Sun 27 Apr 2008
I had the opportunity to attend the inaugural meeting of the iProvo Review Committee last Tuesday and walked away with more information than I knew what to do with. The Committee was kind enough to distrubute current figures for iProvo subscribers, information on municipal networks, and, best of all, a copy of the consultant's reports from CCG Consulting and Franklin Court Partners. There was a lot of information to soak in and it took me about three 2 hours of reading to slog through it all. After digesting where the network is and where it is going, I can come to only one conclusion: iProvo is the most valuable thing the city has ever done for its residents with a potential to save city departments and residents millions of dollars each year it operates.
Since the strategic summit last December, iProvo has netted a gain of just 180 new accounts over the October 2007 numbers reported at the time. This is very far from their goal of signing up 60 a week with growth spread largely between single-family homes and businesses. The growth in MDUs has stagnated, likely hitting the saturation point as they were the first adopters into the system with take rates nearing 53%. This stagnant MDU customer base, which accounts for over half of iProvo subscriptions, experienced a 8.2% drop in ARPU during this same period, from $12.74 to $11.70. Not only are MDU numbers halted, they're becoming less valuable as the high churn causes a large number of disconnects and reconnects.
The picture is rosier when looking at single-family homes. Subscriptions during the same period increased about 2.5% and the ARPU jumped about 2%; it's the opposite picture of the MDUs and customer loyalty seems to be stronger with long-term residents. The take rate is now up to 23.8%, a figure that leaves a lot of room for growth. In the last year, there has been no net loss of subscribers on a month-to-month basis.
Business accounts are gaining a significant amount of strength. Not only did subscriptions jump a whopping 11.5%, the ARPU shot up 10.3%. Each business is now worth more to the network than 8 MDUs, yet the commercial take rate is a scant 16% leaving a lot of room for improvement. A significant number of businesses have switched both their Internet and phone services to an iProvo provider. In the last twelve months, only one month saw a decrease in subscribers from the previous month though the following month made up the difference.
Installation Costs
There's also interesting number on the installation costs. iProvo current eats $800 on installation, $500 to run cable from the curb to the premises and another $300 for the "portal" device used to interface with the network. This installation is over split over two days with iProvo handling the physical connection to the home and the service provider doing the installation inside the home. The consultants reports have noted that this bifurcated installation has lead to higher costs and a lower rate of initial customer satisfaction. I know I already dislike taking a day off to meet an installer for a "noon to 5" window, but doing it twice is just silly.
Some good news is that once a location is installed, it can receive service with few, if any, installation costs. According to Council Member George Stewart, chairman of the Committee, the portals are left on-premises when service is canceled and, per his knowledge, can be activated remotely without the need for a visit from a technician. This means that as the network grows, the installation costs to the network will decrease substantially.
So what can iProvo do in the meantime to improve cash flow? The consultants recommend using third-party technicians that assess the cost of installation to both iProvo and the retailer, but I think they need to look at UTOPIA's new plan for some ideas. Ideally, the retailer should bear all installation costs and offer customers the option of paying an up-front fee, rolling it into the cost of monthly service or some combination of the two. The retailer could also opt to eat part or all of the charge in a cell-phone like model or offer differing packages to give customers more options. With additional retailers, this could lead to interesting models where installation can be borne by the customer, the retailer or a combination of both.
Overall Financial Picture
Provo's problem appears to be a matter of cash flow caused by high churn that causes installations to not pay for themselves in the numerous MDUs that subscribe. At the current ARPU, an MDU takes a whopping 69 months to cover the costs of installation, longer than the time it will take the average student resident to earn a degree from BYU. A single-family home takes 22 months and business accounts cover the cost of installation in just 9 months. By moving the cost of installation into a combination of up-front fees and increased service costs, this will improve cash flow substantially for future customers, yet it still doesn't help with the existing customer base. With current growth trends, the subscriber base simply will not grow to cover the deficit that iProvo faces if changes aren't made.
