# ChatGPT vs Google: same winner, different shortlist

> ChatGPT and Google AI Overviews put the same brand first in 93% of 1,655 buyer categories in Parse's October 19, 2025–April 25, 2026 sample, while sharing just 0.25 of their wider shortlists. This compares category totals, not individual answers.

HTML: https://parse.gl/research/chatgpt-vs-google-same-winner-different-shortlist · Site index: https://parse.gl/llms.txt · Published: 2026-06-30 · Updated: 2026-10-12

Two AI engines often name the same category leader while building very different shortlists underneath. How much do ChatGPT and Google AI Overviews agree at #1 compared with the rest of the list?

We aggregated recommendations from ChatGPT and Google AI Overviews between October 2025 and April 2026 across 1,655 categories with enough observations on both engines. The comparison is between category-level leaders and shortlists, not pairs of individual answers.

## Key figures

- **of categories: same #1 brand on both engines:** 93%
- **mean shortlist overlap:** 0.25
- **matched buyer categories:** 1,655 (48,302 brands)
- **brands named per category (ChatGPT vs Google):** 23.9 vs 36.4

## The engines agree on the winner 93% of the time

Across 1,655 buyer categories, ChatGPT and Google AI Overviews had the same leading brand 93% of the time. Individual answers can still differ; this study compares leaders across repeated answers in each category.

That is the agreement that matters. The brand a buyer is most likely to walk away with is the same on both engines in more than nine categories out of ten.

- **of categories share the same #1 brand on both engines:** 93.0%

> The disagreement isn't about who wins. The two engines pick the same leader almost every time.

## But the shortlists barely overlap

Agreement collapses the moment you look past the winner. Treat each engine's recommended brands as a list and the mean overlap between them is just 0.25, roughly a quarter of the names in common.

Same winner, different rest of the list. The runner-up, the third name, the trailing also-recommended brands: those are where the two engines go their own way.

- **mean shortlist overlap between the two engines:** 0.25

> Winning the #1 slot mostly transfers across engines. Making the rest of the shortlist does not.

## Same winner, different field

The engines agree on #1 in 93% of categories, but overlap across the full recommended lists is only 0.25. They share the leader, not the rest of the shortlist.

**Where the engines agree, and where they don't**

| Item | Value |
| --- | --- |
| Agree on #1 brand | 93.0% |
| Shortlist overlap | 0.25 |

_Agreement on the single #1 brand versus overlap across the whole recommended shortlist._

## Google names about 50% more brands

The reason the shortlists diverge is partly that they are different lengths. Google AI Overviews names a median of 36.4 brands per category; ChatGPT names 23.9. Google's answer is a broader marketplace of options, which leaves far more room for names ChatGPT never reaches.

| Item | Value |
| --- | --- |
| Google AI Overviews | 36.4 |
| ChatGPT | 23.9 |

_Median number of distinct brands recommended per category._

## ChatGPT is the decisive engine; Google spreads the credit

Breadth has a trade-off. ChatGPT's #1 brand carries 53.3% of all the recommendations in its category; it concentrates on a winner. Google's #1 holds only 41.5%, spreading its recommendations across the longer list. ChatGPT picks one brand; Google spreads across many.

| Item | Value |
| --- | --- |
| ChatGPT #1 | 53.3% |
| Google #1 | 41.5% |

_Share of a category's recommendations captured by each engine's own #1 brand._

## When they split on #1, the split runs mostly one way

In the 7% of categories (116 of them) where the engines crown different winners, the disagreement runs mostly one way. Google's #1 brand is missing from ChatGPT's narrower list 14.7% of the time. ChatGPT's #1 brand is missing from Google's longer list only 1.7% of the time.

In other words, Google almost always already contains ChatGPT's pick somewhere in its longer list, but ChatGPT frequently doesn't have room for Google's. The difference comes from list length, and it means a Google-favored brand can be invisible on ChatGPT entirely.

| Item | Value |
| --- | --- |
| Google's #1 absent from ChatGPT | 14.7% |
| ChatGPT's #1 absent from Google | 1.7% |

_Among the 116 split categories: how often each engine's #1 is absent from the other's list._

> A brand Google loves can be missing from ChatGPT outright. The reverse almost never happens.

