OnSIP – The smart approach to business phone service

Interview by Mike Sullivan of Sully’s Blog

Junction Networks was founded in 2004 and delivers VoIP communications services using superior telephony technologies to over 5,000 small, medium, and large businesses. The company was founded on the premise that the future of business communications would rely heavily on IP technologies. OnSIP is a hosted business VoIP service. Customers connect phones to OnSIP over the Internet from any location for everything from extension dialing and inbound calls to acd queues and voicemail services. It’s the leading open SIP platform for business.

Mike Oeth is the CEO and lead of the Sales/Support team at Junction Networks.  He graduated from Harvard College in 1991 with a degree in Economics and went to work for a software company based in Greenwich, CT.  Since then, he has always worked for a technology company in one form or another – cable companies, fiber optic providers, Internet service providers, and application service providers.  His roles have ranged from technical support, sales, business development to executive management.

Rob Wolpov is President/CFO and lead of the Marketing Team at Junction Networks.  Rob holds his M.B.A from Columbia University. He received a B.S. from the Wharton School of Business at the University of Pennsylvania.  He began his career as a software engineer for the Frustum Group, a provider of leading foreign exchange trading software. He has planned and launched multiple technology ventures including Nemex, Inc. and ASP Junction, which was purchased by LivePerson, Inc.  Before forming Junction Networks, Rob worked as a consultant in the Financial Services practice for IBM Business Consulting Services, advising some of the largest financial institutions today.

John Riordan is CTO and lead of the Engineering Team at Junction Networks.  John first got involved with computers in the late 70’s – Apple II, Commodore PET, TSR-80. He studied computer science and cognitive science at MIT in the late 80’s. “While I spent a good deal of time trying to learn all I could about how brains work, I ended up getting sucked into doing work under Andrew Lo at Sloan.”  After school he spent a few years doing software engineering on Wall Street, jumping into the internet in 1994.  Since that time, he’s been an Internet entrepreneur and engineer. He has spent most of his career doing internet service and software engineering.

MO:
My simple explanation of OnSIP is that it’s a VoIP service for businesses.  But it’s much more than that.  Tell me about the service, what it includes and why this is different from the high cost competition.

Mike Oeth:
OnSIP is a business communication platform.  It is built, from the ground up, to be integrated with other services like Outlook, Salesforce, Highrise, etc.  It’s far from a stand-alone entity.  We want our customers to be able to integrate our system with all the other platforms they use.

Rob Wolpov:
It’s the glue that keeps a modern day, Internet savvy business together.  It’s grounded in Internet concepts and foundations: Pay for what you use, when you want it, only when you need it.  It’s available anywhere you want, as long as you are connected to the Internet.

For a business of 30 people in multiple physical locations, this may be the first time that a “corporate-style” communications platform is available at digestible prices.  Previously, small and medium businesses were shut out of sophisticated phone systems because of the high capital cost, extensive deployment requirements and ongoing support and maintenance.

Now, a business of any size can have a corporate presence that used to cost $100,000 for no upfront cost or ongoing support commitments.  A team can be distributed all over the world, but stay connected to the corporate phone network at all times.  And, it’s as easy to set up and maintain as a web hosting or email hosting account.

John Riordan:
It is so much more than that. It is higher quality and more reliable. It costs less. It is smarter and faster. It is overhead reducing. It is performance enhancing. It is more attractive and more interactive. It kicks ass.

MO:
Where did the idea of building OnSIP based on open technologies originate?  Does this give you the freedom to make improvements and enhancements to the system more easily?

Mike Oeth:
This is really John’s genius.  John Riordan, our CTO, built an Internet Service Provider business using open source:  In the late 1990’s, John manipulated Apache, open source web server software, and combined it with other open source platforms.  Through his success, we learned the benefits of open source software for service providers.

