Citation in ChatGPT Search
Ranking in Bing
Ranking in Google
https://www.nike.com/gb/w/sale-air-max-shoes-3yaepza6d8hzy7ok
1
1
https://www.sportsdirect.com/nike/nike-air-max
5
6
https://www.jdsports.co.uk/collection/nike-air-max/sale/p/trainers/
Not in the top 100
Not in the top 100
https://www.amazon.co.uk/s?k=nike+air+max+trainers+men+size+10
Not in the top 100
4
https://www.lovethesales.com/nike-air-max-sale
Not in the top 100
11
This query was another instance of Google ranking URLs for the domain, e.g., JDSports, for the query in a good position – but not the URL cited by ChatGPT Search.
Travel Searches
“Best Holiday Destinations 2025”
Much like a number of queries in the travel sector, this triggered an image pack above the carousel.
Image from author, November 2024
Having worked with a large number of travel companies and supported the industry through events and the Institute of Travel & Tourism, I can honestly say this is a very eclectic list of destinations to consider.
The citations are very publisher-heavy, with other travel companies (and travel blogs) in the search results section.
Citation in ChatGPT Search
Ranking in Bing
Ranking in Google
https://www.independent.co.uk/travel/news-and-advice/lonely-planet-best-destinations-2025-b2633986.html
2
Not in top 100
https://www.lonelyplanet.com/best-in-travel
1
11
https://www.thesun.co.uk/travel/31237176/lonely-planet-top-2025-english-hotspot/
Not in the top 100
52
https://www.thesun.ie/travel/13945757/european-destination-named-top-holiday-spot-2025/
Not in the top 100
Not in top 100
https://www.thescottishsun.co.uk/travel/13767687/african-holiday-destination-big-tui-hot-jet-lag/
Not in the top 100
Not in top 100
“Luxury Hotels In Napa Valley”
To my surprise, this query triggered the SearchGPT Map.
Image from author, November 2024
The Telegraph.co.uk result was referenced twice for two different hotels in the SearchGPT Map Pack.
Citation in ChatGPT Search
Ranking in Bing
Ranking in Google
https://www.vogue.com/article/bardessono-hotel-and-spa
Not in the top 100
Not in the top 100
https://www.telegraph.co.uk/travel/destinations/north-america/united-states/california/napa-county/napa-valley/hotels/
6
Not in the top 100
3
Not in the top 100
https://www.vogue.com/article/halehouse-spa-at-stanly-ranch
Not in the top 100
Not in the top 100
https://www.telegraph.co.uk/travel/destinations/north-america/united-states/california/napa-county/napa-valley/hotels/
6 (Duplicate URL)
Not in the top 100
Reviewing the recommendations and sources, there doesn’t appear to be a logical connection as to why these sources and (specifically) these locations were chosen.
Google’s hotel feature is a lot more comprehensive and interactive, so I don’t see this threatening the hotel search journey any time soon.
“Abu Dhabi Grand Prix”
Sports tourism is growing in popularity, and F1 is a globally popular sport.
Searching for specific races is also a tricky query, which is what I wanted to test out.
Searching for “[location] grand prix” has a number of common interpretations. It could be an informational search, a navigational search, or with transactional intent and wanting to research ticket prices.
Image from author, November 2024
I feel like SearchGPT understood this, as the five websites referenced as citations cover these bases quite well, and it doesn’t lean into being too heavily informational or transactional.
This is a similar result to any Grand Prix query, e.g., Bahrain GP, Belgium GP, Singapore GP.
Citation in ChatGPT Search
Ranking in Bing
Ranking in Google
https://www.formula1.com/en/latest/article/fia-and-formula-1-announces-calendar-for-2025.48ii9hOMGxuOJnjLgpA5qS
Not in the top 100
Not in the top 100
https://f1experiences.com/2025-abu-dhabi-grand-prix
8
13
https://www.p1travel.com/en-GB/motorsports/formula-1/abu-dhabi-gp-paddock-club-2025**
Not in the top 100
Not in the top 100
https://www.etihad.com/en-gb/abu-dhabi/formula-1/formula-1-abu-dhabi-grand-prix-packages
4
Not in the top 100
https://www.bbc.co.uk/sport/formula1/67537239
11
Not in the top 100
From the list of referenced webpages, the **P1Travel URL posed an anomaly across all of the searches I’ve done to try and break SearchGPT. The URL returns a 404:
Screenshot from p1travel.com, November 2024
Travel SEO has been evolving for a number of years now, and in my opinion, LLM platforms pose the greatest potential to disrupt this industry more consistently than others.
Other Query Observations
Over the past 48 hours, I’ve made notes of queries made throughout the day through Google and repeated them on Bing and SearchGPT.
“England V Samoa Score”
Despite being at the match itself, what triggered me to look at this query was my wife commenting on the score – and the score she quoted was wrong.
So, I replicated her search on Google and found the below result:
Screenshot for search for [england v samoa score], Google, November 2024While the “Top Stories” section does bring accurate results, Google’s feature at the top of the SERP brings the result of a game over a year ago – in a different sport.
Now, I can understand why Google is falling down here.
Rugby Union is the second most popular sport in the UK, and Rugby League is the 11th most popular. So, choosing to show this sport over the other would be acknowledging a greater common interpretation of the query, and looking to cater to a larger potential search audience.
I then wanted to see how ChatGPT Search fared with the same query:
Screenshot of search for [england v samoa score], ChatGPT Search, November 2024While it isn’t as visually prominent, the generative text cited from the BBC contains the correct scoreline as well as other factually correct information.
