CostSheet
Methodology

How we build our estimates

CostSheet estimates are built from real, verified public data sources — not national averages, not guesses, and not AI-invented statistics. Here's exactly how each report is put together.

1. Government wage data

Every hourly labor rate comes directly from the U.S. Bureau of Labor Statistics' Occupational Employment and Wage Statistics (OEWS) survey — the same dataset economists and researchers use. We pull the actual mean, median, and percentile hourly wages for the specific occupation tied to each service, in the specific metro area for each city we cover. When metro-level data isn't available for a narrow occupation, we note that rather than substituting a guess.

2. Local contractor data

Contractor counts and average ratings come from the Google Places API, reflecting real, currently-operating businesses in each city — not a directory of unverified listings.

3. AI-assisted writing, data-grounded

Our articles are drafted with AI assistance, but every factual claim about cost, wages, or contractor availability must trace back to the verified data above. Our system prompts explicitly instruct against inventing statistics, and a separate quality-review pass checks every draft for ungrounded claims before publishing. Content that fails this check is held back rather than published.

4. Ongoing accuracy

Wage and contractor data is refreshed periodically rather than written once and left static. Each report notes when its underlying data was last verified.

What we don't do