Bing, not Google, forms the brand recommended by ChatGPT

In this tutorial, we’re going deep instead of wide. We are focused on one question: why is the product not in one ChatGPT notification out of ~70 times?

We chose one command: “What are the best hotels in New York City?”

We analyzed mentions, citations, fanouts, and SERPs on Google and Bing. We also ran a GPT memory analysis, but it made no discernible difference in mentions, citations, or fanouts.

What we did and what we found

We chose NYC hotels because it is a crowded, mature market with juggernauts and up-and-comers. And we have no connection to a luxury NYC hotel location – we purposely chose a place where we can live purposefully and learn from the ground up.

After running the command “what are the best hotels in New York City” 68 times, we identified which hotels consistently appeared and which were almost invisible.

We chose the Baccarat Hotel as our “client” because it only appeared once (1.5% of the time), despite strong reviews and clear alignment with the information’s purpose. We wanted to know why – and whether it could change that.

Key findings:

• You can dominate query fanouts in Google SERPs and drive performance under the ChatGPT brand.
• Bing is very important. The ranking of Bing articles for fanouts is very directly related to the mentioned ChatGPT — not just quotes.
• With a vertical dominated by third-party content, you face sophisticated digital PR techniques to increase visibility.

Note: A complete list of methodology appears in the appendix.

Speaking of Baccarat vs. The Fifth Avenue Hotel shows just how wide the difference in ChatGPT visibility can be.

Baccarat Hotel appeared once in 68 trials (1.5%).

The main players were large luxury hotels such as the Four Seasons Hotel New York Downtown.

ChatGPT also identified boutique hotels as a subcategory, generating a second list from its responses. Boutique hotels like Baccarat are usually small and not part of large chains.

Within this boutique category, Baccarat is still a bit active. The Fifth Avenue Hotel, an upscale shopping complex, appeared 13 times, citing 20% ​​of the time, compared to Baccarat’s 1.5%.

Shadow cannot explain the difference in appearance

We first checked to see if there was anything in the hotel’s history or reputation that could explain the gap. As the chart below shows, nothing significant has been done:

Overall, Baccarat has been around for a long time and has more reviews. For quality, Fifth Avenue Hotel has no limit on Google reviews and only a small limit on Expedia reviews. The only area where Baccarat lags is price — but that’s unlikely to be a problem when the Ritz-Carlton, a consistent non-boutique winner, is listed at $1,100.

Further reinforcing the understated status of Fifth Avenue: one of its top Google results (level 2) was a Wikipedia page for a separate Fifth Avenue hotel that closed in 1908, creating a possible business confusion similar to the two Danny Goodwins.

If the Fifth Avenue Hotel was the one to lose, it would suggest an unfounded brand with business confusion. But the opposite happened – it was successful in ChatGPT.

So what was the problem with the Baccarat Hotel?

Winning Google SERPs from query fanouts doesn’t help, but winning Bing SERPs does

When ChatGPT performs a web search, it sends a series of queries that you can debug with Chrome DevTools. In this lesson, examples include:

• [Best hotels in new york city]
• [Top rated luxury hotels in new york city recommendations]
• [Best hotels in nyc top luxury and boutique hotels new york]
• [Best luxury and boutique hotels in new york city recommendations reviews]
• [Best hotels in new york city nyc top hotels]
• [Top hotels in nyc luxury boutique best places to stay new york city]

In total, we released 25 unique quiz fanouts.

What we saw in Google’s SERPs

If we only look at the articles that dominate the fanout SERPs in Google, we can expect that Baccarat will slightly surpass Fifth Avenue in ChatGPT. That didn’t happen.

In the table below, Baccarat “wins” three of the top 10 pages that appear the most, while Fifth Avenue Hotel “wins” two. The other five do not include anything. “Win” means any of the following:

• Appearing when the other is invisible.
• It appears at the top of the page.
• Having a great feeling.

Data:

What we saw in Bing’s SERPs

Conversely, if we only looked at the articles dominating the fanout SERPs on Bing, we would expect Fifth Avenue to surpass Baccarat on ChatGPT — and it did.

In the table below, Fifth Avenue “wins” five of the eight URLs that appear the most.

Note: The table includes two fewer URLs because the Bing SERPs were slightly different for these fanouts.

Data:

Correlation between Bing visibility and product mentions

Bing ranking strongly predicts ChatGPT citations — 87% match top Bing results, Seer Interactive found. Our case study supports this and extends it.

We examined the relationship between fanouts (Viewer focus on awareness) and brand mentions.

Example quote: “For a luxury boutique experience: listings like the Fifth Avenue Hotel or the Crosby Street Hotel regularly make ‘top NYC’ lists from travel planners.”

Speaking is often more important than quotes. Most people will not follow the quotes but will remember the top recommendations.

There is an ongoing debate about whether fanouts shape the responses and chatter of ChatGPT, or whether they support the responses generated from the training data. For example, Leigh McKenzie argued on LinkedIn:

• “The quotes you see below? Those are displayed after the answer is generated, not before. It’s an afterthought. The model didn’t choose your brand because it found your URL. It generated the answer based on what it already knew, and pointed to the sources that support it.”

In contrast, our data coincides with Beehiiv’s research, which suggests that quotations create implied conditions.

Training data does not seem to be a problem for Baccarat. Compared to Fifth Avenue, it’s older, has more reviews, and has the same high ratings in all major arenas. It lacks a strong presence in Bing results for fanouts and citations, which seem to lead to fewer mentions.

A simple flow might look like this:

• Brand rankings on Bing → ChatGPT fanouts pulls Bing pages → ChatGPT integrates training and Bing data for mentions

Coda: A tale of two Forbes articles, or why details matter

In this vertical, third parties such as Forbes and Condé Nast dominate the space. Visibility depends on who is talking about you, so you need a solid communication strategy – not just updates to your content.

Our data shows that “targeting Forbes” is not specific enough.

The top result that appeared on both Bing and ChatGPT was the same Forbes article. On Google, the most common fanout result was also a Forbes article – but a different one.

As we have seen, entering a Forbes article on Google is unlikely to provide a meaningful improvement. Baccarat “wins” on that piece.

Binging into Forbes’ article, where Baccarat is not mentioned, can make a difference. This requires a sophisticated surgical approach based on Bing data.

Normal will not work; details rule.

Appendix: Methodology

Model: We told GPT-5.2 the results are fast and manually extracted. We have not implemented APIs within ChatGPT.

Number of repetitions: We ran the same information 68 times.

Notify: “What are the best hotels in New York City?”

Settings: We tested three memory states:

• Saved memories are disabled
• Saved memories are unlocked, using actual unrelated user memories
• The saved memories are open, with one memory about needing gluten-free places to go

In all experiments, we turned off the “reference conversation history” to avoid interference in the repetition.

We expected differences based on memory settings but did not find them, so we treated all tests as a single dataset.

What we released:

• All questions for fanouts.
• Full text output of ChatGPT.
• Quotes.
• Google SERPs for all fanouts.
• Bing SERPs are all fanouts.