Guest commentary: what went wrong with muni Wi-Fi, what cities can do now

It is too easy in light of the recent news that MetroFi is selling their networks to begin to say that the municipal WiFi (note that I said Wi-Fi and not wireless) downturn is all  because of EarthLink and MetroFi. Easy to pick on the companies that have come out and said, “This isn’t working and we are closing shop”.

I think that in reality there is an ecosystem of pundits, vendors, consultants and others that also bear some responsibility for where things stand today. What I would like to do in this commentary is lay out a brief summary of the  events that I believe have led us to where we are in the market today.

What went wrong

In 2005 the City of Philadelphia chose EarthLink to build their network. Well known story, no need to comment further on this decision. What happened next though was more interesting. Here was now an opportunity to take what was a proposal from a company that, though they were very experienced in providing DSL service, had virtually no experience in wireless networking, and turn it into a repeatable business opportunity. In fact, it became a business model that was touted at conferences and seminars throughout the United States and the world.

Thus began the  “momentum buying activity” that was characteristic of the tulip craze, Internet stock market craze and similar episodes throughout history.  Many cities wanted to do what Philly was doing, and rightfully so. After all, cities compete with each other in the same way that companies do and what city wants to be seen as falling behind with respect to technology.  And if all you need to contribute is city assets for a wireless network, why wouldn’t you? We also soon started to hear about how other cities were leveraging their wireless networks to attain real cost savings: eliminating T1s, getting rid of their mobile data plans, seeing increased productivity gains from their employees and so on. A new era of enhanced municipal services via wireless broadband had dawned.

About this time, summer 2006,  is when one of my clients, Grand Rapids, Michigan, was presented with a dilemma: do we follow the path that other cities had taken and choose EarthLink or do we go with another company that was in the words of one colleague of mine, “a horse of a different color” – Clearwire. This was a tremendous decision for us, so we decided to do what we thought was appropriate in this situation: go visit some networks, check references and make sure that these companies were able to do what they said that they could do, delivering the applications that were important to my client.

We became convinced,  based on the results of our due diligence,  that Clearwire had demonstrable experience in meeting the mobile public safety needs of my client. However, we also realized that the cities that had been touted as offering municipal services via wireless broadband were still far from undertaking that step.  Contrary to what we had seen in the press releases and heard at the conferences.

I also looked at the  financial model, specifically the view that reflected the service provider’s business case, playing with the critical metrics: node density, penetration rates and time to reach the inflection point in penetration rate. EarthLink had presented to us a business case that assumed 22% penetration by End of Year 4, a figure that they assumed to be conservative.

Over the fall and in early 2007 I played with the model as new figures began to emerge about node density (40 is the new 30) as well as anecdotal numbers regarding initial penetration rates in Wi-Fi markets. Pretty quickly it became apparent that the model was broken. Not just for EarthLink, but for any service provider that was constrained by the following conditions:

1.    Minimal financial commitment from the City or County
2.    Limited or no anchor tenancy
3.    Limited due diligence performed by the City or County in terms of municipal applications and cost savings that may be attained by using the wireless network.
4.    Expectation that the network would cover 90-95% of the City with wireless coverage as opposed to just in the areas where there was a solid business case.
5.    Negotiations that can probably be best described as “let’s squeeze the service provider dry”.

As I have discovered there were a few other industry colleagues that were coming to that conclusion at the same time, if not earlier.  In fact, some of these are the same colleagues that have always been arguing that the Philly model is broken. Not making some “knee-jerk reaction” as the writer in this blog would argue.

So, you may say, big deal. You figured out earlier than some of us that, as they say in the South, the Philly model is a “dog that don’t hunt”.

What surprised me is to find out how little due diligence had been done by other cities in regards to their choice of a wireless ISP. It appeared that more often than not cities were making a choice based on the following logic: “Well if they are good enough for _______ (fill in the name of any large city that chose the same vendor as you did), then they must be good enough for our city”. Neither they nor their consultant appeared to have visited reference networks nor done the financial analysis from the municipal and service provider perspective to make sure that it was a Win/Win situation based on conservative assumptions. Because as we have seen, and as my colleague Craig Settles loves to say, “If you don’t take care of your service provider, they will not be around to take care of you.”

My other concern is that there is still a line of reasoning out there that cities can somehow get something for nothing. For example, Muniwireless points to Cablevision’s Wi-Fi offering as an example of free. Unfortunately, I think someone forgot to read the fine print. It is free for those who are already paying an average of  $121 per month. In the words of Mr. Gump, “ I may not be a smart man, but I know what free is.” This is definitely not free.

