Loplat DOOH Audience Intelligence

The description states that Loplat DOOH Audience Intelligence is a deployed geospatial product connecting 126 real digital billboards in Seoul with cell-level footfall and demographic insights. The author claims the project was built using OpenAI Codex, which assisted in data engineering, statistical modeling, validation, product design, security, and deployment — all by a single CEO over 7–10 days. The system supports audience analysis and media planning but does not claim direct campaign attribution.

This is a self-reported account of a proof-of-concept product built with AI assistance. No evidence of revenue, customers, or traction beyond the author’s own description is provided. The most important open question is whether this represents a scalable commercial offering or remains an experimental prototype.

The description states that Loplat DOOH Audience Intelligence is a deployed geospatial product that connects 126 real digital billboards in Seoul with cell-level footfall and demographic insights. Users can select a billboard, inspect its image and media details, and choose nearby H3 cells to analyze:

• Hourly weekday and weekend footfall
• Estimated age and gender distributions
• Combined audiences across multiple selected cells
• Billboard dimensions, street context, and estimated viewing direction

The product serves approximately 336,000 weekday footfall cells, 323,000 weekend cells, and 55,000 demographic cells per period.

The description states that the project began as a response to a real business question: whether Loplat’s mobility data could help solve the audience measurement gap in digital out-of-home advertising. It positions itself as a tool for media planning and audience analysis rather than campaign attribution.

It evolved from an initial feasibility study into a full product, integrating footfall and demographic data into one interface with real billboard imagery and viewing context. The author notes that Codex helped accelerate development across multiple domains without losing the original business objective.

The description states that the target users are advertisers and media owners who need better evidence about who is present around a screen, when footfall peaks, and how audiences change over time. It supports audience analysis and media planning rather than direct campaign attribution.

No explicit customer segments or personas are defined beyond this general use case.

Not evidenced. The description does not state any pricing model, monetization strategy, or business model details.

The description states that the product was built using GPT-5.6 (Codex) and involved:

• BigQuery extraction and filtering
• H3 footfall and demographic modeling
• Automated and human validation
• Protected aggregation API
• Interactive dashboard
• Browser QA and Cloud Run deployment

It also mentions challenges such as distinguishing presence from movement, estimating demographics responsibly, and combining spatial resolutions. The system maps H3 resolution 11 cells for footfall and H3 resolution 10 for demographics.

The description states that the product is deployed and serves approximately 336,000 weekday footfall cells, 323,000 weekend cells, and 55,000 demographic cells per period. It was built in 7–10 days by one CEO with Codex assistance.

No evidence of revenue, customer adoption, or usage metrics beyond the author’s own account is provided.

Not evidenced. The description does not mention competitors or market positioning relative to other solutions in the digital out-of-home (DOOH) audience intelligence space.

• The product is described as built by a single person using AI tools, raising questions about scalability and institutional knowledge.
• No evidence of revenue, customers, or traction beyond the author’s own description.
• The system uses raw mobility data from Loplat, which may raise privacy or data governance concerns if not properly anonymized or governed.
• The use of Codex for full-stack development raises uncertainty around quality control, reproducibility, and long-term maintainability.

1. What is the source of the mobility data used? Is it anonymized and compliant with privacy regulations?
2. How was the demographic modeling validated beyond human review?
3. Are there plans to expand this product beyond Seoul or to other markets?
4. What are the key assumptions underlying the footfall and demographic estimates?
5. Has the team considered how to scale beyond a single person working with AI tools?