Stacey pinged me and pointed to this report about the state of Internet Access Services (pdf) issued by the Federal Communications Commission. It has a lot of fascinating data, but the problem is that the report has data that is 14 months old, which makes it pretty much worthless.
Sure, government officials and politicians may use the data, but who really cares if there were 71 million fixed broadband connections at the end of June, 2009? Likewise, the fact that nearly 41 million folks used cable broadband that long ago is pretty much worthless.
I’m not sure why the FCC even bothers to collect this data if it can’t release it sooner. It’s nice to have the data, but it’s just not useful so late in the game. The only place where 14-month old stats make any sense is in a baseball fantasy league. Yup, those stats made me pick Boston Red Sox pitcher Josh Beckett for my 2010 fantasy team. Fat lot of good that did.
If FCC wants to build a case for better broadband, I personally feel that they should try to collect data faster and deliver it faster than any research firm. Otherwise, they should outsource to commercial research companies that can do the job faster. Why not use that information? Unless I’m missing a big point, it seems there isn’t that much of a variation: The FCC says at the end of the second quarter of 2009 there were 71 million broadband subscribers. Research from Leichtman Research Group, which doesn’t include satellite, Internet connections puts the number at around 70 million.
PS: In case you want to know, at the end of June 2010, there were about 73.5 million broadband subscribers, according to the Leichtman Research Group.
Related GigaOM Pro Content (sub req’d): Who Will Profit From Broadband Innovation?
14 months is an improvement, the Wireless Competition Report used data that was 18 months old; but it doesn’t stop there, the FCC tends to omit facts that don’t support its point of view. The Wireless Competition report omitted the $20 Billion spent on spectrum auctions from its tally of investments in mobile broadband networks. Duh.
Om, you are expecting too much from a federal agency.
We’re glad you noticed our most recent Internet Access Service report, and we wanted to provide a bit of background on timing.
We’re working hard to get our Internet Access Service reports out on a more timely basis, and believe we’re on track to do that—the next report, covering data as of December 31, 2009, should be out soon, after a much shorter cycle than the last one. But it’s worth noting what goes in to these reports. Perhaps the biggest issue is one of scale: this report isn’t based on a sampling of data from around the country or an estimate of what we think is happening, like most private-sector reports typically. Rather, it’s built on a census of all broadband providers. In a typical cycle, that means we have to receive (and process) more than 8,000 submissions.
Such a collection gives us a solid foundation for building policy on real data. But it does take time.
Most providers need time to get their data in shape for filing. So the data for December 2009 wasn’t due to us until March of this year. Following that, we work with filers to scrub the data and make it as accurate as possible. We must contact providers individually to square anomalies and prod procrastinators. And when we, or providers, find mistakes in submissions, providers need to re-file their data. We’re still working with a small number of providers on the December 2009 data.
It’s also worth noting that our new requirement for data submissions on the census-tract level – which gives us a more granular view of broadband subscriptions than we’ve ever had before – has taken providers a while to adjust to.
The good news is that earlier this week was the deadline for providers to submit the June 2010 data; we’re well on the way to getting it all and hope to lock that data set soon. Unfortunately, we haven’t heard from all providers yet, but we’re working on that
Steve Rosenberg, Chief Data Officer, Wireline Competition Bureau, FCC
8,000 ISPs doesn’t sound like a “duopoly.” Don’t say that!