What cities need to do

Where I think we need to be focused, rather than continuing to look for ways of how we can entice cities with the promise of a telecommunications prince that will come and sweep them off their feet and make all their free Wi-Fi dreams come true, is in the following areas:

1.    Tell the truth: If cities and counties want to use broadband to provide enhanced services, it will require budgetary dollars in the short, medium and long term.  There will  be exceptions, but this will be the rule of thumb in the future. Not what cities necessarily want to hear, but my philosophy is that as a consultant I am not paid to tell cities what they want to hear, but what they need to hear. They also need to hear about real examples of where these investments are paying off in terms of municipal employees using networks to provide services. There were very few in 2006, but there are a lot more now.

A good example of one that is built on the right model is Gainesville Regional Utility, which plans to use the network for traffic synchronization, mobile public safety and other applications. The project has been capitalized properly using municipal funds($3-$4 million) and the fact that 22 service providers/integrators responded to the RFP is a sign that the business model works.

2.    Be honest with challenges: Was very refreshing to be on a conference call with several municipalities last week and hear them speak openly about deployment challenges as well as how these challenges are being overcome. This industry has been plagued by too many glowing reports of networks that just don’t measure up to their PR.

3.    Stop the bluffing:  Too often municipal negotiations are like high stakes poker games. I would encourage a more open dialogue, because if one party comes out feeling like they got the short end of the stick the project is already off on the wrong foot. In addition, if the City doesn’t have the money, say so and do not waste  the valuable time and money of the service provider that is responding to the RFP.

4.    Do the financial due diligence: Run the financial models from the perspective of the service provider and the municipality  to make sure that the business case works for both parties.

5.    Focus on operating expenses: I would assume  based on a figure of 20-30 percent of CAPEX that the operational expenditures were going to cost $2-$4 million per year to operate the network. Not something that Philly bargained for when they chose EarthLink. Other cities are facing the dilemma of $10K per square mile per year just for pole attachment and electricity usage. More thought needs to be given to  how to mount more radios on non-utility assets to eliminate these fees.

6.    Focus on capital expenditures: How can we get the node density and corresponding CAPEX per square mile to move downward, not upward? I think there are some promising developments in terms of smart radio technologies and the introduction of the 802.11n standard in client access radios.

7.    Focus on hybrid licensed/unlicensed networks: Licensed spectrum for mobile public safety and backhaul, unlicensed spectrum for public access.

8.    Consider partnerships with WiMAX service providers: I would suggest that cities explore this option now that WiMAX technology has moved from testing to the market launch phase. The capital expenditures per square mile lend themselves to a variety of network ownership models that are economically feasible.

– – – – – – – – –

Karl Edwards is the Cofounder and COO of Excelsio Communications, a municipal wireless broadband consulting group. Excelsio brings an informed approach to the wireless consulting space that is based on actual wireless network deployment experience. Karl has overseen the design of multiple wireless area networks to date as well as consulted with municipalities and counties such as Grand Rapids, MI; Decatur, GA; Alpharetta, GA; Marietta, GA; Rome, GA; Statesboro, GA; Racine County, WI, and Gainesville Regional Utility which are in various stages of their wireless broadband initiatives. Karl earned his B.S. in Electrical Engineering from the University of Utah and an international MBA from the University of South Carolina. He is also a Proxim Certified Wireless Broadband Engineer.


  1. Very good summary – we are in exactly the same space here in UK – lots of debate and some good success stories and a lot of ???? what shall we do to make it work, oh its too risky!!

  2. One of the areas that fascinated me from the beginning was the fact that municipal broadband assumed only mesh radio equipment. The other was the assumption that even at 40 AP’s per square mile, many people still needed indoor equipment. There were many other false assumptions made technically and financially that also contributed to the demise. When these silly business plans came to market, we stepped back as well as many other companies with designs that were much more solid from technical/financial position. For example, we have a design that is deployable at less than $5K per square mile and even worse case (hills, trees, tall brick houses), less than $10K per square mile. Yearly maintenance and electricity will be less than $50 per unit per year although that varies for each area assuming no pole rental. We also aren’t installing 40 units per square mile which was another major mistake.

    Now that the two major failures are out of the way, we are deploying our systems again with the intent on making a profit. Here is what’s even more funny, my little 6 man company made more profit last year in WiFi services than MetroFi and Earthlink combined. Amazing what you can do if you don’t have to give everything away.

  3. What a great commentary, Karl. The industry needs more thoughtful analysis like yours. Keep up the great work!

  4. Really good summary Karl… as you have always been a tremendous asset to this expanding marketplace and for those exploring large scale broadband wireless initiatives.

  5. For those of us that have been in the wireless business for a long time, I think most of us recognized this as the re-incarnated Hotspot model, just on a grander scale. It was no surprise to the seasoned wireless business person that this would, in time, fail also. On a number of levels, starting with the technical level….

    What I can’t get over personally is why the municipalities thought this would be a long lasting and fruitful relationship. Are they that gullible? And who are these guys that are selling this snake-oil to the Muni’s? Do they have a lick of business sense?