Enter City Savings
The financial picture seems bleak until you start looking at the potential benefits of the network to various city departments. The Energy Department stands to eliminate costs of around $704,000 per year through automated meter-reading and remote connects and disconnects of service. Even more exciting is the potential savings to Provo City Power customers. The fiber-optic system will give the city immediate feedback on power outages and allow them to remotely turn on and off devices during peak usage, much in a manner similar to Rocky Mountain Power's CoolSaver program. This means power customers will save upwards of $5.5M on their electric bills or $161.76 per year per household. This savings more than covers the shortfall from telcom subscriber revenues and results in massive savings for the citizens of Provo.
The water utility also wants to get in on the remote meter-reading game. They stand to save around $250,000 per year and will, for the first time ever, be able to read meters in the winter. They can also get instant feedback on water quality, pinpoint potential leaks and see water pressure statistics in real-time. Combined with the power savings, Provo could end up saving upwards of $1.5M annually via the use of the network with an investment of $8.9M in new meters for both utilities. In the long run, the network makes a lot of sense of the city.
This just scratches the surface of the potential savings. The smart management of the power grid could help reduce demand for new capital-intensive power plants and help reduce air pollution in the city. The school district could save $78,000 per year with a switch to a new VoIP system. Electronic parking meters and traffic light photo enforcement can bring in anywhere from $180,000 to $275,000 annually. These just scratch the surface of how the city can save far more than it currently spends on the network.
New Revenue Streams, New Providers
The consultants also made a number of suggestions for new revenue streams including burglar alarm systems, co-location, off-site data backup and more. The biggest change they found to be critical to subscribing new customers, especially the lucrative business customers, was to add new providers. I hate to say it, but I came up with that last one back in December. This is going to be especially important as Franklin Court believes that neither of the current providers will stay solvent long enough to continue to provide services for an extended period. The prospect of another HomeNet fiasco is disturbing, especially given the vacuum that it created in its wake.
Back in December, we were told that Provo was negotiating with several providers, three who had submitted an RFP and 5 who had merely expressed interest. Two months prior, the Daily Herald told us who those providers were that had submitted an RFP: Emery Telecom, FiberNet, and Utah's favorite home team, XMission. These long delays in adding new providers need to be overcome in order to retain customers and attract new ones, especially the business accounts. I've heard many stories from iProvo users that they would really like to get XMission in their homes, especially since they now offer a VoIP product.
Part of the problem may be Provo's own doing. They current stipulate that all providers be triple-play capable and dictate what Internet speeds providers can offer. This would likely scare off ISPs and phone providers with little experience and even less will to do video.
Strength in Numbers
Another consideration should be joining the network to UTOPIA. Given that both existing iProvo providers currently operate on that network as well and XMission has been in negotiations, it should be a problem to transition the customers over. They're also already leasing services from UTOPIA's fiber-optic backbone, so there's an established working relationship.
You might wonder why this makes sense. Based on a conversation I had last Thursday with an employee of a UTOPIA contractor, they've been in negotiations with a major triple-play provider (I'm talking millions of customers) to join UTOPIA. They said they'd join if UTOPIA hit 50,000 served addresses and bring with them a reputation for excellent customer service and extensive experience in video, voice and data services. Since Provo can't grow much past 34,000 served addresses, they could only gain access to this provider if they joined UTOPIA and became part of a larger network. Another major double-play provider is also interested in joining if the network grows to that size. Provo should keep its options open in order to gain access to these nationally-recognized providers and their expertise.
Further Consultant Recommendations
There were a number of other problems that the reports also touched on. It seems that the telecom department had a poor working relationship with the retailers and infighting amongst divisions of the department is common. All too often, employees are unitaskers and do not assist with other tasks as needed. Fixing this problem will require a significant culture change and likely someone to fill the shoes of managing that department. There's also a recommendation to start cross-training employees to allow more flexible staffing. This includes training electrical employees and fiber installers in each others job functions.
It was also noted that retailers regularly use iProvo staff to troubleshoot problems on their end. This results in a large cost to the city which can be alleviated by allowing retailers to directly provision services on the network, make small service changes on their own and start charging for support when the issue is deemed to be the fault of the retailer. They also recommended consolidating the Energy and Telecom NOCs into a single unit with appropriate chargebacks to each department and reducing the NOC hours from 24/7 to normal business hours.
