Federal Communications Commission

A New Way to Measure Broadband in America

April 9th, 2010 by Dave Vorhaus - Expert Advisor, Economic Opportunity

One of the central components of the National Broadband Plan is to increase the amount of information consumers have about the performance of their broadband service.   Transparency makes markets more competitive as consumers are armed with the information they need to choose providers and understand the service they receive.  We have already started that process by posting tools on our website that can measure point in time speeds, and using comScore and 477 to show aggregate figures on broadband connections. However, speed measurements from these tools are often affected by many factors, including slow home computers, internet traffic patterns and network congestion.  As the New York Times and others have pointed out, consumers need to know what level of service their ISP is actually delivering, independent of other factors, and how that performance compares to the advertised maximums of a connection.

A few weeks ago we announced the first step in collecting that information for consumers, the solicitation of bids for a third-party to independently test absolute speeds that providers deliver.  We are pleased to announce that the bidding process resulted in the selection of SamKnows Limited to partner with the FCC on broadband testing.  SamKnows brings a wealth of experience in measuring and analyzing broadband speeds, having completed a similar project for Ofcom, the regulator in the United Kingdom, last year. In a couple of weeks, we will be asking for consumers from across the country to voluntarily install hardware in their homes (on an opt-in basis) that is capable of measuring broadband performance.  The measurements will give us results across a broad swath of providers, service tiers and geographic areas.  More details on how to volunteer will follow in the coming weeks. We are tremendously excited about this announcement, the next step in the process of increasing transparency and competition in the broadband market and better informing consumers about their broadband service.

The FCC will also release a Public Notice in the coming days with details on SamKnows’ technical approach and methodology to allow for comment and new ideas. We’d like to encourage everyone to provide us with thoughts and comments about the approach, both here on the blog as well as in direct response to the PN. And, of course, when the website is announced -- sign up to participate in the panel!

23 Responses to “A New Way to Measure Broadband in America”

  1. MKam says:

    The most connected individuals in America have basically 2 choices for broadband. Competition is the only thing that will catch America up with the rest of the world. I can currently get Cable (COX) or a DSL line which is barely broadband, and I live within a apartment complex within city limits. I wonder why my service hasn't been increasing but the price has? $60/month for a 3 Mbps connection is not going to get 'America Connected'

  2. Joe A says:

    MKam, you sound like a COX employee.

    DSL is very competitive and comes in many flavors from many providers. Often CLECs provide the best service, 8-15Mbps for $20-40/mo.

    Just because your ILEC (primary local phone company) doesn't provide great speeds doesn't mean DSL is junk.

  3. Guest says:

    Oh, the irony. The first widespread meaningful testing of connection speeds will probably lead to the first widespread instance of intentional traffic discrimination. This testing is begging to be gamed by the ISPs, who will have every incentive to preference this traffic over their network so as to not look slow.

  4. Alan Weinkrantz says:

    Here's something that can really help change - the wired home network and the new standard that's coming down the pipe. Check out -

  5. Leland Standford says:

    Yes. Sign me up I will be looking for the info. No problem participating in this. BTW: I have always believed the 'information superhighway' should be treated just like physical highways. Were are the privately owned toll roads in this country? Why do individual corporations control our broadband? Why don't local municipalities run wimax networks much like they have a streets and sans dept to run the physical roads? Why aren't the backbones run like our interstate highways? We treated the Internet providers like we did the rail barons and now we pay thier tolls. Phooey.

  6. Guest says:

    I have had broadband with both Cable and DSL, neither of which were capable of their advertised speed. I am currently paying for a 15Mbit connection but only recieve speeds of roughly 1.5 to 2 mbit per second. Both my cable and DSL providers state that it is my fault for some reason. I am excited about hearing the news of the test hardware you will be offering, and will gladly volunteer.

    Hopefully the providers advertised speeds will be more accurate, and constant in the future, Thank you for your effort to understand these issues that have been unresolved for several years.

  7. Toymaster says:

    I live they say on the fringe, yet I live less then five miles from downtown, of my CO so I cannot get the 3mbps or 6mbps only the 1.5mbps which is now barely broadband. This is with AT&T, yet I don't want Comcast as they increase their rates yearly for the same speed. So, how will the FCC contribute to getting America Connected? Two choices is not competition.

  8. danapop says:

    This is a good step. The problem with speed tests is that cable scams them with 'powerboost'-a technology designed to make speedtests look good! Basically, they provide a burst of speed for about 30 seconds then slow down to the crap speeds again-essentially 'gaming' the speed test.

  9. Brett Glass says:

    I'm surprised. A decent off-the-shelf Internet router costs about $50, and one that's specially made to measure a user's broadband performance certainly wouldn't cost less. Multiply that by the required sample size of 10,000 broadband customers and you get... $500K, the size of the SamKnows bid. This leaves nothing for software development, test development, data analysis, interaction with test subjects -- even postage! Something is very odd here. Could someone please explain how a study costing only $500K could possibly meet the bid requirements and produce useful results?