As a service provider, you are not beholden to some third party for upgrades and bug fixes.  If you need to either enhance something or fix something, as long as you have the developer skills in house, you have that flexibility and power.  Open source does, however, force you to become a software company as opposed to just a service provider.  From the start, you have to know that differentiation.  If you choose the open source route, you’ll need to continue to invest in and grow your development team along with the business.  To work as an open source company, you have to know that at your heart, you are a software company.

Secondly, open source fits well with our company ethos.  We do not lock our customers into any particular hardware or even in long-term contracts.  We want our customers staying with us because of our merits, not because we have them locked down.  It also allows us to influence the way we charge for our product.  Since we do not pay a third party a per-seat licenses like most providers, we can then have a completely different pricing model.  Our customers can have as many phones and users and extensions as they need.

Rob Wolpov:
Mike is a genius.  One day, he stormed into a meeting between John, myself and some Broadsoft reps. He told the Broadsoft reps to leave and then demanded we come up with a lower cost alternative.  Enter open source!

Just kidding.  It was a no brainer from day one.  We intended to build a platform that was not tied to any one vendor, application, system etc.  The important thing to consider is not that the components are built on open source; rather, it’s that the entire platform is modularized such that ANY component can be swapped out for something else.  We did this to take advantage of the best of bread solutions, regardless of vendor, open source availability, etc. It just so happens, that open source technologies have many advantages.  Using them frees us from vendor licenses. As a result, we can pass on savings to our customers. Also, as you asked, we are not tied to any vendor road map or product decision. As a result, we can respond very quickly to customer requests for new features.

John Riordan:
Hard to say. Al Gore maybe – didn’t he invent open technologies? [Editor’s note: Please see Mike’s response for a more helpful answer :) ]

MO:
How portable is this system?  Do users need to be in the office to get calls or are there other options that allow them to be more mobile or work virtually for their company while still leveraging the benefits of this system?

Mike Oeth:
OnSIP is completely portable.  The same user can be registered on up to 10 different devices.  That means you can have a phone in the office, a phone in the home office, a soft phone on the laptop, and an app on the iPhone – all of which will ring at the same time (if you choose) when someone calls your work phone number.  Most days, I work from the home office.  It’s great because I have 100% of the functionality I’d have in the office, but have the flexibility to not have to commute into NYC every day.  So, it’s portable, and in some respects a green solution.

This goes back to our open source roots.  Since we built the software ourselves, we do not have to play by someone else’s rules and restrict our users and how they use the system.  We want our users to be able to use are system as flexibly as they use e-mail today.  Today, you can get your Gmail on any Internet-connected PC and on your Blackberry or iPhone.  Your company phone service should be this flexible; and, with OnSIP, it is.

Rob Wolpov:
There is no requirement to be in any physical location.  Just like getting your email anywhere, on any computer tied to the Internet, OnSIP works anywhere.  You can have 100 employees in 100 different locations, all making and receiving 4-digit extension calls, taking calls from the corporate auto-attendant, transfering calls, and even making outbound calls from the company “phone system”.

John Riordan:
I’ll defer to Mike on that one.

MO:
Are there any advancements or predictions you have for the future of VoIP technologies or telecommunications in general?  What will the world look like 10 years from now as it relates to the need for communication with other individuals or businesses and the technologies available to do that?

Mike Oeth:
IPv6 will allow for seamless phone call transfers from cell, to WiFi and back.  With that, I don’t think that phone service will be recognizable 10 years from now.  It will be phone, video, SMS, chat, PBX extension, and desktop sharing all rolled into one.  Today, my iPhone can be used as an extension on our PBX.  This means if someone calls my extension while I’m out of the office, and have a softphone application running on my iPhone, I’ll receive that call.  That’s just the beginning of what will be available in 10 years.

I can’t wait to read what John wrote for this.  I predict John will write about he ubiquity of SIP addresses and the combination of services with presence (availability) information.  My take on that from listening to John talk about it is that SIP addresses will merge with e-mail addresses, SMS and chat and the device will decide whether you’re trying to make a SIP phone call or IM or e-mail depending on your location and current task and the availability of the other party.