Both consultants also recommend opening up the full 100Mbps of bandwidth to each premises and allow retailers to figure out what speeds they can offer. Going to a "dumb-pipe" wholesaler will allow retailers to be creative with the packages they offer and will make the network more enticing to new entrants. Certainly this kind of plan also needs to be paired with a plan to upgrade to 1Gbit connections at each serviced address to support additional services.
One recommendation by CCG is to have the city start paying to reserve some of the network capacity for future use. This seems like an unnecessary move given the high capacity of fiber and the ease with which the connections can be upgraded to 1Gbit or 10Gbit speeds. I feel it's more of a stop-gap "make the loan not a loan" measure that, regardless of acceptability in the marketplace, will not be politically feasible in Provo.
Conclusions
While a lot is made of iProvo's up-front losses of around $2M per year, it would appear that the potential savings and new revenue for both residents and the city of $8M+ per year far exceed these numbers. Even so, it behooves Provo to take some decisive action to stop the up-front losses with a concentration on adding new providers to the network to increase subscriber rates. This means decisive action, not more of what we've already seen. Provo's leaders need to take swift and prudent action to turn iProvo into a truly open network with vibrant competition between providers and few restrictions on services being offered. Anything less means that the success in overall savings will forever be overshadowed by the losses on subscribers.
12 Responses to “ The State of iProvo: Network Will Save City and Residents Millions Each Year ”
Comments:
Leave a Reply
Trackbacks & Pingbacks:
-
Pingback from Opinion Piece in the Daily Herald » Free UTOPIA!
May 2nd, 2008 at 11:54 am[…] opinion piece is adapted from my more in-depth analysis of the situation with iProvo and was published in today's Daily […]
April 27th, 2008 at 9:06 pm
Nice write up Jessie! You have to believe that there are real cost savings that could be achieved by combining some of the operations of UTOPIA and iProvo. Clearly a larger customer base would be more attractive to future service providers.
April 28th, 2008 at 12:13 pm
Very nice write up
We already see the benefits for public infrastructure (like roads) and I haven’t met anyone who isn’t interested in Utopia or iProvo.
The biggest problem I’ve seen is people don’t know about the service being available.
A little advertising would go wonders
April 28th, 2008 at 4:01 pm
An interesting summary of the meeting. I found the reference to XMission as “Utah’s favorite home team,” puzzling. I’m not sure what that means. In UTOPIA they’ve gained little against the other service providers. I doubt much would change if they were on iProvo. Quality of service is impacted more by the networks than by the providers so iProvo users wouldn’t see much difference, if any, compared to the providers already there.
I don’t see how the network will ever attract new providers, especially a large nationally recognized one when the profit margins on iProvo are so small compared to what the big national providers are familiar with. Why would the big boys want to come and play where margins are less than 20% compared to the 60% or better margins they make today. Their size doesn’t bring scale on the iProvo network so they would have to market like the existing providers, or lose money for a very long time before generating one third the margin they do elsewhere. It simply doesn’t make any sense under the current wholesale model Provo offers.
Personally, I believe this is why the network is struggling. There is no financial incentive for service providers. There is no scale on their wholesale model. A provider can have 100 customers or 10,000 customers but the wholes price remains the same. I’ve never seen a wholesale pricing model that treats all customers the same no matter how big they are. Is it any wonder that the service providers spend so cautiously on marketing? When that is fixed then providers can see how to make money and new providers will be more serious about joining.
April 28th, 2008 at 5:41 pm
KLT,
Utah has a fairly large technical population and XMission is a very technical friendly company. However they also in general have very good support and a good setup of their own. Yes the Fiber Network be it iProvo or Utopia is the same from Provider to Provider but they all choose how to bring the customers out to the rest of the world, and as a general rule I think XMission does a better job.
The difference between profit margins on iProvo and Utopia vs. a network they build and operate independently is that they don’t have the cost of building the network. The only cost of getting a new customer on the network is advertising / getting them to sign up.
