Local Business Expansion Network & SEO Improvements — August 11, 2026

Today, I focused on launching a local business expansion network and implementing significant SEO improvements to enhance our spatial AI capabilities. After a long 15-hour day, I shipped eight commits, six of which were new features or improvements, and two fixes. The work was crucial as it aligns with my vision of building a one-man-show capable of achieving a billion-dollar valuation. The local business expansion network is a major step towards this goal, enabling us to tap into a thousand Census cities with drip-controlled, sharded sitemaps.

The most significant feature I implemented was the US business expansion network, which now allows for the integration of 1,000 Census cities into our platform. This will enable businesses to better connect with local markets, and it’s designed to be scalable. I also added a new 'local-expansion' network card that works with our growth-rollout engine, ensuring that we can manage this data efficiently. It’s exciting to see how this will empower local businesses and enhance our operational capabilities.

Another crucial feature was the enrichment of business leaves with real OpenStreetMap (OSM) detail tags and computed distances. This means that thin no-website leaves now carry essential information such as hours, cuisine types, wheelchair access, and more. This data is invaluable for users searching for local businesses, as it provides a richer context about each listing. The computed distances to city landmarks will also help users make more informed decisions based on proximity.

On the SEO front, I tackled some critical issues that had been affecting our Google Search Console (GSC) performance. I added an image to the tag-page Product JSON-LD, which resolved a critical issue with 408 items flagged in GSC. Fixing this problem was essential to ensure that our merchant listings are accurately represented. Additionally, I implemented hashtag inventory sitemaps, creating a full 160,000 enriched-tag universe sharded into 40,000 per file. This will allow us to maintain a small vouched quality tier while also expanding our reach.

However, the day was not without its challenges. I encountered a significant bug when I realized that point of interest (POI) database results were not entering the osm2 cache. This oversight led to uncached probes storming the osm_poi, which I had to address promptly. After diagnosing the issue, I was able to implement a fix that ensured proper caching of POI data, which is crucial for the overall performance of our platform.

In terms of performance, I also focused on improving the root sitemap rendering process. By implementing Incremental Static Regeneration (ISR) for the sitemap on an hourly basis, I managed to reduce the cold render time to approximately 28 seconds. This enhancement is vital as it ensures that our sitemaps are always up-to-date, which is crucial for both user experience and SEO.

Reflecting on the day, I am proud of the progress made, but I am also conscious of the challenges that lie ahead. Building this business solo with the help of AI tools like Claude Code in VS Code has been a rewarding yet demanding experience. I am reminded that every bug fixed and feature shipped brings me closer to my vision of a billion-dollar valuation. Each commit is a stepping stone toward creating a robust spatial AI portal that can truly serve the needs of local businesses and their customers. The grind is real, but so is the ambition, and I am committed to seeing this through.