dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
9956
share rss forum feed

Telecomnet

join:2010-02-10
united kingd

Initial investment for setting up ISP

I have been asked to find the initial investment to setup an ISP. I have divided total investment into start-up costs and operating costs. I have also found some key items like equipment cost, marketing cost, salaries, rent, internet backbone cost etc. Can someone help me to find tentative costs for those items. I will appreciate if someone can add other key items and related costs. Is there any websites I can go through for help. I have been searching from last couple of days but no luck. Any help will be appreciated. thanks


jimbouse

join:2011-10-01
Bryan, TX

$10 million dollars....

Make the check out to "Jim Bouse".

.... Seriously .... How are we supposed to help you with your question if you don't give us any idea of the area you are trying to cover, the equipment, the speeds, your geography, etc....

I started my WISP on about $5000 (but I have a full-time job to pay the bills as the WISP grows). I am slowly (6 Months) getting to the point where the WISP is paying my personal account back.


gunther_01
Premium
join:2004-03-29
Saybrook, IL
reply to Telecomnet

»WISP Forum README
--
»www.wirelessdatanet.net



WHT

join:2010-03-26
Rosston, TX
kudos:5

1 recommendation

reply to Telecomnet

You need at least enough to get enough subscribers on line in 6 months.

You really really need a consultant to help you with this. Asking about a several hundred thousand dollar invetment in a public forum is a poor idea.



Inssomniak
The Glitch
Premium
join:2005-04-06
Cayuga, ON
kudos:2
reply to Telecomnet

Yea it's a really broad question based on so many things. I've heard ppl fail a wisp with 50 grand, yet I started mine with a thousand bucks and succeeded.
--
OptionsDSL Wireless Internet
»www.optionsdsl.ca



John Galt
Forward, March
Premium
join:2004-09-30
Happy Camp
kudos:6

1 recommendation

reply to Telecomnet

Expect to spend money like a "fire hose"...and get it back like a "eyedropper", especially if you're starting from scratch.



Lightwave
Premium
join:2010-06-03
Tilbury, ON
reply to Telecomnet

Lets see... $145 for two Linksys routers, $8.44 for 2 tuppeware boxes (to house the routers outside) and existing DSL connection... so..... I suspect $150.00 if you get the tupperware on sale! 4 yrs later, 4K customers + and its self sustaining, never borrowed a dime. I think I might have tossed in another $500 or so for a couple of Engenius radios, but all the same, and tupperware was actually "ZipLok" brand.. still have them

So can be done... but YOU might wish to eat during that growth phase, so perhaps a few more bucks in the bank will help.



Inssomniak
The Glitch
Premium
join:2005-04-06
Cayuga, ON
kudos:2

said by Lightwave:

... but YOU might wish to eat during that growth phase, so perhaps a few more bucks in the bank will help.

Are you saying you didn't?
--
OptionsDSL Wireless Internet
»www.optionsdsl.ca


Lightwave
Premium
join:2010-06-03
Tilbury, ON

Kraft Dinner & Inlaws


OHSrob

join:2011-06-08

2 edits

1 recommendation

reply to Telecomnet

We started ours with ~$8,000 and about 4 servers, 2 desktop's and 5 switch's. One netgear managed switch, 3 cisco ios switch's and 2 hp procurves and all of this I had acquired from ebay over the years.

That said 2 years later and 86 subscribers with 12 sites later and im still not making any money just paying the company's bills and putting it all back into the company.

I don't expect to make any money for another 3-5 years, It is VERY expensive to do things right and as you grow the cost to do things right also grows. I am anticipating running out of bandwidth on 2 key backhaul's and needing to bring in more ethernet circuit's from bell to key towers within 1-2 years at present growth.

Also don't screw around, Do things right and do things once. And don't try to do things that you can't offer it just makes you look like a joke and gives all unlicensed wireless a bad name.

I started my wisp because the wisp I previously worked for refused to do things right and insisted that everything was done on a shoe string budget with equipment from the local office depot.

I kid you not they refused to buy a managed switch, if it wasn't linksys or dlink and featureless they wanted nothing to do with it.

When shit hit the fan and their 6 year old tranzeo backhauls started to fail and began to require hourly reboots to keep them up and over 300 people had internet down more then it was up.