I do think the Municipal model is a good one, though it has some hurdles to overcome, I hope it makes it.
April 28th, 2008 at 6:36 pm
Luke,
I’ve spent over 20 years in technology and I’m very familiar with XMission, but when it came to picking a fiber service provider I found Mstar to have the superior connection to the Internet. Off-net I would agree Xmission does a better job, but on-net Mstar seems to have the superior solution. I’ve had their service for a year and only had a couple outages, and both were UTOPIA, not Mstar.
I too hope the municipal model works and can be a good one, but they will need to see it more like infrastructure, similar to roads, water lines, etc., if it is going to work. If the cities are trying to profit from it they will kill the project.
Your comment that the service providers don’t have the cost of building the network is false. They just don’t get the benefit of paying for it. Today the service providers are paying a large amount of what they collect to be on the network. The cities don’t plan on paying for it unless they have to use the taxes to cover the bond. My understanding is that to date no taxes have been used.
The majority of what a customer pays for services is going to the network to cover the cost of construction. It seems that a service provider would rather have the cost of construction because eventually construction is paid for and now the profits increase dramatically. Service providers doing business on-net never have that kind of scale because they will always pay that access fee. That creates a multiple year pay back for service providers today with nothing changing in the future. The big question nobody is asking is how do providers ever get whole? Until someone answers that question they can keep bringing on new providers and they will keep on doing the same thing. I don’t know how the networks think they can get different results when they continue to do the same thing. The networks and the service providers are going to have to find a middle ground where both can win. Right now it looks like nobody is succeeding and if the providers can’t succeed then the networks fail. I seem to be one of the few who actually see that.
April 29th, 2008 at 12:45 am
KLT: Most Utahns are quite happy with XMission as a retailer, especially residences and small businesses that I’ve met. MSTAR, on the other hand, has been the subject of mixed reviews just like Veracity/Nuvont.
Something that was discussed at the Telecom Summit in December was an incentive for retailers to meet specific service levels and enjoy reduced wholesale rates. The consultants reports made no mention of these proposals and I’m left wondering if they evaporated into thin air. I certainly agree that the model must be made attractive to retailers.
iProvo will always have trouble attracting new service providers because it can’t grow beyond its current size. Some of the 20 providers that UTOPIA is currently negotiating with have stated their intention to join the network once a certain number of homes (usually 50K) have been passed. Provo can’t break this number without banding together with a larger network because the city has no physical room to grow that much.
While your points on network construction are well-taken, you can find a low-speed version of UTOPIA in the dial-up market. It was an open network that allowed literally dozens of competing providers to offer similar services at similar price points and many of them made good profits until broadband came around. While the big cable and phone companies would certainly rather enjoy the vertically-integrated monopoly, small-to-medium companies appreciate being given a ready-to-use infrastructure. There is certainly precedent for successful retailers that do not rely on owning the wholesale end.
May 2nd, 2008 at 10:02 am
I have a hard time believing some of the reports’ projections for saving for such things as remote electric meter reading and connect/disconnects. My concern is the cost to set up such a system. There has to be a cost to put a new meter in every home that allows reading over fiber, plus a cost of $800 per home or business to connect each location that isn’t already an iProvo subscriber.
On the other hand, if it’s possible to sell the iProvo network to a private party, then the private party has some reason to believe there is significantly more value in the network than the price paid. If the city thinks it can get 50 cents on the dollar for the network, I would expect the buyer to have a plan to make it worth the whole dollar.
May 2nd, 2008 at 10:47 am
Jared - The city doesn’t have to hook up every business or home with fiber to read meters. This can be done wirelessly which is very inexpensive. There are more than enough APs in homes and business in Provo to pick up the wireless feed and send the data back to the energy department. Now factor in the cost of this solution vs. trucks and maintenance, meter reader’s salaries and the soaring cost of fuel to stop and go all day long. There is a big savings once the city breaks even on the meter reading equipment and that saving goes on for years with minimal maintenance to the infrastructure.
