Tales from the Towers, Chapter 37: Talk is cheap but Ebay is cheaper

Anybody starting a business knows that funding is the biggest problem.  Angel investment and venture capital for first round funding are about as scarce as shade in Phoenix.  That is of course, unless your Uncle RUSs or Aunt CAFee open up the government (meaning poor taxpayers) purse strings (lookup OpenRange, Main Street Broadband) to people who haven’t quite mastered the intricate and mystical techniques of the Excel spreadsheet with the common sense hyperlink.  Apparently passing a Civil Service Exam, getting elected President, or getting a political appointment by a Community Leader (for definition see President) out of an ivory tower classroom (see Solyndra, Amonix, Solar Trust, Evergreen Solar and 10 others that I know of) makes you think you are smarter than educated equity investors who have spent years learning business management and analysis skills.   Since OBAMACARE has already passed and OBAMAGREEN is an unmitigated taxpayer disaster to the tune of at least $30 billion dollars (they should have hired Bain Capital to make these investment decisions, at least they were profitable), it looks like OBAMANET is next on their list because the others have worked out so well.

To start my first business, I sold a motorcycle for $1000.  Even 20 years ago though, that wasn’t going to do much so I had to make that $1000 not only go as far as possible, but to make my podunk Peoria living room office appear publicly as professional to my clients as Microage (RIP) on the 4th floor of  Phoenix corporate center.  To that end, I dumpster dived the Radio Shack computer centers and came up with printers, monitors, a tape backup, and the phone/computer version of the Crown Jewels, the PC based voice mail system board.  The PC voice mail system supplemented a phone system that would have been far more than my startup budget alone.  When you called my number, you got the following and today mostly annoying phone menu with the following:

1)     Push 1 for Sales

2)     Push 2 for Technical Support

3)     Push 3 for Accounting

4)     Push 0 for the Operator  (who will coincidentally sound the same no matter which button you push)

I either answered the phone on all options until I could afford an office person or it went to voice mail for the appropriate department.  This was big stuff in 1990 and got me into some large accounts fairly quickly.  I basically presented a company to the public that looked like it was more than a one man Joe’s garage and computer repair.  Not all of us can be Apple but they showed it’s possible.

Startup WISP operations with limited budgets can do the same thing.  We covered the concept of making an angel investment look good for a second round of investment funding.  However, what happens if you can’t even get to the angel investment level and have to stretch even penny like it it’s the last Twinkie on earth (my money is on the Twinkie)?  In addition, you might be competing with entrenched providers that make customers more comfortable about long term viability.  There are very few areas where you live and not run into some competition.

As a refresher, I’m helping a friend get his WISP started up.  His budget is based on how big his paychecks are.  That means it’s taking about 4 months to get the basic system running based on a budget of about a few hundred dollars a month.  Because of this, we are analyzing every connector for cost.

The plan is to start with a company presence that demonstrates financial stability and confidence.  Fortunately hanging around the dumpster isn’t the only option today as today anyone can set up an IP based phone system for $30 a month.   What is important is that you get someone with a professional sounding voice to do your announcements.

Computers and network equipment are best purchased on Ebay or reputable surplus computer stores.  My favorites are dual-core computers like IBM/Lenovo T60/61’s or any of the Lenovo dual-core desktops.  Loaded with the operating system should still keep you below $200 for any of these units.    I just bought a T60 for $100 a couple weeks ago in fact.  I finally upgraded my T43P (which will go onsite somewhere as a local monitoring device running Dude) a couple weeks ago and I’m typing this article on my new (to me) T61P.

None of my desktops, sans monitor, cost more $150 and for running accounting/office/web apps, run great.   HP switches cost as little as $15 on Ebay and come with lifetime warranties.   I haven’t paid more than $200 for a fully loaded IBM Rack-Mount server for years and have yet to have one fail.   If it does, I have 2 more on the shelf that I paid about $100 (shipping gets you here though, have to watch that).   Antivirus is supplied by Microsoft (for free) although I use Symantec for my Exchange Server.

Business cards are also cheap with companies giving away 250 business cards for free.  You can even print your own out that look professional.  Letterhead is the same way.  Conservationists would cry if they saw how much old letterhead, brochures, and sales material we would throw out when we moved.  A good laser printer or Inkjet is a much better investment today in the digital age for small printing volume.

Next get a couple company shirts and hats.  First impressions are important so meeting a client with a company uniform is impressive.  The more professional you look, the more they respect you and the more you can charge.  Although I’m partial to Bermuda shorts and a tank top in Phoenix, my business dictates something more professional.

The last part of presentation is your vehicle when you do an installation.  Don’t’ show up with a fold out ladder in the trunk of your Ford Taurus.  If you don’t own a van or minivan that you install stickers on, rent one for the first few installs and slap on a magnetic sign when you go onsite.  Another option is to try to find a really cheap cargo van that might have a cage and shelves already installed.

