Back in 2011 the FCC began collecting real-world user broadband data from customized routers, then issuing reports on which ISPs were failing to deliver advertised speeds. It's one of the few FCC policies in recent years that has truly paid dividends for consumers. The first report "named and shamed" several larger ISPs like Cablevision for offering horrible peak performance, offering less than 50% of advertised sustained speeds at peak hours. The tactic did wonders -- by the second report Cablevision had boosted that performance to 128%.

Today the FCC released their third annual report naming and shaming ISPs whose peak bandwidth performance isn't up to snuff. According to the latest FCC data, just four ISPs tracked by the agency managed to deliver faster than advertised speeds during peak hours: ViaSat (137%) Verizon FiOS (118%), Cablevision (115%), and Comcast (103%).

As three of these companies attest, the FCC's practice of naming and shaming works -- if there's regional competition. Verizon FiOS and Cablevision in particular have spent the last few years trying to one up one another in terms of offering speeds faster than what is advertised on promotional materials.

The FCC says they're now including satellite broadband in their study for the first time; ViaSat only cooperating after they launched their faster Exede service. HughesNet's Gen4 service is not included in the rankings -- probably a good thing for HughesNet considering the massive complaints we've seen about under-performing connections from the company's newest Gen4 offerings. Satellite's advertised speed rankings of course don't take into account the often draconian daily usage limits these users face, or some of the compression used by these operators.

As for the rest of the ISPs in the FCC's study, most are at least getting close to what they advertise in terms of speed delivery. Cable offered 99% of advertised speeds at peak times to 85% for DSL. As the chart above illustrates (click to enlarge) there's several notable under-performers, including AT&T, Windstream, CenturyLink, and other telcos with little to no competitive incentive to upgrade their networks. The lone exception appears to be Frontier who, after getting slammed in the first two reports has made significant performance gains -- though still only offering an average of 87% of advertised speeds at peak times.

According to the FCC, the agency plans to expand the tests to include additional metrics and other technologies like wireless. The agency historically has avoided naming and shaming carriers on price, as the industry violently fights the FCC on collecting and publicizing that data for obvious reasons. As the ViaSat Exede example obviously highlights, collecting data on which ISPs impose usage caps would also be invaluable.