The Dunleavy Organization

Data · Analytics · Digital Delivery

Cost of Living Reality Check

A public data product for comparing occupational wages and regional price levels across U.S. states — explore market medians without inventing a salary, or evaluate a concrete job offer after cost of living.

Click image to expand
Role
Sole developer
Timeline
2026
Type
Data product
Status
Live

Problem & Outcome

People get job offers and relocation questions without a clear, data-backed view of typical pay in a state and what an offer is worth after regional prices. Early exploration used SQL Server, Power BI, and a WordPress embed — a weak fit for a consumer product.

Outcome: an owned web product at costoflivingrealitycheck.com with two modes — Explore wages (map and state table, no salary required) and Evaluate an offer (COL-adjusted offer vs medians, illustrative take-home). Official BLS and BEA sources with labeled data vintage.

What You Can Do on the Site

Mode Inputs Outputs
Explore wages Occupation; state optional National medians, U.S. map, sortable state table
Evaluate an offer Occupation + offer $ + state COL-adjusted offer, vs state/national median, rough take-home

Map color can show state median wage, COL-adjusted median, or BEA regional price parity (price level). Tax figures are simplified federal + employee FICA for illustration — not tax advice.

Architecture

Desktop analytics pipeline publishes small JSON artifacts. The public host runs an API and static UI — no open database and no Polars on the server CPU.

Hosting & Operations

Live host: same family as this site and SinglePlayerGamers — Ubuntu, Caddy reverse proxy, Cloudflare edge, deploy from Windows.

Runtime: systemd service for the API; production dependencies are API-only (prebuilt publish JSON on the server).

Measurement: GA4 on the COL domain (separate property from this consulting site).

FastAPI MapLibre Polars (desktop) Caddy + Cloudflare

Tech Stack

Chosen for an honest public-stats product under home-lab constraints.

Layer Choice Why
Sources BLS OEWS, BEA RPP, MERIC Official public stats; annual vintage labeled in the product
Transform Polars on the desktop Fast local analytics; not required on the web server
API FastAPI + Pydantic Clear contracts for search, market, compare, offer check
UI HTML, CSS, JS + MapLibre Interactive choropleth without a paid map vendor for MVP
Earlier path SQL Server + Power BI + WordPress Lab / archive — not the live product

Takeaways

Full ownership of a public data product: ingest official sources, model and publish, build an interactive map interface, and run it on the same hosting pattern as other Dunleavy properties — with product judgment (explore without forcing a fake salary) and honest methodology.

Explore the live wage and COL tool?

Brief project notes are enough — I reply by email.