I was told to lie to the subscribers faces and tell them that the issue was at their house or with their routers and my boss's told me that we were never getting the equipment to I needed to fix it there mess they called a network. (all they cared about is if they were getting payments from people not if the internet worked right).

Before that even on the best of day's there was tcp congestion control was kicking in during the peak hours across most of the access points, All critical backhauls and most of there internet connections. During the off hours it was even kicking on and off on most of the major infrastructure.

There internet connections at the pop's that fed tower sites ranged from a 2meg adsl line running 30 peoples internet that was marketed as 2 megabits per second down to the subscribers. To 50 meg down 2 up cable lines with 200+ people on each behind nat. (They also had ml-ppp adsl that I had setup with 12 lines in one pop and 4 in another all overloaded but they at least had public addressing).

They were foolish and offered unlimited to people wirelessly. They had subscribers with as much usage as 600GB/mo, the top 10% were all over 150GB/mo. (excluding the top users on the 2meg adsl line). They had one access point with 30 subscribers on it that had 2 subscribers using more then 50% of the bandwidth 24/7 to infringe copyrights.

I had tried from day 1 when I was there to get them to get a 100meg Ethernet circuit and consolidate the 3 pop's and 19 separate unreliable Residential/Small Business internet connections into one. But I was met with only hostility from the owners regarding that as they thought the price of doing business in this field properly was not worth it and preferred to keep the costs as low as humanly possible to maximize profits.

Feed up with their bullshit and completely wrong attitude of just trying to make stuff sort of work for free or near free cost in ways that violate all best practices for networking and radio.

I started my own wisp and vowed to never drop a packet, to always have at least 20% capacity free on all parts of my network during all hours so congestion control will never kick in and To always deliver on my advertised speed and to never compromise on network design or implementation to save a buck.

Ive had to suspend signup's many times to address capacity issues so far until they were resolved. (I consider it an issue if its over 80%). I also do not allow a small percentage of my users to monopolize the radio time by letting everyone go unchecked. I have a limit but unlike other providers around me when the limit is reached I slow the connection to 1.5 megabits per second down and 128k up.

Suspending signup's can be very harmful to the cash flow but it is truly something I feel every isp should do when capacity becomes an issue.

If your passionate about rural broadband and really want to go out and give it your all. And you have a good grasp of networking at the very least. (at minimum know what the osi model is and properly understand it) I would say go for it mikrotik or ubiquiti is a good place to start for the wireless stuff, ligowave has some nice gear as well too canopy is another choice as well if speed is less important to you.

For networking gear hp has a wide selection of quality products as well as cisco.

I recommend you listen to my words and ensure that you do not end up going down the same path as my previous employer did. It will do nothing other then make you and your company look bad.

edit: If I were to start over again I would probably go mikrotik from the start the OmniTik U-5HnD would be a good place to start with 4 cpe's to go with it.


Chele

join:2003-07-23
kudos:1
reply to Telecomnet

Your question is way too vague, bandwidth alone can vary as much as $900/mb within 100 miles where I live. Gone are the days where you could start with just a T1, Netflix and Youtube are the things people care about the most now. Anyone can tell you that you can setup a POP for as little as $350 in equipment, however, equipment will be the least of your problems. Acquiring reliable and reasonably priced bandwidth will be a big issue, even a dealbreaker. Then you have access(or lack of) to good sites, free bandwidth and free gear mean nothing if you don't have a decent spot for a POP. On the upside, equipment options and pricing have improved drastically since we started almost 10 years ago.


jim_p_price7

join:2005-10-28
Henryetta, OK
Reviews:
·HughesNet Satell..
reply to Telecomnet

There's no cookie-cutter template for this.

Most of us in here started our WISPs because there was a need in the community with no one to fill it. So we started with what we had (5K in my case) and made up for what we lacked with hard work and no life.

I still work a fulltime job and only put about 5 hours a week into my WISP. I'm approaching 75 subs and could pay myself a decent salary for my time, but choose not to for the time being. All the $ goes right back into the business to grow it.

Before you even THINK about money you need information and a plan. What area(s) are you going to serve? How many houses will your prospective transmitter sites pass over? What speeds are you going to offer? Who is your competition?

How much to spend comes after you know the answers to those questions, and have used them to formulate a plan for getting from point A to boint B. It's like a trip. If I don't know what I'm driving or where I'm going, there's no way to know how much the gas is going to cost.

Go back to the drawing board.