Our first cargo van was a Ford E-150 with 100,000 miles.  It had cages and shelves already installed but and cost about 1/4 of what I would have paid for a new one.  A couple hundred dollars on Ebay for a factory cruise control and we were set.  That van has exceeded 240,000 miles (with Mobil 1 Oil changes every 7-10K miles) and except for an intake gasket and brakes, is still going strong.  If you can’t budget that, look for a really cheap pickup truck with a shell or cheap cargo mini-van.  My next vehicle is going to be something like the Ford Transit although I might pick up a used Chevy HHR  (I really like that 30+ miles per gallon). Also keep  a box of Milk Bones in the van which really makes a good impression to dog owners.  Claritin is my favorite cat option, for me, not the cat (I’m allergic).

With this tight of a budget, not only do we have to watch startup costs, it’s also critical to make sure that monthly costs stay within his current expense level.  It’s a chicken or egg problem.   Since your deliverable product is data circuits, those have to be on the shelf so to speak, to open the store.  That means a commitment of some sort of contract.  Since we know our maximum monthly budget is a few hundred dollars a month, we have to find cable or DSL circuits to be competitive unless you are in a very remote area and the clients have not other options.  In that case, you can get a T-1 circuit and be competitive with satellite which I’m about to do for the second time.  Opening any business involves licenses, corporate filings, etc… but we are going to focus on the installation hardware and the monthly technical costs.

If you start the business out of your house and can talk the neighbors into not calling the zoning board, you can put up a pole on your property as high as 40’.   Rigid galvanized pipe sections can be bought inexpensively at any plumbing or fencing supply house.  We bought 2-7/8”, 20’ sections for about $130.  If you are only planning on going up 20’ or just plan on putting an omni-directional antenna on top, then you can use a lighter gauge.  In our case, we wanted 30’ and the ability to add 4 sector antennas later.  However, I’ve done 40’ with 4” aluminum pipe successfully with a rigid 10’ inner sleeve.  Where you mount the pole against the house and how high the upper brace is also dictates how much wind load you can handle or how thick a pipe you need.  A good rule of thumb is to never plan for more weight than your friends can handle.  Whatever you do, be safe.  If it’s not easy, don’t do it.  On one project, the local penal institute was more than happy to bring out lots of assistants in orange jumpsuits so keep your options open.

Pole bracing can be handled by channel bar against the house.  Here is another area where it has to be really thought through.  A couple wood screws holding channel bar to your stucco wall with a 2×4 stud behind it isn’t going to cut it.  Best to find a place where you can use all-thread, go all the way through the wall, and put a second section of channel bar inside, wedged around 3-4 wall studs.  Channel bar connected inside the house typically doesn’t match the indoor décor though. This is also a little tricky to get by the wife unless you are big into tapestry.  A garage might also be a better place to brace against.

To make our pole easy to take down and put back up, I like this method.  If you can dig, put a second bracing pole 5’ in the ground with the main pole between it and the wall.  This bracing pole would be better with some concrete but if that’s not possible, go deeper.   In Arizona, digging a 5’ hole in some places either involves heavy machinery or explosives, both which my wife took away from me when we got married (to my Homeland Security friends with no sense of humor, that’s what they call a joke).  It’s also probably not worth the fear of letting any of your buddies who have the highly tuned physique of a Weeble developed from years of intense WarCraft battles, swing a pick-axe within 6” of your outer wall.   The end result should be the channel bar clamping the pipe at the top and a second channel bar below it, the all-thread goes through the buried pipe at the bottom and through the lower wall/channel bar.  The pipe can then be tipped over pivoting on the lower all-thread rod by 1 or 2 guys when needed.

Add a couple of 3-5’ grounded copper rods around it which are available at Home Depot.   Use some 2-6 gauge copper wire to attach the rods to the pole and bury the wire several feet to the poles.  You should also attach it to the house ground if there is a connection available.  I’m sure I’m going to get hammered on this by the tower pros but it’s more than sufficient for a start-up application to get the business off the ground.  I know I also shouldn’t have to say this but I’m going to anyway because I’m partial to not  getting sued, this isn’t for climbing. Home Owners insurance is also up to you.

If there are a lot of trees around, the minimum I would start with is an Ubiquiti 2.4GHz 13dbi omni-directional antenna and a Rocket M2.  If there are very few or no trees, then start with a Ubiquiti 5HGz omni antenna and Rocket 5M.  This should be sufficient for about 30-60 clients depending on bandwidth.   The 2.4GHz will get through the trees better but 5GHz gives you less interference and more channel options later.  If you have lots of trees and 2.4GHz isn’t working, the only other real option you have is 900Mhz.  WiMax is available but with a budget that couldn’t afford one CPE per month, that isn’t happening.  900MHz is not a good choice unless your yard looks like the Amazon Jungle either because our costs go up on the AP and CPE side.  Even worse, if you are in an area that case DSL and/or cable, then you probably have some type of 900Mhz interference.  Either way, it’s probably going to cost you one months of budget to figure out which one you are going to be able to use.

