause visibility or that they have no effect.
Does ChatGPT shopping data include reviews in every category?
No. Profound records a star rating and review count for all 50 top cosmetics products, but for only 3 of the top 50 golf products and for none of the top 50 baby gear or skincare products.
| Category | Top products checked | With rating and review count |
|---|---|---|
| Cosmetics | 50 | 50 |
| Golf equipment | 50 | 3 |
| Baby gear and furniture | 50 | 0 |
| Skincare and makeup | 50 | 0 |
For that reason this study reports review statistics for cosmetics only. A single review threshold across all four categories cannot be set from this data.
What do Part 1 and Part 2 show together?
Part 1 shows that a few retailers fill most ChatGPT shopping results, and Part 2 shows that individual products with relatively few recorded reviews can still appear prominently, so brands should monitor both retailer distribution and product-level visibility.
| Study | Question | Main finding |
|---|---|---|
| Part 1: ChatGPT Shopping Index | Which retailers and brands appear in ChatGPT shopping results? | The top three retailers hold 59.3% to 79.5% of results in each of four categories |
| Part 2: ChatGPT Review Study | Do product reviews and ratings relate to visibility? | No consistent minimum review count was observed, and the rank correlation between reviews and visibility is 0.13 |
What should brands do with these findings?
Brands should treat reviews and ratings as one input rather than a prerequisite, and build product pages and feeds that make each product easy to match to a question. The study points to four actions.
- Collect reviews, but do not wait for a count. Three of the top 50 cosmetics products have fewer than 100 reviews, so a small review base did not keep them out of the results.
- Watch the star rating. 49 of the 50 products have a rating of 4.0 or higher, and 36 have 4.5 or higher, but the study did not compare them with products ChatGPT rarely shows, so it cannot show that high ratings drive visibility.
- Check that ratings reach the product listing. Ratings and review counts were recorded for every top cosmetics product and for almost none elsewhere, so brands outside cosmetics should confirm their review data is attached to their product listings.
- List every product variant clearly. Retailer-specific listings, such as the 18 “at Nordstrom” baby gear products, appear as separate products and can each earn visibility.
5W is a PR, creative and AI Visibility agency, and we get you in the chatbox. The team runs consumer PR programs and beauty marketing programs. The full series is on the 5W AI Visibility Index page.
How did 5W build the ChatGPT Review Study?
5W built the study from Profound’s shopping product reports for ChatGPT prompt runs between August 3 and November 1, 2026, and the figures here are 5W’s analysis using Profound’s tools.
| Item | Detail |
|---|---|
| Window | August 3 to November 1, 2026, inclusive |
| Engine and geography | ChatGPT only, United States region only |
| Sample | The 50 most visible products in each of four categories: cosmetics, skincare and makeup, baby gear and furniture, and golf equipment |
| Review analysis | Cosmetics only, because ratings and review counts were recorded for all 50 top cosmetics products but few products elsewhere |
| Review source | Profound reports the rating and review count that AI engines see for each product. Each product record carries one rating, one review count and one retailer product URL, and for the top three cosmetics products the URL points to the Ulta Beauty product page while several merchants are listed as sellers. Profound does not state which retailer’s reviews the count reflects, so the figures are the numbers attached to ChatGPT’s product listing and not totals across all retailers |
| Product counts | 3,685 cosmetics products appear in the default product report. A version of the report that adds merchant offers returns 3,654 products, and the visibility of the top product differs by less than 0.1 points between the two |
| Prompt set | The 1,114 prompts that 5W tracks across the four categories, the same set used in Part 1 of the index |
| Statistics | Median, thresholds and Spearman rank correlation, calculated by 5W on the 50 cosmetics products |
| Brand and listing counts | Counted by 5W from the product names and brands recorded in the top 50 lists |
| Not measured | Sales, price, availability, checkout, or the minimum reviews needed for a product to appear at all |
Four limits apply. The study analyzes only the 50 most visible products in each category, so it cannot set a minimum review count for appearing at all. Correlations on 50 products that already rank highly are likely to understate the effect of reviews across all products. Ratings and review counts are shown as Profound recorded them, which may differ from the numbers a shopper sees on a retailer page. And the study shows associations only, so it cannot establish that reviews or ratings cause a product to appear.
Conclusion
The 5W ChatGPT Review Study found that the cosmetics products ChatGPT shows most have strong ratings but show no consistent minimum review count, that review counts track visibility only weakly, and that rating and review data is recorded for cosmetics far more often than for other categories. The study describes the products ChatGPT shows and does not test what causes a product to appear.
5W is an AI visibility agency that helps companies get into the chatbox. 5W runs AI Search (GEO) programs for brands across consumer, B2B, financial services, healthcare, and technology, building the machine-readable footprint that gets brands cited, not just ranked. Learn more at 5wpr.com/practice/generative-engine-optimization.
Frequently asked questions
What is the 5W ChatGPT Review Study?
The 5W ChatGPT Review Study is 5W’s analysis using Profound’s tools of the star ratings and review counts of the products ChatGPT shows most often in cosmetics shopping results. It is Part 2 of the 5W ChatGPT Shopping Index and covers the 50 most visible products between August 3 and November 1, 2026.
Do products need many reviews to appear in ChatGPT?
No consistent minimum review count was observed among the products ChatGPT already shows. Three of the 50 most visible cosmetics products have fewer than 100 reviews, and 13 have fewer than 500. The study cannot say what minimum applies to products outside the top 50.
Where do ChatGPT product review counts come from?
Profound reports the rating and review count that AI engines see for each product. Each product record carries one rating, one review count and one retailer product URL, and Profound does not state which retailer’s reviews the count reflects. The figures are the numbers attached to ChatGPT’s product listing and not totals across all retailers.
What star rating do ChatGPT’s top beauty picks have?
The median star rating of the 50 most visible cosmetics products is 4.5. Forty-nine of the 50 have a rating of 4.0 or higher, and the lowest rating is 3.8.
Which products does ChatGPT show most?
The most visible product is e.l.f. Cosmetics Power Grip Primer in cosmetics (4.3% of shopping answers), Stokke Tripp Trapp High Chair 2 in baby gear (5.8%), DRMTLGY Brightening Eye Masks in skincare and makeup (12.1%), and Callaway Elyte X Driver in golf (7.9%).
Why does the review analysis cover only cosmetics?
Profound recorded a star rating and review count for all 50 top cosmetics products but for only 3 of the top 50 golf products and none of the top 50 baby gear or skincare products. Cosmetics is the only category with enough review data to analyze.
How does 5W get companies into the chatbox?
5W gets companies into the chatbox by running search, media, technology and PR together, all in house. The search team structures pages, product data and schema so ChatGPT and other AI engines can read them. The media and PR team earns coverage in the publications those engines cite, and the video, Reddit and social teams publish in the other sources they cite. 5W’s proprietary technology tracks where a brand appears in AI answers, so each change can be measured against data like the figures in this study. See 5W’s AI visibility agency practice and its GEO programs.
Does a high review count guarantee ChatGPT visibility?
No. Review counts tracked visibility only weakly in this study, and the study does not test cause. Reviews and ratings are one input among many, and sales still depend on price, availability and checkout.
Does this prove reviews do not affect ChatGPT rankings?
No. The study analyzes only the 50 most visible cosmetics products and shows associations. It found no consistent minimum review count among them, but it did not compare them with products ChatGPT rarely shows, so it cannot show that reviews have no effect.