WHT

join:2010-03-26
Rosston, TX
kudos:5
reply to Telecomnet

Reality check here, »forum.ubnt.com/showthread.php?t=35378


LLigetfa

join:2006-05-15
Fort Frances, ON
kudos:1
reply to Chele

said by Chele:

Gone are the days where you could start with just a T1...

Ja, while it may be nostalgic to read about some of the upstarts from a decade or so ago, you probably wouldn't survive starting the same today unless you were the only game around.

I still shake my head when I read about some WISPs providing sub-megabit speeds and some of the prices they are demanding for it. I left one WISP that provided only 512 kbps when I could finally get 10 mbps.
--
Strange as it seems, no amount of learning can cure stupidity, and formal education positively fortifies it. -- Stephen Vizinczey

jcremin

join:2009-12-22
Siren, WI
kudos:2

said by LLigetfa:

I still shake my head when I read about some WISPs providing sub-megabit speeds and some of the prices they are demanding for it. I left one WISP that provided only 512 kbps when I could finally get 10 mbps.

My "standard" plan is 768k for $40/mo. Show me a way to offer 5 megs for that price in this area without running myself out of business and I'd gladly do it.

LLigetfa

join:2006-05-15
Fort Frances, ON
kudos:1

said by jcremin:

My "standard" plan is 768k for $40/mo. Show me a way to offer 5 megs for that price in this area without running myself out of business and I'd gladly do it.

Obviously there will always be regional variation. In my case the WISP that I switched to is also a cellular carrier that has lots of very tall towers and very deep pockets padded with government handouts. I'd say that may be hard to compete with.

They are also my upstream provider for my hotspot in town.
--
Strange as it seems, no amount of learning can cure stupidity, and formal education positively fortifies it. -- Stephen Vizinczey


DaDawgs
Premium
join:2010-08-02
Deltaville, VA
reply to Lightwave

said by Lightwave:

Lets see... $145 for two Linksys routers, $8.44 for 2 tuppeware boxes (to house the routers outside) and existing DSL connection... so..... I suspect $150.00 if you get the tupperware on sale! 4 yrs later, 4K customers + and its self sustaining, never borrowed a dime. I think I might have tossed in another $500 or so for a couple of Engenius radios, but all the same, and tupperware was actually "ZipLok" brand.. still have them

So can be done... but YOU might wish to eat during that growth phase, so perhaps a few more bucks in the bank will help.

That's what I am talking about.
--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."


DaDawgs
Premium
join:2010-08-02
Deltaville, VA
reply to Telecomnet

Well after having built one or two for other people, I've decided to do one of my own. I'll start with a tilt over tower and $2,500.00. That will probably be enough for me to pick up 250 or so subs at about $20.00 per month. Overhead will be minimal and when I get to that point I will go for a faster connection or just let it run and stop adding subscribers. There are a lot of ways to make the business of reselling bandwidth work.
--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."



totalaccess
Premium
join:2002-10-04
Austin, TX
reply to Telecomnet

As you can read there are several "things" needed for success of building any WISP, including but not limited to: bandwidth, installers, network gear, and so on.

In most cases people start out with 5k-10 investments but have to find a niche market where there is no wifi or wireless. Then you have marketing and sales to think about.

I've build several hundred Wisp's (for others) over the decade, and seen demands flex in bandwidth due to VOIP and streaming video, so beware customers esp residential are extremely greedy and want to pay as little as possible.

Take time and actually do a cost cost analysis study on your WISP. If you don't have one, let me know (message me) and I will send you over a sheet we created to help start-ups. Then once you know your costs (hardware, towers, bandwidth, installers etc) then write a business plan....

Yeah things typically change for Wisp's pretty fast, but basic business rules do apply ( taxes, permits etc) and you need to also have "working" billing solution in place. Hope this helps....

