By Simon Young
In June, The Golf Business reported on a study by AI Score that tested four UK multi-site golf operators with realistic golfer prompts on ChatGPT and Perplexity. The results were striking.
St Mellion Estate did not appear at all on Perplexity for “best golf club in Cornwall”, while ChatGPT ranked it in the top four with an accurate, positive description. Redlibbets was invisible on Perplexity but came in the top five on ChatGPT. One club was weak on both, and another appeared on both but was ranked last each time.
For a club manager, the lesson is that there is no single AI ranking
to check. So what can a club sensibly do?
I work in answer engine optimisation (AEO), a term coined by Jason Barnard. I am not in the golf industry, so this is written from the outside in. I will try to be clear about what is established and what is still evolving.
What is established
Answers differ between assistants and between prompts. The June study showed it for golf, and anyone can see it by asking two assistants the same question. That means one prompt run is an anecdote, not evidence.
A single check that shows your club, or misses it, tells you very
little.
Assistants can only repeat what they can find. If your visitor policy, society packages and membership categories are buried in a PDF or spread across platforms with conflicting details, an assistant has little reliable to say.
Nobody can guarantee placement. None of the major assistants publishes exactly how it chooses which clubs to name, and as far as I know none sells placement in its answers.
How to monitor
Pick a fixed set of real golfer questions. For illustration only, they
might be: “Where can I play a visitor round near [town] this
weekend?”, “Which clubs near [town] welcome beginners?”, “Where could we hold a society day for 24 players?” and “How much is membership at clubs near [town]?” Your office staff will know the real ones.
Run the same questions once a month across more than one assistant. Log the date, the assistant, whether your club appears, where, and what is said about it. Note anything inaccurate. After a few months you have a trend worth acting on.
What to publish
One plain, dated page in the club’s own words: the course, how
visitors and societies book, green fee guidance, practice facilities,
membership categories, dress code and the clubhouse. Keep it
consistent with your listings and booking platforms.
Specific detail beats adjectives. “Welcoming” is easy to say and hard to repeat with confidence. A clear visitor policy, with days, times and how to book, is something an answer engine can actually repeat.
What is still evolving
How each assistant weighs a club’s own website against reviews, press, directories and booking platforms is not public, and it changes. The study’s own commentary suggested results differ by query type, with branded searches faring better than generic ones like “best club in [area]”. That fits what I would expect, but one study of four operators is a starting point, not a rule. How much enquiry volume comes from AI assistants, rather than Google, is also hard for mostclubs to measure today.
What to realistically expect
Clear, consistent information makes it easier for any assistant to
describe your club accurately. It will not guarantee you are recommended, and changes may take weeks or months to show, if they show at all. The honest measure is the monthly log: is your club appearing more often, and is what is said about it right?
Simon Young is the founder of Question.Marketing
Leave a Reply