How to Read Broadband Data

The broadband data on PlainBroadband comes from the FCC Form 477 dataset, the primary federal source for broadband deployment information in the United States. Understanding what this data represents, and what it doesn't, is essential for making accurate comparisons between providers, states, and technologies.

What Is FCC Form 477?

Form 477 is a data collection program run by the Federal Communications Commission. All facilities-based broadband providers in the U.S. are required to file reports twice per year (as of June 30 and December 31) detailing where they offer broadband service. Providers report at the census block level, meaning they indicate which specific geographic areas they can serve and at what speeds. This creates a granular map of broadband availability across the country.

What Records Mean

On PlainBroadband, the "records" count represents the number of individual deployment filings a provider has made with the FCC. One record typically corresponds to one census block served by one technology type. A provider serving 1,000 census blocks with cable internet would have 1,000 records for that technology. If the same provider also offers DSL in 500 of those blocks, that adds another 500 records.

This means record counts are not subscriber counts. They measure the breadth of a provider's coverage footprint. A provider with 10 million records is reporting broadband availability in a very large number of census blocks. Providers offering multiple technologies will have more records because each technology in each block is counted separately.

Understanding State-Level Data

Each state page on PlainBroadband shows several key metrics. The provider count indicates how many distinct broadband companies have filed deployment records in that state. Average speed is the mean of maximum advertised download speeds reported by providers. Technology percentages (fiber, cable, DSL) show the share of records for each technology type within the state. These percentages do not represent the share of population with access to each technology. They reflect the share of deployment filings, which is a measure of how much infrastructure investment each technology has received.

Limitations of the Data

FCC Form 477 data has well-known limitations. Until the transition to the Broadband Data Collection (BDC) system, providers could claim coverage of an entire census block even if they served only one location within it. This led to overstatements of coverage, particularly in rural areas with large census blocks. The speeds reported are maximum advertised speeds, not typical or guaranteed speeds. Actual user experience depends on network congestion, home equipment, and distance from infrastructure.

The data is a snapshot as of the filing date and may not reflect recent infrastructure changes. New deployments, discontinuations, and service modifications that occurred after the filing deadline will not appear until the next reporting period. Despite these limitations, Form 477 remains the most comprehensive source for understanding the broadband landscape at a national and state level.

How PlainBroadband Processes the Data

We download the raw Form 477 datasets from the FCC, aggregate them by provider, state, and technology, and organize the results into searchable pages. No data is modified, interpolated, or editorially adjusted. All numbers on PlainBroadband trace directly back to the FCC filings. When we compute derived metrics like averages or percentages, the calculation is described on the relevant page.

Tips for Using the Data

When comparing providers, look at both record count and state count. A provider with many records in a few states has deep regional coverage. A provider with fewer records but many states has a wide but potentially thinner footprint. When comparing states, consider that larger states naturally have more records because they contain more census blocks. Use the technology percentages and average speed metrics for meaningful state-to-state comparisons.

For more on our data pipeline and methodology, see the methodology page.

Frequently asked questions

Where does this data come from?

All figures on this page derive from official federal data — primarily the U.S. Bureau of Labor Statistics, U.S. Census Bureau, U.S. Department of Health and Human Services, and U.S. Department of Labor. We cite the underlying agency and series in the methodology section. No proprietary aggregators are used.

How often are figures updated?

Each series follows its own publication cadence. We refresh our database within 30 days of each upstream release. Specific update timestamps appear in the page footer where available; the methodology page documents the cadence per data series.

Can I use this data for my own analysis?

Yes. The underlying federal data is public domain. Our presentation, calculations, and editorial commentary are licensed for individual reference. For commercial republication or large-scale data extraction, contact us at the email listed on the contact page.

What if the figures here disagree with another source?

Different sources use different methodologies, definitions, geographic boundaries, and reference periods — disagreement is normal and informative. Our methodology page documents exactly which series and reference period we use for each metric, so you can reproduce or audit the figures against the upstream agency directly.

Calculator Plug your numbers in Methodology How we computed this Research Original analysis By state Drill into your jurisdiction

Worked example: reconciling FCC and Ookla data for a single address

A consumer evaluating an address might pull three data sources. The FCC National Broadband Map reports that the address has Provider A offering 1000/50 Mbps fiber and Provider B offering 300/10 Mbps cable, both available. Ookla's crowdsourced Speedtest data for the surrounding ZIP code shows median residential download speeds of 195 Mbps and median upload of 22 Mbps — well below the maximum offered speeds. The FCC's Measuring Broadband America panel reports Provider A delivers 109% of advertised on average and Provider B delivers 98%. Synthesizing the three sources: the address is well-covered, both providers have capacity, but the Ookla median suggests most subscribers in the ZIP are not buying the top tier. The FCC's National Broadband Map (broadbandmap.fcc.gov) is the official source for coverage; Ookla and other crowdsourced datasets show actual realized speeds across households actually subscribed.

Broadband data source reliability matrix

Data sourceDiagnostic weightWhat it measures
FCC National Broadband Map (BDC)25%Provider-reported coverage by location
FCC Measuring Broadband America (MBA)25%Actual delivered speeds on FCC routers
Ookla Speedtest Open Data15%Crowdsourced consumer-run tests
M-Lab NDT (federal partnership data)15%Crowdsourced via Google search box
State broadband office maps10%State-level coverage validation
Provider's own coverage check5%Marketing-led address lookup
BroadbandNow / similar aggregators5%Affiliate-driven plan comparisons

Challenging a National Broadband Map entry

One of the most underused features of the FCC's National Broadband Map is the public challenge process. If a household pulls up its address and the map reports a provider offering service that the household cannot actually receive (a common situation with claimed-but-unbuilt cable franchises and overstated fixed-wireless coverage), the household can file a Location Challenge directly through the map interface. The FCC accepts challenges from individuals, ISPs, state broadband offices, and local governments. Successful challenges remove the inaccurate claim from the map and reshape future BEAD funding eligibility for that area. Filing a challenge takes about 10 minutes and requires uploading basic evidence (a screenshot of the provider's "service not available at this address" message is typically sufficient). The process matters because BEAD funding flows to areas designated unserved or underserved on the map; an inflated provider claim can disqualify an area that genuinely needs subsidy. Challenges have already removed millions of inaccurate coverage claims since the map relaunched in 2022.