GHz Wireless
"Nothing But Net - 2012"



Jerm

join:2000-04-10
Richland, WA
kudos:2
reply to WHT

said by WHT:

Reality check here, »forum.ubnt.com/showthread.php?t=35378

Great link WHT! Hits the nail on the head!


treichhart

join:2006-12-12
reply to totalaccess

Are you really sure taxes and permits are required? because last time I checked it really depends on the state because I know state of ohio dont require taxes on broadband unless your sell telephone services with it. What permits are you talking about?



teleadmin

@weirvalve.com
reply to jimbouse

Thanks a lot for your reply. I am doing a consultancy project and I am about to finish market feasibility study and other analysis required to setup ISP for 1,000 customer. I have even calculated some of the costs required to setup ISP. But I am bit confused with couple of costs and thought someone who is already working in the area will be able to help me. I just want to know which type and how many routers, switches, firewall is required. Also, the internet backbone cost and if there is any other things if you think I need to consider will be much appreciated.
Thanks



WHT

join:2010-03-26
Rosston, TX
kudos:5

Is your cost estimate up to $80,000 yet?


OHSrob

join:2011-06-08

1 edit
reply to teleadmin

said by teleadmin :

Thanks a lot for your reply. I am doing a consultancy project and I am about to finish market feasibility study and other analysis required to setup ISP for 1,000 customer. I have even calculated some of the costs required to setup ISP. But I am bit confused with couple of costs and thought someone who is already working in the area will be able to help me. I just want to know which type and how many routers, switches, firewall is required. Also, the internet backbone cost and if there is any other things if you think I need to consider will be much appreciated.
Thanks

If you were more serious about this you would realize you can't just jump right into this and shoot for 1000 people.

You need to first find a target market without cable, adsl or a competent wisp (All of my competitors are morons that's why I started where I did), plan and budget your link's and ensure you have clear line of sight + Fresnel zone clearance to get from where there is fiber optics located at a low cost to your target rural market.

You should have this planed before your even considered hardware vendors ore equipment.

The internet is a much more utilized place then it use to be, You use to go by one subscriber per megabit but HD youtube and netflix changed that its now wildly unpredictable depending on how many netflix users you have.

Heck for routers/access concentrators you could go mikrotik, cisco, quagga on a linux distro, juniper, nortel, HP,Alcatel-Lucent, Avaya, Nokia the list just goes on. It all depends on what you understand and how much you want to spend. (and how well you want things to work especially between different vendors).

edit: To service 1000 people properly without traffic shaping the shit out of everything its going to cost you a minimum of $100,000 on parts and internet transit alone in the time it takes to build it to that point.


TomS_
Git-r-done
Premium,MVM
join:2002-07-19
London, UK
kudos:5
reply to teleadmin

said by teleadmin :

Thanks a lot for your reply. I am doing a consultancy project and I am about to finish market feasibility study and other analysis required to setup ISP for 1,000 customer. I have even calculated some of the costs required to setup ISP. But I am bit confused with couple of costs and thought someone who is already working in the area will be able to help me.

Im sure plenty of people can help. And maybe I am a bit of a cynical prick here, but if this is a consultancy project, youre probably being paid to put this report together, yet youre trying to get some of it (the network aspects) without paying anyone else for the consultancy they are giving to you? How is this fair? :-P

Perhaps as a consultant, you can pay some of this money you are earning to a consultant who is more familiar with the topic you are trying to cover here, and that can come up with a couple of network designs using different vendors equipment to cover various budget scenarios, different topologies with different levels of redundancy, produce a BOM for each, etc?

I just want to know which type and how many routers, switches, firewall is required.

Cisco, Juniper, Brocade, MikroTik, and tonnes more. Take your pick. What features do you need? How much is the budget?

How many and what type isnt exactly a recipe you can just cook to, its somewhat specific to each and every network, and influenced heavily by the topology of the network. It all depends on the capabilities you want your network to have, what features you want to offer to customers, the type of technical expertise you or your customer has or can acquire to run and maintain it

Plenty of people around here have built their networks on MikroTik routers and are doing fine. I used to work for a company that built its network using Cisco equipment. Those are about the most opposite ends of the scale as you can get price wise, but both probably have some/most/all of the features that you're likely to need.

As I said above, trying to write a specification for something you understand little about is probably something you are best leaving to a professional who understands it like the back of his hand.

Also, the internet backbone cost and if there is any other things if you think I need to consider will be much appreciated.
Thanks

What is available in the area that your customer is going to service? Rural areas can be extremely bandwidth scarce, and access to any decent amount of bandwidth can alone be enough to make a project unfeasible. Your physical interface options might be limited, as in there might not be a simple ethernet jack to plug your MikroTik router into. You might have to buy a more expensive modular router that can take a T3 or maybe an OC-3.