The downfall then comes that the news stories being pulled through relate to a different game (that happened the previous week).
As I reviewed all of the citations provided by ChatGPT Search, I noticed that it pulls through news stories relating to the same incorrect match that Google cited prominently in the SERP feature.
But the AI has been able to distinguish the two and provide a timely result, potentially showing a greater understanding of queries that may be time-sensitive and the concept of QDF against leaning towards the “more common” dominant interpretation.
As Bing plays a prominent role in SearchGPT, it’s no surprise to see Bing also prioritizing the more recent incident of England vs. Samoa through the Bing News feature, although Bing does then feature the same match from over a year ago in the special sports feature.
Screenshot of search for [england v samoa score], Bing, November 2024
“Manchester United v Chelsea”
To test true timeliness around live events, I’ve performed (and compared) this search query during the live soccer game.
I actually found the ChatGPT Search result to be the most informational and useful – outside of reporting the live score.
Screenshot of search for [manchester united v chelsea], ChatGPT, November 2024The ChatGPT overview provides a lot of information in a single snippet – including broadcast information.
If you search for broadcast information on Google and Bing, you don’t get this depth of information directly in the results – and unfortunately, a lot of publishers bury this information deep in an article, even if that article is written (and headlined) to target the query specifically.
To see if ChatGPT Search would show the in-game score, I tried repeatedly with variations of “Manchester United v Chelsea live score|in game score|current score|score now.”
But despite Bing showing the live score in a SERP feature and a number of sources being accessed by ChatGPT that contain the information, all it would give was variations of the below response – refreshingly asking the user to visit the source website for the information.
Image from author, November 2024
“Cheese Not Sticking To Pizza”
This was a prominent query that Google’s AI Overviews fell down on earlier this year, drawing scrutiny from the mass media.
Image from author, November 2024
I’ve tested a number of the queries that tripped Google up, and so far, SearchGPT seems to avoid hallucinations and confusing results.
Do I Think This Will Disrupt Google?
Google has been facing disruption since the mass market’s relationship with mobile devices changed.
Moving away from desktop and utilizing apps and browsers to research, shop, and be entertained opened the doors for new platforms to compete for screen time and attention.
On a recent podcast I ran, John Mueller said that when Google first started to push towards mobile-first, a lot of SEO pros and webmasters were skeptical and vocal that users wouldn’t shift from mobile to desktop to view their websites.
SEO as a practice is only 30 (or so) years old – and in 30 years, how users interact with businesses has changed drastically.
AI platforms, such as SearchGPT, will disrupt Google as adoption increases and the “early majority” of consumers start to use LLM platforms daily for tasks they otherwise would have engaged the wider Google ecosystem with.
We can draw some parallels to the Model T, and when Henry Ford introduced the car, it was seen as a novelty and not for the masses – with people seeing them as “faster horses” than a revolutionary mode of transport.
As SEO professionals, we can see the “revolution” part of it in motion, but we must remember that to the mass market, a lot of AI is merely gimmicks such as removing things and people from the background of images on phones or using things like Circle Search.
SearchGPT will contribute to the overall disruption, and it might pioneer the “what’s possible, “much like Jeeves was a pioneer in mass market internet search engine adoption.
Also, I don’t feel this would be a complete answer if I didn’t ask SearchGPT directly if it will disrupt Google:
Image from author, November 2024
Which Search Is Best?
In testing SearchGPT felt as though it had a better understanding around time sensitive queries and used more recent sources.
Removing visual experience from the equation, SearchGPT is better as objectively you want recent content when you’re searching for information around holidays, gifts, products for “now,” and not viewpoints from over a year ago.
For local search, ChatGPT Search still has a long way to go and falls short of were Google and Bing are.
In my opinion, information was better in ChatGPT Search than in Google and Bing, but not on all queries. Overall I still feel it is behind Google’s AI Overviews and Bing’s Generative Search in terms of answer queries behind text-driven informational results – and the ChatGPT Search interface is a little bit too detached from what we’re used to seeing from a “Search engine”.
Closing Thoughts
Brett Tabke has referred to SearchGPT as the “final” search engine, and I agree with this.
While SearchGPT currently offers certain advantages over Google and traditional search models, it is still a work in progress.
However, its potential trajectory is one of the key insights to take away from SearchGPT.
This development indicates that user interaction with the internet is evolving once again, and as SEO pros, we must adapt as an industry.
The rise of platforms like TikTok highlights the shift toward multi-modal search, urging us to move beyond a narrow focus on the “engine” aspect of our work.
The core principles of SEO, particularly technical SEO, will always remain relevant, as AI crawlers must crawl, identify, and access content similarly to traditional search engine crawlers.
I believe it’s time for us to adjust our SEO terminology, shifting the conversation from ranking to referencing, and realigning what the overall success of an SEO campaign is.
And we finally start taking credit for upper funnel visibility and touchpoints and are not measured solely on bottom-of-the-funnel conversions.
Rather than asking the questions “How do we optimize for SearchGPT?” or “How do we optimize for LLMs?,” we need to be looking at our market and the different LLM platforms and asking questions like:
How does our audience interact with SearchGPT, and how will SearchGPT stop our audience leaving the ChatGPT ecosystem?
What information does the user get if they stay in this ecosystem versus what they would receive “traditionally”?
When (and if) the user leaves this ecosystem, what is their new entry point in our perceived funnel? How well adapted is our funnel for new-to-brand (NTB) touchpoints at this stage?
More resources:
Featured Image: Roman Samborskyi/Shutterstock