  10. Tyler says:

    Americans would definitely benefit from increased competition among broadband providers as it would lead to cheaper fees for broadband service, an incentive for companies to provide an edge on services and innovation and drive the industry forward. It was only a few years ago that a <a href="">broadband speed test</a> would report back a measly 512kbps on my line, but now I can easily reach speeds varying from 6Mbps to 13Mbps or even 20Mbps depending on where I am, at home or work and that has opened the doors for many interactive and media rich services to be utilized through the broadband connection.

    We should look at the privatization model in the UK where the original sole provider, BT was made to sub-license it's telephone exchanges to smaller firms at commercially viable rates so that a whole host of companies could enter the broadband market as broadband service providers over the existing backbone network.

  11. Stuck on dial-up says:

    In rural areas like were I am there are 2 internet connections 1 is Dial-up and the other is Satellite internet with limits to it and extremely high tier pricing like for example around $80 US for monthly charge for 1.5Mbps and a 17GB bandwidth usage cap. Were i am in California from what i am getting off of forums and rumors that Verizon which is in my area 'Phelan, Ca" does not even have FiOS on the books for this area maybe because it is a rural area, the citys on both side of Phelan, Ca have reports of FiOS eaither in place or being worked on. In Phelan for DSL was available but all the slots for it are all full and there is more houses in range for DSL then DSL slots. The only other choice i can see is EVDO, EVDO can't be your main High-speed internet source because of the 5GB limits. Also after the schools let out a few short minutes after the libary's computer stations are FULL and you have to wait for a computer by making a reservation and you only get 60Min to do what you need but keep in mind of this over 10 computers on 1 business DSL line with things like youtube, Myspace, facebook, flash games from like nickeloden being accessed that like bogs down to Dial-up speeds from 5KB/s to under download speed with being bogged down.

  12. Marathon says:

    For rural areas, the best choice is a quality fixed wireless solution. With today's fixed wireless technology, the customer can expect a consistent 3+ Mbps in each direction. The next generations of radios bumps the speed to at least 10 Mbps in each direction. The key to making this work is a provider that builds a quality network and not one that is under financed. Another critical element is competent personnel who can engineer a quality network and then run it to quality standards. Fixed wireless is superior to satellite. A challenge to building a quality network is adherence to strict engineering standards. Fixed wireless has the lowest cost per square mile of coverage of any technology that delivers quality service with high throughput and low latency.

  13. Hephaestus says:

    The USs ISPs have choosen not to provide customer information to the US government to create a broadband map by address, company, and connection speed. This has cost the US tax payers $350 million dollars and been farmed out to the telcom industry and friends. The resulting data will be biased and show what the telcoms want it to show. This is to maintain their local monopolies. The data will also be years out of date when it is compiled.

    Since most internet users use search on a daily basis. The simple low cost solution to both the broadband mapping and speed mapping is to contact google, yahoo, microsoft, and all other search companies used in the USA.

    For the broadband mapping project the questions to be asked are.
    1) What is your Address including apartment number?
    2) Who is your service provider?
    3) What is your connection speed?
    4) What is your current monthly cost for internet access?
    5) Do you have any other options of internet access?
    6) If yes to the above question, what are the other options for internet access?
    7) Record the persons IP address and do a trace route back to it.
    8) Leave a Cookie for the browser
    9) Pull the modem-router info if possible

    For the speed mapping part of the National Broadband Plan. At the end of the questionare do a speed test on the internet connection. Randomly do a speed test based on the cookie making sure the trace route goes back to the same modem-router and-or area of the ISPs network. Next plot it on google maps for all to see.

  14. Guest says:

    I live in rural eastern Kentucky with no cable, no DSL. I have wireless 6 dn/1 up for just under $100/mo. The catch is I had to install a 140 ft microwave tower for about $6000 to get service, plus about $10-11 per month for electric service to the tower. There is fiber from the nearest tower to the internet through my ISP's NOC. The fiber provider charges my ISP $100/mo for 10 capacity, There is a separate provider for internet access on the far end of the fiber. Performance is near rated except Monday AM, Firday PM and school holidays. You guys have nothing to gripe about until you have to spend $6000 just to get service.

  15. Guest says:

    Ok, the rural dwellers have a problem getting high-speed internet, and I do sympathize with them. But I would like to point out that I live a city, about a mile from my ISP, and today I measured my speed with the FCC Ookla tool to be 221 kilobits per second. I had a nice 1.4 Megabit/Sec DSL connection until the carrier "improved" things - for the ISP they promote. I hope I get a chance to put the monitoring hardware in my house, and that it measures the speed I'm really getting.

  16. Guest says:

    I've known for the last seven years that I''m not getting the speed advertised by our broadband company. They openly admitted this when the exuberant rate hikes began and I noticed the lack of customer support when my service did go down.

    I was told there were now residential and business accounts, the business accounts were given preferential treatment over residential even though we were both paying the same price at that time. Now they offer several packages at varying prices, however, you need a business lisence to obtain the preferences of that package which continues to supercedes residential.