You will have to determine the providers that are available to you in the area you are trying to service, and contact them and find out what they can offer you to work that one out.

The company I used to work for invested hundreds of thousands, if not millions of dollars building high capacity backhaul through our service areas, as bandwidth from any of the available providers in the area was simply too expensive to buy - too expensive as in more than the hundreds of thousands to millions that *we* spent *building* backhaul...

So maybe you need to do the same and build your own backhaul network over considerable distance to get access to cheaper bandwidth, which will likely be located closer to more major towns and cities.

There is so much that we cant tell you because as I said, this is not cookie cutter type stuff. Though as we have done, we can give you hints. You'll have to do the leg work and fill in the blanks.

rconaway8

join:2005-11-10
Phoenix, AZ

If you are building a business plan that has investors, unless you get to 1000 users paying you $40 per month, don't do it. Anything below that is a one-man hobby so to speak. That being said, the smallest business plan that I would approach investors with will require $300-$400K. That is private angel investors. In fact, I wrote an article on this dealing with single and multi-round investment models and issues.

»www.triadwireless.net/index.php?···emid=272

Growing organically with a small operation is fine but be honest with yourself. Maybe your potential market is X but lack of investment funding to grow quickly and your competitors looking for new areas which might end up being yours, means you may not have 5 years to grow slowly as capital becomes available. If you want to work with institutional investors, you better have some really good friends and a $200M business plan that needs $10M to start. And you better have something different than "I'm going to throw up a bunch of radios on a few towers" mentality.

Justin (WHT) and I have spent months analyzing a business plan with a potential of 3000+ users to try and reduce the Capex. The best we have done is about $250K to cover 500 square miles.



WHT

join:2010-03-26
Rosston, TX
kudos:5

said by rconaway8:

Justin (WHT) and I have spent months analyzing a business plan with a potential of 3000+ users to try and reduce the Capex. The best we have done is about $250K to cover 500 square miles.

And that's not counting 1,000 hours spent in examining housing density per square mile (with 100 foot resolution granularity) for best placement of tower sites. That effort alone reduced the CAPEX by $100,000 to get to the $250,000 figure.

We haven't even gotten to a shopping list of equipment yet. As I have always said...an equipment list is pretty much at the bottom of your list of things to do.


DaDawgs
Premium
join:2010-08-02
Deltaville, VA

1 edit

said by WHT:

said by rconaway8:

Justin (WHT) and I have spent months analyzing a business plan with a potential of 3000+ users to try and reduce the Capex. The best we have done is about $250K to cover 500 square miles.

And that's not counting 1,000 hours spent in examining housing density per square mile (with 100 foot resolution granularity) for best placement of tower sites. That effort alone reduced the CAPEX by $100,000 to get to the $250,000 figure.

We haven't even gotten to a shopping list of equipment yet. As I have always said...an equipment list is pretty much at the bottom of your list of things to do.

I don't know. I'm thinking that you are making it sound a lot more expensive than it has to be.

--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."


DaDawgs
Premium
join:2010-08-02
Deltaville, VA
reply to rconaway8

said by rconaway8:

Justin (WHT) and I have spent months analyzing a business plan with a potential of 3000+ users to try and reduce the Capex. The best we have done is about $250K to cover 500 square miles.

That is absolute horse hockey...

A 200M business plan? Jebus..

I can cover 500 square miles for far less than $250,000.00 investment but *A START UP DOES NOT NEED TO COVER 500 SQ. MI.*
--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."


DaDawgs
Premium
join:2010-08-02
Deltaville, VA

1 edit
reply to Telecomnet

For Gods sake, please give me a 10M start up and I will run every competitor for fifty miles out of business....

That number is just insane. There are three WISPs in our area. There are perhaps 35 thousand households lit up by those three WISPs. Now the area they cover is probably close to 2000 square miles... Just so we keep things in the same units.

Total invenstment in the three companies is maybe 8 million. They service in the aggregate about 6 thousand subs at about $30.00 per month average.

These are not start ups, these are established companies.

And you know, I'm just dumb enough to put up my own WISP in the middle of all of it because I think I can take enough share to make it work.

Don't let some know-it-all tell you it can't be done for less than 10M.
--
Once we IPv6 enable every device on the Internet we will have toasters, baby monitors, and security cameras joining the bot nets which today are populated only by idiots that can not refrain from clicking, "Yes I would like to see those titties..."