## The headline split: FanDuel vs DraftKings

The cleanest example of a genuine split is Daily Fantasy Sports, where the two engines crown opposite household names in a near tie. ChatGPT's #1 is FanDuel, with 32% of its recommendations in the category. Google's #1 is DraftKings, with 33%. Two engines, two famous brands, one percentage point apart.

| Engine | #1 brand | Share of recommendations |
| --- | --- | --- |
| ChatGPT | FanDuel | 32% |
| Google AI Overviews | DraftKings | 33% |

_Daily Fantasy Sports: each engine's #1 brand and its share of that engine's recommendations._

## More clean splits, including Google backing its own product

The same pattern recurs across categories where two credible brands are close enough that each engine lands on a different one. Note the last row: in Business VoIP, Google's #1 pick is its own product, Google Voice, a reminder that an engine's defaults can include the house brand.

| Category | ChatGPT #1 | Google #1 |
| --- | --- | --- |
| Daily Fantasy Sports | FanDuel | DraftKings |
| Meal Kit Delivery | Home Chef | Blue Apron |
| Online Couples Therapy | Talkspace | ReGain |
| Continuous Glucose Monitors | Dexcom | Veri |
| Business VoIP | eVoice | Google Voice |

_Categories where ChatGPT and Google crown different #1 brands._

## The most contested industries are where the budgets are

Sort categories by how often the engines agree on #1 and the bottom of the list is striking: it is data, martech, and AI, the highest-budget software categories. These are crowded fields with no single runaway leader, so the engines have the most room to differ on the winner.

| Industry | Categories | Agree on #1 |
| --- | --- | --- |
| Data & Analytics | 50 | 84.0% |
| Content & Publishing | 26 | 84.6% |
| Sales & Marketing | 36 | 86.1% |
| Administrative Services | 39 | 87.2% |
| Internet Services | 55 | 89.1% |
| Artificial Intelligence | 67 | 89.6% |

_Lowest #1-agreement industries. Click a column to sort._

> The shortlist disagreement is widest exactly where buying budgets are biggest.

## Where one brand dominates, there's nothing to argue about

At the other end, several industries hit essentially total agreement on #1. Consumer Electronics, Gaming, Clothing & Apparel, Energy, and Science & Engineering all reach about 100%; Sports is 97.7%. These are categories with one obvious dominant brand, which leaves the engines no room to disagree about the winner.

| Industry | Categories | Agree on #1 |
| --- | --- | --- |
| Consumer Electronics | 41 | 100% |
| Gaming | 17 | 100% |
| Clothing & Apparel | 17 | 100% |
| Energy | 22 | 100% |
| Science & Engineering | 17 | 100% |
| Sports | 43 | 97.7% |

_Highest #1-agreement industries._

## Each engine has its own house favorites

Some brands win the #1 slot across many unrelated categories: the engine's defaults. ChatGPT's most frequent winner is edX, crowned #1 in 11 categories, followed by Walmart (7) and Zapier (6).

Google's defaults are a completely different set: Zoom (8 categories), Wiz (7), eBay (6), and WordPress (6). Almost no overlap. Each engine has its own default answer for which brand to name first.

| Engine | Brand | Categories won |
| --- | --- | --- |
| ChatGPT | edX | 11 |
| ChatGPT | Walmart | 7 |
| ChatGPT | Zapier | 6 |
| Google | Zoom | 8 |
| Google | Wiz | 7 |
| Google | eBay | 6 |
| Google | WordPress | 6 |

_Brands crowned #1 across the most categories, by engine._

## The fragmentation is structural, not noise

This is not a side effect of a small sample. The pattern holds across all 1,655 categories and roughly 100,000 recommendations, and it concentrates exactly where stakes are highest. Same winner, divergent shortlist, widest in data and martech and AI: that is a stable structural feature of how the two engines build an answer, not a quirk of any one query.

- **Categories analyzed:** 1,655
- **Distinct brands:** 48,302
- **Same #1 / shortlist overlap:** 93% / 0.25

## What this means for AI visibility

Winning a category's #1 slot mostly carries across both engines: the two agree on the leader in 93% of categories, so a brand that earns the top of one usually leads the other. That's the single-winner game, and it's sharpest on ChatGPT, which puts 53% of its recommendations behind its #1 pick.