Rob Wolpov:
The demise of the 10-digit numbering scheme will continue.  People will call one another using addresses, just like they do with email.  We’ll look back and say “remember when rob at ACME Corp had that funny number we had to remember all the time in order to talk to him in real time?”

John Riordan:
I think it is a safe bet that the inflation adjusted cost of bandwidth will decline. Anyone want to make me a market on that one?

Also, I predict telephone numbers will decline, and people will use SIP addresses.  The addressing scheme in SIP is one of the most compelling aspects of the protocol. Like email addressing, SIP addresses can be associated with an individual and routing is controlled by DNS. For example, Joe User’s email address is joe.user@acme.com. Likewise, Joe User’s SIP address can be joe.user@acme.com. In the same manner as they would send an email to Joe, anyone with internet access can make a call to Joe’s SIP address. Joe can talk all day and receive as many calls as his internet access will support. And Acme can add additional SIP addresses at will and control where call to those SIP addresses go by updating their DNS.

By moving from telephone number based addressing to domain based SIP addressing, the telephone company is eliminated. Calls occur peer-to-peer, and whoever owns the domain controls all the SIP addresses in that domain. There are a some serious caveats in all that; but, it is fundamentally the case. It is simply more efficient to communicate peer-to-peer over the Internet. Assuming SIP remains the de facto standard, it seems a reasonable to speculate that over time that domain based SIP addressing will take hold. Free World Dialup initially demonstrated the concept years ago. But, there is a critical mass aspect to the whole thing. Some day, a tipping point may be reached; and, if it is, there will be accelerated adoption. Perhaps Google may trigger it by making all the gmail.com addresses SIP addressable.

MO:
What were the factors that brought these three great minds together?  How did you meet one another and come to the conclusion that Junction Networks would be the venture you launched?  Would the outcome have been different if any one of you were not in the equation?

Mike Oeth:
In late 2003, the company I was working for went through Chapter 11.  I was let go during that process and was hiring myself out as a technology consultant and a home handyman.  Yes, these are nearly diametrically opposed occupations.  But, luckily I’m nearly as good with a hammer as I am with a keyboard.  When I found an opportunity to install a VoIP system for one of my technology consulting engagements, I started researching the field.  After a little research, I brought Rob on board.  After Rob and I had made some headway, we brought John into the loop, and he quickly got up to speed on Asterisk and saw the similarities between Asterisk in 2004 and Apache in 1996.  From that point on, we were hooked.

Each of us brings something unique to the table.  That helps us stay out of each other’s hair; but, without one of us, the landscape would be radically different.  I fully believe in crowdsourcing.  We don’t always agree; but, when we disagree, we work through it and always come up with a solution that was better than anything any one of us could have come to on our own.  Ideas don’t get watered down in the compromise – they are built up and strengthened.

Rob Wolpov:
We all had our own personal motivations for being here although we certainly shared a common desire to build a successful technology business.  We were all interested in leveraging the Internet to deliver a service that made sense to small and medium businesses. All three of us lived through the dot-com birth and despite the bubble’s bursting, we all shared the common belief that the Internet was just getting started as a technology platform.

Back in 2004, there was some buzz about Vonage, Skype and some other internet telephony.  At the time, we had some concern about the Internet as a real time communications medium. But, those fears were quickly put to rest when we got our business started.

As we each bring unique experiences to the team, the combination results in consistently rational thinking.  Not much happens around here that doesn’t make real business sense.  We’ve all seen the fly-by-night venture funded companies who are short on products and revenues, but long on buzz and PR.  We tend to stick to what works, what our customers want, and what makes good business sense for us and our customers.  And, we have an eye on what we think the future will bring.

Oh, and if I weren’t a part of the team, Junction Networks would be doomed.

John Riordan:  It was all Mike’s idea. But he needed some lackeys to deal with the grunt work of operating an internet service so he could focus his great mind on bringing the technology into existence… Rob and I looked good on paper. He generously let us onto his coat tails.