    You are able to purchase faster speeds (less hopps around the globe to get to a local addy, ie: your broadband server; I do know how to trcroute) which I don't understand given this is suppose to be on a fiber optic line somewhere underground. Its all a big gist as far as I'm concerned. They still use your cable line which is based on the underground phone lines that may exist in your area. So, just how fast is fast here(?): after taking the test it wasn't much of a surprise, after all, I do remember dial-up speeds; we're at appx. 1111kps dl, ul 245, and 242 ms. Of course, our router could be slowing things down a tad, and the fact that we have two pc's connected to it. But this is not even close to what is advertised, and/or quoted when I call for trouble-shoot and to report outtages that take place at least two times a day.

  17. Guest says:

    Same poster as above: So how do I sign up for this FCC experiment?

  18. Brian says:

    Dave, there is a SUBSTANTIAL problem with the company you selected to perform the broadband speed tests for the FCC. I remember reading about the initiative in the news and signing up to assist in the FCC trial.

    I've been getting these emails from "SamKnows" telling me to "click here" to test my ISP speed. The problem is that not only does the email look like spam, and emplores users to "click here" for a common spammer tactic, but there is extremely limited mention of the FCC or the fact that this is something I signed up for. Yes, the email "SamKnows" is sending out states "We’re pleased to say that the details you originally supplied match the type of broadband connection we’re looking to measure". But, there is ZERO credibility implied in this statement that I actually signed up for this.

    If you don't fix the email message being sent out by SamKnows, and fast, you will get considerably fewer people engaging in your test than you should be getting. The taxpayers are paying for this study and we deserve to have it orchestrated with competence.

  19. briand805 says:

    signed up for this months ago.. was told i would be informed in a few weeks about receiving the equipment and what not a couple months ago.. that was the last i have heard from them. great job! I am sure our tax paying dollars are hard at work here.

  20. Spinflow says:

    My household applied to be a part of the program, and from application to acceptance…to receipt of the new, Wireless-N router…to confirmation that the program IS receiving data from our unit…to the successful ability for me to login and view the data my unit specifically is collecting…EVERYTHING has been smooth as BUTTER. i HAD ZERO issues setting up…the router works without error…endlessly…still to date without needing but ONE reboot (Because of a power outage and a reboot on my cable modem at that)…

    No issues. Super High quality router at no cost…and absolutely no issues whatsoever with my internet service or them being in-between.

    I had ONE question for support concerning my login information to their "dashboard" site where users can view their own statistics daily…and they responded and had my question answered within a couple hours.

    So I'm extremely happy, and all for about ten total minutes of my time invested in ALL steps…and now with a great router to use for free.

    No complaints…thanks taxpayers! Much appreciated! I hope the data turns out to benefit us all.

    Charlotte, NC

  21. Brian says:

    Dave, you guys had better FIX the issues with the email being sent out by SamKnows if you expect to get a reasonable response to the study. Please see my comments above.

    Frankly, the only way I even knew that the message wasn't a spammer trying to get something of mine was by doing a Google search for "SamKnows" and "FCC". How many people do you think are going to go to that trouble? Seriously!

    Fix this! Have the email message to user's originate from the FCC.GOV mail systems. Or provide a link in the email that clearly and unambiguously (i.e. text email, not HTML) points to an FCC.GOV URL for more information.

  22. Guest says:

    Hi to the posters above I am a volunteer for samknows internet testing in the UK I don't know the technical details of how it works but they send you a router which you plug one lead from your router connected to the internet into its internet port. This leaves you with four ethernet connections in the samkows router to use for your network the sam knows router is passive and when it detects you are not using the internet sends and receives packets of data to test the connection available at random times throughout the day. It then sends this information about the quality of the internet connection to Samknows servers to give a comprehensive set of readings about quality and speed of this particular line and isp. The isp has no idea when the readings are taken or by whom. So he can't '"game'" the results in his favour as someone suggested. We are annonymous customers spying on the isp's ability to deliver advertised services. See how they like the tables turned lol.

  23. Paul in Atlanta says:

    I signed up for this, and am I ever sorry that I did! The router they sent did not work, there was no tech support evident, and I can't even get them to tell me how to send the hunk of junk back to them. I have been assured, essentially, that I hooked it up wrong and that it's all my fault, but the problem there is that I happen to be a well degreed electronic engineer. Hopefully our government isn't sending the contractor a pile of money for their assistance, as it is definitely not being provided.

Leave a Reply

Capture The Phone Numbers Using Your Camera Phone

If you have a camera and a 2D matrix code reader on your mobile phone, you can capture the FCC Phone numbers right to your phone by following these three easy steps:
Step 1: Take a photograph of one of the codes below using the camera on your mobile phone.
Step 2: Use your phone's Datamatrix or QR Code reader to decode the information on the photograph. Please note, these code readers are device specific and are available to download on the internet.
Step 3: Store the decoded address information to your phone's address book and use it with your Maps or GPS application.

Datamatrix and QR FCC Phones