The rest of the shortlist is engine-specific. Because Google names about 50% more brands per category, runner-up visibility lives disproportionately on Google. If you aren't the category leader yet, Google's longer list gives you far more room to show up.

Split your strategy accordingly: compete for the decisive #1 where one brand owns the answer, and build shortlist presence on Google where the field stays wide. They're different games on different engines.

Related: [Compare competitors across ChatGPT and Google AI](https://parse.gl/use-cases/compare-competitors) · [Research who AI recommends in your market](https://parse.gl/use-cases/research-a-market)

## Why 93% agreement doesn't predict the next answer

That 93% figure compares the most frequently recommended brand in each category after pooling the full study window. It doesn't compare two individual answers. Both engines can put the same brand first in category totals while naming very different brands in any single response.

For example, Brand A can lead both engines over a month even when each engine names it in fewer than half of its answers. To check whether your brand is reliably visible, look at its appearance rate and repeat-run history alongside its aggregate rank. Keep Google AI Overviews separate from Google AI Mode when comparing this earlier study with newer reports.

## How we measured this

We took every buyer category that drew at least 20 distinct brand recommendations on both ChatGPT and Google AI Overviews over the window from October 19, 2025 to April 25, 2026: 1,655 matched categories, meaning categories measured on both engines, spanning 48,302 distinct brands. Recommendations were counted once per answer, brand, and need, so that a brand repeated within a single answer counts once. For each category we ranked each engine's recommended brands by frequency, compared the two #1 brands, and measured overlap across the full recommended lists.

Agreement on #1 means both engines' top-ranked brand is the same. Shortlist overlap is the Jaccard index of the two brand lists: the share of brands common to both, out of all brands either engine named. Figures describe a measured sample of answers from the two engines in scope over the stated window; they are not a claim about Perplexity, Gemini, Copilot, or current live state. Some brand strings in the underlying answers are product-SKU level and were cleaned to brand names for display.

Related: [Why one AI visibility score is misleading](https://parse.gl/blog/why-one-ai-visibility-score-is-misleading)

## Method

We compared the brands ChatGPT and Google AI Overviews recommended across 1,655 buyer categories that drew at least 20 distinct recommendations on both engines between October 19, 2025 and April 25, 2026, spanning 48,302 distinct brands.

## Data

- [Dataset CSV](https://parse.gl/research/chatgpt-vs-google-same-winner-different-shortlist/dataset.csv): Headline metrics behind every figure in this report.

## Sources

- [AI platforms diverge on brand recommendations, BrightEdge](https://www.brightedge.com/news)
- [Google AI Overviews and ChatGPT surface brands differently inside category answers, Search Engine Land](https://searchengineland.com)

## For agents

- [Study dataset (CSV)](https://parse.gl/research/chatgpt-vs-google-same-winner-different-shortlist/dataset.csv)
- Any brand's shares and rivals, with 95% ranges: `https://agent.parse.gl/v1/tools/check_brand?subject=%3Cbrand-domain%3E`
- A brand's shares by topic, with 95% ranges: `https://agent.parse.gl/v1/tools/query_metrics?args=%7B%22measures%22%3A%5B%22share_recommended%22%2C%22share_named%22%5D%2C%22subject%22%3A%22%3Cbrd_id%3E%22%2C%22by%22%3A%5B%22topic%22%5D%7D`
- Search the index: `https://parse.gl/api/public/search?q=<words>`

Reuse: Figures and tables on this page are free to republish with attribution: a link to https://parse.gl/research/chatgpt-vs-google-same-winner-different-shortlist beside them. A standard followed link (dofollow) is preferred.

- No account: Public figures and up to 5 examples per list, about 120 brands a day per address.
- Free account: $5 in credit, your brand's first check (20 buyer questions) and a free weekly check.
- After your first top-up: Exports and deeper search results. The first top-up starts at $5.
- Live checks: $0.01 per question per engine ($0.05 on Gemini and Claude).
- Full public answers and pages: 25 free per day, then $0.005 each.
- Public answers older than 90 days: $0.001 each the first time your workspace counts or lists one, then yours.
- An agent can save work before signup: `POST https://agent.parse.gl/v1/workspaces`
- Connect an assistant: `https://agent.parse.gl/mcp`

[Prices](https://parse.gl/pricing.md) · [Map of the site](https://parse.gl/llms.txt)
