We audited every café and restaurant in Tiong Bahru.
Here’s what we found.
Last week, we did something a little unusual. We pulled every café and restaurant in Tiong Bahru from Google Maps — 115 venues in total — and ran each one through our GEO Readiness Checker.
GEO stands for Generative Engine Optimization. It’s the practice of making your brand visible to AI tools like ChatGPT, Gemini, and Perplexity — not just Google. And right now, most Singapore F&B brands have no idea where they stand.
So we checked. Here’s what we found.
The headline numbers
Of the 115 venues we identified, 56 had a website we could run through the checker. Of those, 44 returned a full GEO score. The average was 62.5 out of 100 — which sounds reasonable until you look at what’s missing.
The bright spots: almost every venue had GA4 installed and scored well on trust signals. The foundations are there. The GEO-specific signals are not.
The finding that surprised us most
We expected big chain brands to dominate. More marketing budget, bigger teams, more resources — surely they’d be better optimised?
Not quite.
When we split the 44 venues into chain brands and local independent brands, the overall average scores were nearly identical — 61.5 for chains versus 63.0 for local brands. But the reasons behind those scores told a very different story.
Chains scored higher on AI competitor visibility purely because their brand names are already embedded in AI training data from years of web presence. That’s not something you can replicate quickly — it accumulates over time.
Local brands outperformed chains on AI comprehension (89% vs 73%), brand consistency (77% vs 64%), and schema markup (53% vs 43%) — the signals you can actually control and implement this week.
The reason is straightforward: local brand owners are closer to their digital presence. They move faster, experiment more, and are more likely to adopt newer formats. Chains move slowly — decisions go through regional HQs, websites are managed globally, and updates take months.
This is the GEO opportunity for Singapore’s independent F&B scene. The playing field isn’t level — it’s tilted in your favour, if you move now.
The three things to fix first
Based on what we found across Tiong Bahru, these are the highest-impact changes any F&B brand can make today:
1. Create an llms.txt file. Place a plain text file at yourdomain.com/llms.txt that describes what your business does, who it’s for, and what content AI can use. This is the single most commonly missing signal — and the easiest to fix. It takes about 30 minutes.
2. Add schema markup. Implement a LocalBusiness or Restaurant schema on your homepage using JSON-LD. This tells AI systems exactly what you are, where you’re located, and what you offer. Without it, AI has to infer — and inference is unreliable.
3. Build a FAQ page. Write answers to the 8–10 questions your customers actually ask: What are your opening hours? Do you take reservations? What’s your most popular dish? Is it halal? AI tools actively look for question-and-answer content when building responses about local businesses.
None of these require a developer or a big budget. They require about a day of focused work — and they’ll compound over time as AI search grows.
What’s next
Tiong Bahru is just the beginning. We’re planning to expand this audit across Singapore’s key lifestyle and F&B neighbourhoods — Dempsey, Holland Village, Orchard, Tanjong Pagar — and publish a full Singapore F&B GEO Readiness Index later this year.
Is your brand AI-visible?
Run a free GEO Readiness Check and get a full report in 2 minutes — with a prioritised action list to improve your score.
Check your AI visibility →No credit card required · 2 min to complete
The brands that get this right now will have a head start that’s very hard to close later. The window is open — but it won’t stay open forever.