Why I built The Data Bulletin
I kept having the same small argument. Someone on TV says the economy is the best it has ever been, or the worst. Someone at the table repeats it. And I would think: the real number exists. It is public. It is free. It was probably released last Tuesday at 8:30 a.m. Eastern. Almost nobody at the table has seen it.
That gap bothered me more than it probably should have. The government publishes nearly everything: unemployment and prices from the Bureau of Labor Statistics, growth from the BEA, energy costs from the EIA, crime from the FBI. All of it sits in official databases with free APIs. Meanwhile the places people get their numbers are headlines and clips, which mostly carry opinions about the numbers.
So I built the thing I wanted to exist. The Data Bulletin shows the latest official U.S. figures the moment each agency releases them. Every chart names its exact series and links back to the source, so you can check any number in one click. Everything carries a "last updated" stamp, because official data comes out on schedules, monthly, quarterly, weekly, and agencies revise past figures. Pretending otherwise is how a site loses trust.
The part I care about most
When a politician or pundit quotes a statistic, Data Check puts the exact quote next to the official numbers. Each check gets one of exactly three findings:
The word "lie" never appears on the site. I can't know what someone intended. I can only show what the data says and let you compare for yourself. As far as I'm concerned, that restraint is the product.
There are also things the site refuses to do. It never ranks states against each other on crime, because the FBI itself warns that reporting practices differ too much for those comparisons to be fair. Trends and national context, yes. Rankings, no. Nothing publishes without a human reading it first. And there is no scraping: if a source has no official API, the section waits until I find one that does.
The other reason
I wanted to learn to build things properly. I bought the domain in July 2026 and made myself a rule: code the site for real instead of reaching for a website builder. It was almost called TheDatanomics, because the name I actually wanted was taken. The better name shook loose eventually.
Since then the project has forced me to learn serverless functions, API caching, keeping keys out of the browser, and what happens when a federal CSV arrives with Windows line endings at midnight. A course would have been faster. It would not have taught me as much.
Today the site runs 26 charts from nine official sources: FRED, the FBI Crime Data Explorer, the EIA, NOAA's storm archive, NASA's wildfire satellites, the WHO, the World Bank, and the OECD. The pipeline refreshes itself as each source publishes. My job is to keep it honest.
If you spot an error, tell me, and the correction will be visible rather than silent. That is the whole idea.