This past November, the FCC released new broadband maps that provide a snapshot of the state of broadband deployment in the United States. This is part of our evolving effort under the Broadband DATA Act to give us an accurate picture of where broadband is and is not across the country. Back in November, I noted that these maps represented more of a beginning than an end for the new era of broadband mapping. In that spirit, it is important to note that we’ve come a long way in the short time since that November release. Here are some key developments and things we’ve learned over the past four months.
Our new systems are up and running and working—especially the challenge process.
One of the big innovations of our new maps, compared to the old maps, is that they were designed to be constantly and consistently refined through an iterative process. This required us to build systems that allow for the steady collection—and integration—of new data. The centerpiece of this effort was setting up a robust challenge process where consumers and key stakeholders like state, local and Tribal governments could review our maps and let us know if our data matched their reality on the ground. Over the past four months, consumer and stakeholder engagement has been growing, and this feedback has been integrated into our datasets. For evidence of how this working, just look at the numbers.
Our maps are built on top of two datasets: a locations dataset, what we call the Fabric, that tells us where fixed broadband could be installed, and an availability dataset that shows what broadband services, if any, are actually available at the Fabric locations.
On the availability side, the FCC’s mapping team has a system in place to collect these challenges and report them to the relevant internet providers who supplied the availability data. Providers then have 60 days to concede or dispute the challenge. To date, stakeholders—primarily state governments—have stepped up to provide more than 600 bulk challenges covering provider reported availability at several million locations.
In the past four months, our mapping team has processed challenges to availability data for over 4 million locations. In other words, on average, we are addressing availability challenges to tens of thousands of locations every single day. Every two weeks, our public map is updated to reflect all availability challenges that have been resolved. In other words, the system is working.
Our next maps will be noticeably better—thanks in large part to even better location data.
The fact that we now have a working challenge process to hold internet providers accountable for their availability data is a sea-change improvement from the FCC’s old broadband maps. Remember, the FCC’s old maps had zero systems for public feedback. The agency simply reported what the carriers said on a census block basis. But an even bigger improvement is that our maps now reflect broadband availability at the physical location level. You can’t offer location-level data without an accurate accounting of locations where fixed broadband could even be installed. That brings me to the evolution of our locations dataset—the aforementioned Fabric. This Fabric was developed based on more than 200 public and commercial data sources ranging from public records to tax assessments to satellite imaging.
Our first version of the Fabric identified over 113 million locations where fixed broadband could be installed. For context on how much more granular this data is than our earlier maps, there are 8.1 million census blocks. We have largely completed work on the second version of the Fabric, and I can definitively say that we have taken a big step forward.
The topline from the new Fabric is that it now reflects over 114 million broadband-serviceable locations, a net increase of 1.04 million. That tells us a few things. First, a net adjustment of less than 1 percent to the number of serviceable locations says that, on balance, the November pre-production draft of the National Broadband Map painted a helpful picture of where high-speed Internet service could be available. But it also demonstrates that we’re not resting on our laurels and, with the help of Fabric challengers and our own assessments, we’ve made important corrections and additions to the data. Let’s dig into some of notable details.
For starters, we’ve added 2.96 million new broadband-serviceable locations. Percentage-wise, we saw some of the most significant increases in Alaska, U.S. territories, and Tribal lands.
At the same time, we have removed 1.92 million locations from the first version. The reason for these reductions ranged from data refreshes to more sophisticated tools to help remove structures like garages and sheds from the total count.
The challenge process for our locations database contributed significantly to these refinements. For example, location challenges from states led to nearly 122,000 of the new location additions. But the majority of location additions and other adjustments were a result of the ongoing efforts of CostQuest, the contractor developing the Fabric. At our direction, CostQuest is continually updating and improving the dataset by refining their models and processes and using better input data sources such as new and more granular parcel data.
Providers used this improved location data for their second required filing of availability data, which concluded on March 1, and it will be displayed on the map later this Spring. While over time we expect future versions of the Fabric to require fewer refinements, these ongoing efforts to improve the Fabric outside of the challenge process will continue and will remain an important tool for the improvement of the National Broadband Map.
We remain on track to release new and improved maps later this Spring.
The amount of activity over the past four months to improve the National Broadband Map is certainly notable. What’s also notable is that this work is happening in a timely manner as required by the Broadband DATA Act and is consistent with our obligation to do a major refresh of our maps every six months and update the map a few times a month when we resolve availability challenges.
This progress would not have been possible without the extensive engagement of state and local leaders as well as the collaboration of NTIA and other federal partners in this effort. Thanks to their partnership, we will release a National Broadband Map this Spring that more closely reflects the state of connectivity on the ground. And we’ll do it again and again and again every six months, constantly strengthening this foundation for smarter broadband policymaking for years to come.