No black box. Eight datasets, one weighted score, refreshed nightly. Here is exactly what goes into it, and which parts are ours.
Every regulated utility reports how long and how often its customers lost power. That gives us outage minutes and interruption frequency at the utility level, which we attach to each business by service territory.
Historical severe weather events with dates, plus live watch and warning polygons. This is what lets a storm-response list appear while the warning is still active.
County-level percentile risk across eighteen hazard types, along with expected annual loss. This is the underwriting-grade context behind the insurance and compliance angles.
Business counts, industry classification, and employment size bands. This tells us what a business actually does, which determines what an outage costs it.
Which providers and technologies claim to serve a location, and how much real competition exists. Thin provider coverage is the core signal for connectivity and failover selling.
Millions of real speed tests, aggregated by map tile. Where measured speeds fall short of what the FCC filings claim, that gap is a selling signal on its own.
Georgia Tech's live internet outage detection, watching routing and traffic signals around the clock. It catches connectivity failures that never make a utility filing.
Solar resource by location and utility rate data. This is what makes the solar and storage ranking honest: the same roof pays back very differently in Arizona and in Ohio.
Each business receives a 0 to 100 score. The score is not a measure of how good a lead is in the abstract. It measures how much this specific business stands to lose when the infrastructure around it fails, weighted for what you sell.
Four inputs carry most of the weight:
A generator dealer and a connectivity seller are looking at the same city and should not get the same ranking. When you choose a vertical, the weights shift toward the signal that drives your product: grid reliability for backup power, provider thinness and the claimed-vs-measured broadband gap for connectivity, electricity rates and solar resource for solar, live weather for storm response, hazard percentile for insurance.
Weather alerts and internet outage detection update continuously. Outage and event data refreshes nightly. The federal datasets themselves publish on their own schedules, typically annually, and we pick up new releases as they land.
We say plainly elsewhere on this site that our raw inputs are public. So it's fair to ask what you'd be paying for. Four things, none of which exist in any public dataset:
The eight datasets don't share a common key. Utility territories, weather polygons, county hazard indexes, broadband blocks and speed-test tiles all describe geography differently. We built the crosswalk that resolves every one of them down to an individual business address. The datasets are public; the join is ours, and it's the part that takes teams months when they try to rebuild it.
A generator dealer and a failover ISP looking at the same city need different rankings. Our weighting profiles per vertical — how much grid reliability, provider thinness, hazard percentile, electricity rates and live weather each count — were tuned against what actually drives buying in each trade. The 0–100 score you see is the output of that tuning, not a generic risk average.
For every business we keep a dated history of the outages, storms and connectivity failures at or near that address. That timeline is what turns a score into an opening line a rep can say out loud: a real event, with a date, at that location. No public dataset ships this; it only exists once everything is joined and kept current.
Live weather alerts stream in continuously, outage and connectivity signals refresh nightly, and new federal releases are picked up as they land. A score built on last year's snapshot decays fast. Keeping 4M+ data points current, automatically, is infrastructure we run so your reps don't have to think about it.
Put differently: the government publishes the ingredients. The recipe, the kitchen, and the fact that dinner is on the table every morning are what Main Street Vitals sells.
Main Street Vitals is not a contact database, and it does not claim to be a bigger list than the incumbents. Its output is ordering and timing: which of the businesses in your territory to call, and why now. Contact information is included so your reps can act on that ordering, but the ranking is the product.
Questions about a specific state or vertical? Ask us directly — if it's a bad fit we'll say so.