A city guide where every recommendation is backed by public data
austintexasthings.comEvery city has a hundred 'best brunch in Austin' listicles and no way to tell which of them is a paid placement. Austin Texas Things takes the opposite approach: the same lifestyle content everyone writes, but with components that pull live government data into the page — this restaurant's actual health inspection score, whether the Greenbelt is currently flowing, what the permit activity looks like on that block.
Local content is a commodity and everyone knows it. A guide to Austin coffee shops competes with thousands of identical guides, most of them assembled from the same three sources, and none of them offering a reader a reason to believe them.
The available differentiator was sitting in public data portals that nobody bothers to wire into content: Austin Public Health publishes every restaurant inspection score, USGS publishes real-time creek gauges, the Texas Comptroller publishes mixed-beverage receipts that reveal how busy a bar genuinely is.
The engineering problem is that this data has to appear inside prose without turning every article into a hand-maintained dashboard that goes stale in a month.
Articles are MDX, so a writer drops a component into the prose — an inspection score, a swim-conditions check, nearby development activity — and it renders live at build or request time against the shared Postgres warehouse. The article stays a document; the data stays current without anyone editing the article.
Scheduled jobs pull Austin Public Health inspections, scrape USGS gauges for the three Barton Creek stations, and load FRED economic series covering 25 Texas metros and 252 of the state's 254 counties, feeding both the article components and standalone dashboards.
The site reads from the same Supabase project that backs the construction-data platform, which is why a lifestyle article can show real permit activity near a venue. One warehouse, two products with completely different audiences and front ends.
Texas publishes mixed-beverage receipts per establishment. That turns into a genuinely novel content angle — which bars are actually busy, by revenue, rather than by whoever bought the placement — using data that is public and nobody thinks to use.
The parts that were genuinely hard — and how they were solved.
The economy dashboards and inspection browser are real reference tools with their own routes and charting, not decoration. They give the domain a reason to exist beyond the article funnel and pull an audience that never wanted a brunch recommendation.
The Greenbelt swim check reads current USGS flow and stage. Anyone who has driven to Barton Creek to find it dry understands why that is worth more than a paragraph describing what the creek is usually like.
Inspection scores, receipts, gauges and economic series all carry provenance back to a named government dataset. The whole editorial proposition is that a reader can check the number, so the pipeline preserves the link rather than flattening it into a figure.