MO:
Launching a VoIP service sounds like a colossal achievement.  What were some of the challenges you faced getting off the ground?

Mike Oeth:
I talk about a post-it note I had on my PC for the first year it read “Origination = in, Termination = out.”  Those are some very, very, very basic telephony terms.  It’s analogous to a note reading PC = personal computer.  But, as someone who has grown up in the digital age, learning old-school telephony was like learning a foreign language.

The great thing for us is that we do not carry any of the old-school telephony baggage with us and are able to break a lot of rules – all for the better.  For example, we don’t have an “uptime guarantee.”  That seems to be a staple of old-school telephony.  That also means, that if you ever admit you have downtime, you have to pay your customers, so the motivation is to hide and disavow any outages to the system.  We do just the opposite: we tweet and blog our outages (which are rare).  We don’t want our customers banging their heads against the wall trying to figure out what’s wrong.  I don’t know of any carrier who does this.

Rob Wolpov:
We had a plan from day one.  BUT, we had many people calling us to ask for custom services, one-offs, hybrid solutions.  It was tough to say no – but we did it often.

Convincing businesses to abandon behemoth companies and well established service providers for our upstart company was somewhat tough.  Convincing them that we would be reliable was no easy task.  But, we stuck to what we knew, did it really well, never over-promised, and always delivered.

John Riordan:
It has been a colossal undertaking. The biggest challenge was trying to get a handle on what the hell we were doing and why.  While we all shared the same view that VoIP-land, in general, was the place to go build, we didn’t have the same vision regarding the details of the landscape.  Ultimately, it took us more than two years and a few course changes to get our OnSIP service off the ground.

We launched a peering service in Aug 2004, and it failed badly.  (https://forums.dedicatednow.com/index.php?topic=2743.0). So, we decided to try something else. As we tossed ideas around, we realized many of them required being able to gateway calls between the Internet and the PSTN. Upon investigating the VoIP trunking market, it became clear that the process of obtaining VoIP trunking service was a really painful for a small company like ours. We realized we could do better.  However, we also predicted that selling trunking services was a crappy business long term. Only a telephone companies can make a good business out of selling trunking, and we had no interest in becoming a phone company. And longer term, we predicted free calling over the internet would drive the cost of phone calls toward zero. So, the trunking service would have to be just a stepping-stone to another service.

We spent the first 4 months of 2005 building the SIP/IAX trunking service. The platform was a single Asterisk instance sitting in a colocation facility that was doing nothing more than acting as a B2BUA between upstream trunking providers and our customers.  Setting that part up was easy. All the hard work went into building a web based portal for customer self-service and real-time prepaid billing; that was the advancement we really brought to the market. We basically applied Internet dialup service business practices to selling trunking and released the service in May of 2005 by listing ourselves on voip-info.org. We had customers instantly.

The next few months were taken up with scrambling around trying to deal with having customers. The scale of sophistication of toll fraud was something we had underestimated and we were taken for a lot of money a few times. Meanwhile, I was busy developing a system that would handle the load as our single Asterisk instance was very quickly run over. I kept things going with an escalating series of really ugly and unsustainable hacks (stuff that I lost sleep over and gave me weird computer nightmares) while I worked on developing the “real” platform. We hired our first employee in Oct of 2005, purchased a bunch of Linux servers from Dell in Nov 2005, and rolled out a scalable “real” platform based on SER, Asterisk, and MySQL in December 2005. That system has been running more or less unchanged since than – just scaled up with more hardware and more network.

Rob and I started the development of OnSIP in Feb of 2006 while Mike focused on operating the trunking service. At this point, we had some consensus about what we wanted to build from a technology standpoint, why we wanted to build it from a business standpoint, and the likelihood that we could build it in a reasonable time frame with cash from the trunking service financing development. I built the system architecture and developed the software. Rob focused on interface functionality, testing, and marketing. We hired software engineers to join the project in June 2006 and spent the summer and fall finishing development. OnSIP was launched in December 2006.

