Brisbane Restaurant Data - 2026

April 24, 2026

As 2026 rolled around, I realised that my list of Brisbane's best restaurants and subsequent data analysis were slowly becoming outdated and needed a refresh. My website was no longer on the cutting edge of Brisbane's culinary scene. 🥲 I re-acquainted myself with Google's infuriatingly confusing Places API and pulled the data, which I've used to create my list of the 2026 Best Restaurants in Brisbane.

Changes

Query Area

The challenge in this project is that Google's API isn't designed for this. Each query to Google's API has a centre and a radius, and you get the 20 most popular results from in that area. You also only get 1000 free queries each month before Google steals your wallet. I threw my previous inefficient implementation away, and with the help of AI, created a new - significantly more optimal - scanning approach. Shown below are the ~900 queries I used to scan Brisbane.

An image of Brisbane with 900 interecting circles overlaid to cover the entire city.

Some points to note are:

Density of queries increases as you get closer to the city. This is because intuitively, restaurant density is higher in the city.
I've excluded D'aguilar National Park because it is distinctly lacking in restaurants.
I've sparsely queried Moggill and Logan, so they can be included in my list. While Logan may be >15km away, it's very quick to reach with the help of the Pacific Motorway, so if there is something good there, I want to know about it.

Here is a zoomed in image of the CBD, so you can see how many queries cover it.

The same map as above, but zoomed in to show just Brisbane CBD, and how the circles are significantly smaller and more numerous in the city.

Last year I simply queried everything within ~15km at the same resolution. In 2025 I had data for ~2700 resturants, but in 2026 I have ~3480. I think smarter querying contributed heavily to the extra data at a lower cost.