For testing purposes, start with the Ubiquiti Nanostation M5’s, which are good for getting a feel for distance and interference.  If you have trees, get a pair of Ubiquiti Nanostation M2’s and run some distance and link quality tests.   If the M2’s won’t get far enough, get one M900 and first use it simply for signal interference testing.  AirView, which is built into the firmware of all these radios, will give you an idea of whether or not you want to even attempt to play in any of these bands.  In 900MHz, if the interference level is below 70dbi, and you will have to watch every direction for about an hour or more (in some cases data collection may only be done once per month), then get a second M900 and start testing range.  All data on range can be extrapolated from a pair of radios.

Egress bandwidth is easy.  Find a DSL wholesale provider that doesn’t have a reseller EULA restriction.  It won’t be cheap, probably about $100-$200 per month for 3-6Mbps.  That may sound like a lot but it’s not a bad starting point.   When you start to expand, you can get more circuits and use a load-balancing router.  When the revenue is sufficient enough, you can move up to a standard fiber circuit.

In my opinion, and believe me, it’s not universal by any stretch and may have some legal issues, blocking illegal torrents is the best way to stretch your bandwidth.  Throw in a cache and bandwidth stretches even farther.  Block all virus and malware traffic or as much as you reasonably can, and your bandwidth goes even farther.  I’ve talked about using a Barracuda Web Filter appliance as the easiest way to do it but you can also get a Barracuda remote cloud device with Web Filter and 50 licenses (minimum you can order) for about $1200.  That’s a lot for a startup but keep in mind that it can get you a 20-1 ratio or higher if you don’t have a lot of NetFlix users.  At 20-1 and $25 per month for a 1-2Mbps service, you are still doing pretty well for a startup operation.  At 50 users, which matches your license limit, your monthly revenue is $1000 which about covers your licenses in 1 month.  License renewals are about $15 per client per year.  Considering you aren’t going to do better than 10-1 without this or some other filter, it’s a good investment.

Don’t quit your day job yet though.  Get the foundation built and gain some experience.  Better yet, make sure that you have the desire to do this full time.  This isn’t like the drums that your parents bought you that you gave up after a few months.  Once you start this, you are now the lifeline to the world for some people and that’s a big responsibility.  Keep telling yourself that you are now the local internet utility company.  If they go offline, they are putting their trust in you to fix it.  If there is a major storm, you might need to bank some sick days to take the day off and get things up and running as quickly as possible for your clients. Telling them you will fix it after work isn’t the most professional of excuses.  For example, we had a tech drive 7 hours round trip on a Sunday to fix a single radio on a roof today.

When you fund your own business, you accept you are going to make sacrifices, whether it’s financial, time, or family.  When the government hands you millions of dollars, you tend not to make those sacrifices.  Although I understand that many of the solar companies went belly up partly because of the government’s inability to show a spine to China, many more simply had uncompetitive market technologies.  WISPs are different.   Just as China targeted our fledgling solar industry, our own government, through the USDA, the RUS, and CAF are targeting the WISP industry. Even worse, these puppet agencies are letting CenturyLink, Verizon, and AT&T pull their strings.  Hopefully that doesn’t slow down the entrepreneurial spirit of people like my friend that still believe in the American Dream of creating your own business.  He should have the opportunity to compete fairly against other competitors instead of his own government choosing sides.

PREVIOUS POST: Tales from the Towers, 36: A Take on Ubiquiti

NEXT POST: Tales from the Towers, Chapter 38: Pick a Tool, Any Tool – what every wireless ISP should know

 

About Rory Conaway

Rory Conaway has been in the IT and Wireless Industries for the past 25 years as an author and consultant. He currently operates a growing WISP operation in Southern Arizona. He consults with investors, manufacturers, and WISPs, and develops financial business models for startups. In addition to writing articles in industry publications such as Mission Critical Magazine, Mr. Conaway also writes the series “Tales from the Towers” that can be found on various such as www.triadwireless.net and www.muniwireless.com. He has also engineered several wireless designs such as S.P.I.R.I.T. and Guerilla Wireless as well as building integrated wireless and video surveillance for airport security, municipal and critical infrastructure, SCADA systems, and hotel/MDU deployments.

Comments

  1. Rory, as simple as 1-2-3. Reminds of all the little things I had to do back in late 1998 when I assembled a home/business 100baseT network with 8 upgraded 1990-1996 vintage PCs and used my Libretto sub-notebook (T80) as the server. How many hours did I spend on the phone with Microsoft, Compaq and Netgear support!?! Keep it up man. Michael

Speak Your Mind