MO:
You all seem to have previous entrepreneurial experience.  What were some of the lessons you carried forward from your previous ventures that helped to make Junction Networks and OnSIP a success?

Mike Oeth:
There have been many lessons.  Most of the lessons have been what NOT to do…

I’ve seen too many technology CEO’s who have absolutely no idea how the product actually works.  I think that’s imperative to know the in’s and out’s of your product.  That doesn’t mean that you have to be in there writing code, but you have to know how all the pieces work and be able to have an in-depth conversation with the developers when necessary.  You should be able to (hopefully) keep marketing, sales, finance and technology all on the same page.  I’ve also seen too many CEO’s who dictate direction.  Instead, you should hire smart people and actually use their input.

For us, without unlimited resources, it’s important to not just find customers, but to find the right customers.  There is a HUGE difference between a sales lead and a good lead.  We use Google ad words and native search results very strategically.  We try to capture customers who are at least a little ways up the learning curve.  If a potential customer is searching for ‘Free VoIP,’ they’ll never find us.  However, it they search on ‘Hosted PBX’ or almost anything SIP related, you can’t avoid us.  Rob has done a brilliant job in engineering our use of keywords and getting the right customers to Junction Networks; but, just that is nearly a full time job and requires daily diligence.

Rob Wolpov:
Raising capital with a business plan will distract you from building your company.

Learn by doing, not by building Powerpoint slides.  If you can build it and sell it, do it.  Later, you can always revise your plan based on real experience and real customer feedback.

There are many sharks out there.  Make sure you treat them has if they have sharp teeth.  Don’t stick your arm in an open mouth.  They aren’t smiling at you.

Good people make good things happen. Hire them.  Bad people drag down the good people.  Don’t hire them.

John Riordan:
Don’t panic.
Profits do matter.
The glass is half full.
Buzz doesn’t pay bills.
Customers do need to pay.
The customer isn’t always right.
Any product is better than no product.
Contracts are a good thing and worth reading.
People will tell you what they think you want to hear.
You will be the sucker, just try not to let it happen too often.

MO:
What would you tell an inexperienced entrepreneur who is contemplating leaving his day job to launch a new business?

Mike Oeth:
Now is a great time.  It’s easy to start an application in the cloud for under $30,000.  You don’t need millions in VC funding. And, in many cases, you’re better off without as you must make hard choices on where to spend that next dollar.

For me, and probably many people in this economy, I didn’t have a lot of choice in the timing.  I didn’t have a day job to quit – but I did know a good opportunity when I saw one.  That’s what you have to be able to do.

Just as importantly, you have to be able to change as the opportunity changes.  Junction Networks started out installing Asterisk PBX’s and then changed to SIP trunking.  Now, most of our effort is focused on our Hosted PBX product.  You have to be able to sense new business opportunities and throw out the old business plan to take advantage of those opportunities.

Rob Wolpov:
–          Make sure you are true to your personal stakeholders (spouse, family, etc.).  You will need more support from them than you think you will.
–          This is not a job.  It’s an adventure.  Seriously, there’s not going to be a steady paycheck for at least a while.  Don’t get into it unless you are ready for a roller coaster ride.
–          Expect the unexpected – good and bad.
–          Try to forecast how long it will take you to reach certain milestones to keep you on track.  Then, double the amount of time it will take for each milestone!
–          Read contracts.  The language is in there for a reason. If someone wants you to sign a legal document, keep in mind that the sales guy who has been promising the world may soon be gone and his/her boss will only have the contract you signed as a guideline to your business relationship.
–          Become as knowledgeable as you can before making any major decision/commitment.  For example, if you are opening a bagel shop, try to find a way to resell someone else’s bagels at special events or work in a bagel shop for a few weeks.  You’ll get to know the business before you spend your lifesavings building out a store with all of the necessary equipment.  You’ll understand the suppliers and customers before you make decisions you can’t back out of.

John Riordan:
Good luck.