Remember if Provo sells the network for less than what they have into it, 50 cents on the dollar as you suggested, the citizens of Provo then will have to pay off the balance of the bond through taxes. That would be about $20 million plus interest. At that price I’m sure there would be many buyers interested. This is a state-of-the-art fiber to the home network. One of the best in the nation, similar to what the rest of civilized world is deploying. If sold to a private party at any price, the middle man (the city of Provo) is removed and the new owners could be at break even with far fewer than 10,000 customers. The plan is really simple if the owners are also the service provider, which is the other thing citizen in Provo will give up - there will no longer be a choice of service providers on the network.
May 2nd, 2008 at 12:15 pm
The problem with wireless meter-reading is that you either have to pay someone to drive up and down the streets to pick up the signal OR you have to install a lot of access points throughout the city (again, leveraging the fiber infrastructure). Even with all of this, wireless is slow and unreliable by comparison, subject to all kinds of electromagnetic interference. The cost to do wireless meter-reading is about the same as pulling fiber to each meter.
As far as selling the network, it seems like a bum deal for Provo to consider doing so. While they might be able to recover the principle still due on the bond, consider that a large part of the network was paid for by federal grants to monitor traffic and improve air quality. The feds might not look kindly on such a deal.
May 2nd, 2008 at 5:16 pm
I don’t want to offend your intelligence but it is a very simple solution to set up the access points throughout the city. Every Access portal in every home and business is a potential access point. Simply plug a wireless receiver into an AP and all meters within range link and send date over the fiber back to the city. Todays wireless solutions can reach many miles and the city isn’t that big so with only a hundred receivers hooked to existing APs, at a max, they can do the meter reading. The main expense is retrofitting the electric meters, but they are physically reading them today so that just require the city workers to instal the transmitters while they are there.
You say wireless is slow, but the data being read is minimal and could all be done city wide in a few minutes. What is slow is driving around the city from house to house manually reading meters. The older wireless technology was unreliable and prone to interference.
I’m working with some wireless technology today that can deliver 40Mbps to 60Mbps up to 15 miles and is very reliable. It has been used by our military and is now coming the market. It will not replace fiber but it has really advanced over what you describe.
I don’t know the actual cost to do a wireless solution for the city because there are so many variables. But I do know that even if it was the same as pulling fiber to each meter, it is a one time expense. Manually reading them is a cost the city will have month after month, year after year with no break even on the expense.
If Provo decides to sell they will do it to cut their loses, but they will not recover the expense put into it and they will be unlikely to even cover the remaining principal. If a bond is like a tradition loan all payments paid to date have gone primarily to interest. I doubt the principal has been reduced much given it was a 20 year bond and they aren’t a quarter of the way into paying it back.
I have no idea what the feds might think or say. I’ve never heard that any of it was paid by federal grants. My understanding is the bond is backed 100% by city taxes. UTOPIA used grants, but my understanding is that to date the federal government has never actually given UTOPIA the money promised, so they too are 100% guaranteed by city taxes if service providers can’t generate enough positive cash flow to stay in business.
May 2nd, 2008 at 9:15 pm
The problem with the wireless technology you describe is that it sounds proprietary. iProvo was built around open standards (like Ethernet) and trying to couple proprietary wireless tech onto it seems anathema to its current design, especially if they require the use of a licensed range.. That seems to leave standard technologies in unlicensed spectrums like WiFi and WiMax on the table, both of which come with signal problems (just ask Earthlink) and limited range.
When I say wireless is slow, I don’t necessarily mean throughput. Latency on wireless is still an issue, especially when compensating for weak signals and dropped packets. It’s also significantly less secure since data is traveling over the airwaves. With fiber, you’ll always detect an unauthorized splice.
Since somewhere in the range of 12,000 homes are already wired to the existing fiber network, it seems to be a better use of resources to wire the remaining 22,000 and go from there instead of doing another overbuild.
iProvo started as a city-only network with federal grants. The bond to turn it into a municipal wholesale network for voice, video and data piggybacked onto the existing rings. Saying that the city should scrap the entire network over the telecom division makes no sense at all since city money didn’t even pay for the majority of it.