While building a client page on Friday night, a “map that didn’t look like a map” pulled me in
At 10 p.m. last Friday, I was redesigning the introduction page for a coffee shop. The owner wanted something “a bit special,” but didn’t want to pay more for design. I was staring at a pixel map of the San Francisco Bay Area when it hit me: local clients may not need something prettier. They may need a map that lets someone instantly recognize “This is where I live, and this is where the shop is.”
The project turned 1,300 square miles of terrain into an explorable pixel scene, including Apple Park, the Palace of Fine Arts, and UC Berkeley. I’ve gotten stuck here before too: I kept treating “designed” as “more polished,” which only pushed the cost higher.
It’s not an ordinary map—it turns the city into content people can explore
Developer Justin Chavez transformed the San Francisco Bay Area into a consistent pixel-art world that people can explore. For local businesses, it is more memorable than a standard map: they can adapt the visual style for neighborhood maps, store-area maps, or event routes. For anyone building a personal IP, their city can become a long-term content platform.
Justin also documented the process and cost estimate while completing the San Francisco map. The part worth copying isn’t the entire model-training workflow. It’s the idea: “Choose one familiar area, then turn it into a map people want to share.”
To replicate this today, I could start for $0 and half a day
Money: Start with one small neighborhood. Tool and asset costs can stay at $0. If I outsource or buy ready-made map assets, I’d usually budget 300—1,000 yuan.
Time: Organizing the area and selecting 5—10 locations takes roughly 4—8 hours.
Technical barrier: Coding isn’t required, but I need to handle image dimensions, perspective, and labels.
First step: Open Canva, Photopea, or Figma, search for a “pixel map template,” and choose a base map for my own neighborhood. If I’m willing to experiment for another two weeks, I could use an image-generation tool to establish a consistent style. But I wouldn’t train a model for the first map. I dove straight into technical details once too, ended up with neither a finished map nor a published post.
At different stages, here’s how I’d choose the approach
If I’m just starting, I’d map the two streets I pass every day. A $0, half-day map paired with a post titled “The 3 Best Places Near My Home” could test whether anyone is interested.
If I already have one or two clients, I’d turn it into a custom product: 500—2,000 yuan per version, starting by asking whether they need a store-area map, event route, or gift guide.
If I’m scaling, I’d create a unified template that designers or assistants can replicate across neighborhoods, while keeping records of data sources and licenses.
This path isn’t for everyone. If I don’t currently have content to publish or a client need, it’s fine to wait. Testing with one neighborhood is much safer than building an entire city in one go.