top of page
Leon Woo's Take
Pink Sea

Telecom Business at 3Com: Why Did it Fail?


Irfan Ali provides answers:


“Given the complexity of the 3Com story, I understand that no frame might be large enough to include all the elements.  Our telecom business is one such element.  Based, initially, on the USR Total Control - Remote Access Server (RAS) platform, this business rapidly grew to approximately $800 million in annual revenue, with gross margin in excess of 60%.  At its peak, this business had 1,800 employees globally.


The success of this business was based on the emerging intersection of market timing and technology.  Given the imminent, wholesale shift to IP based networking in private and public networks, the telecom operators were scrambling to find a carrier-grade solution that could provide a long-term transition for the PSTN (Public Switched Telephone Network).  In this environment, we were able to position the Total Control as a Media Gateway, with an associated SoftSwitch (software for call control, etc), that could provide voice, data, fax, video services over a common IP based platform, for both wireline and wireless networks.


In this process, at one time, we had the top 20 telecom operators as customers, using not only our Media Gateway + Softswitch but also implementing our larger network architecture that included dynamic service creation capability as the third layer.  This architecture effectively separated services/service creation from the underlying network and was a precursor to the smartphone based apps that are prevalent today.


The business model that we used effectively in this case was based on customer-intimacy.  Eric refers to this model in the context of the large enterprise market.  In the telecom market we used this model to work closely with some of the largest operators in the world and, using a standard platform, deliver customized solutions.  In the process, we made ourselves "indispensable" to the customer and created an entry barrier against Cisco.


As we built this business, I worked closely with Eric and Bruce.  As the business grew, Eric demonstrated his strong support for it with full commitment to the challenging task of building the next generation of Media Gateway + Softswitch.  In this context, we took a close look at acquiring Sonus Networks, the VoIP startup founded by Hassan Ahmad and Rubin Gruber.  In the end, I agreed with Chris Paisley that the valuation was too high and we passed.  Sonus subsequently went on to a successful IPO.  Around the same time, Cisco acquired TransMedia and Tellabs acquired Salix, both in the same space.

"Sonus IPO doubles in debut"

"Tellabs buys Salix Technologies for $300 million in stock"

"Cisco buys Transmedia"

Once the decision was made not to acquire Sonus, the company, nonetheless, demonstrated its commitment to the telecom space by approving what had to be one of the largest development projects in its history.  And, Total Control 2000 was born, with an internal code name of Millenium.  The Millenium project included the next generation Media Gateway, intended to be the most scalable product in its class, and a next generation SoftSwitch supporting H.323, MGCP, SIP based call control.  It also included a SIP based service creation capability that would allow the telecom operators to dynamically monetize their singular, new, IP based networking asset for voice, data, fax, video; wireline or wireless.  The new network architecture was called CommWorks.


As this development progressed, for a variety of reasons, we decided to spin out this telecom business through an IPO, using a model that was subsequently validated by Palm; we were about six months behind Palm.  The resultant company was to be called CommWorks.  And, to strengthen the CommWorks portfolio further, we acquired two companies in the IP services space.  The first one was a Canadian based Fax-Over-IP company that we acquired for approximately $20 M.  The company was called LANSource.  The second was a Virginia based company called Call Technologies with a Unified Messaging platform; that acquisition was for approximately $90 M.

Company Overview of LANSource Technologies Inc.

3Com Announces Plans to Buy Call Technologies

And then, it fell apart.  For larger, macro reasons beyond our control.  It turns out that MCI, under Bernie Ebbers, was cooking its books.  And MCI was one of our largest customers. When that was revealed, the legitimate concurrent concern about a slowdown in the telecom market turned into an unmitigated panic, a stampede for the exists, a good old bank run.  Call it what you want, but the telecom market collapsed.  And overnight, I had a business that went from a large cash machine to a large cash drain.  With an immediate pressure to plug the drain, I started the painful process of laying off approximately 1,000 employees.  In this process, I had invaluable assistance and support from Steve Campbell, our CFO, and Alice Bergmann, our VP of HR.  And, while we did this as carefully as possible, there was no denying the carnage that we left in our wake as an iconic mid-western technology success story was dismembered in full public view.   


And, yet, there were factors within our control as well.  The Millenium project was taking a lot longer and costing much more than originally anticipated.  While this is not uncommon for large technology projects, there is no excuse for the fact that we did not anticipate the complexity associated with combining DSP technology with packet engines at a scale that we wanted.  Or the complexity associated with developing a high performance call control engine based on emerging protocols.  I take full responsibility for that.  And, I learned painfully, and first hand, the fine line between confidence and arrogance.  The USR ethos of never saying no, or pausing for legitimate doubt, had served the company well.  And, an environment that did not tolerate bad news had taught the mostly intact engineering team the wrong lesson.  This lesson would come back to haunt us at this crucial time.  I should've recognized this sooner.  But, I did not.  And, for a reason that Eric alludes to in the book.  In an environment as intense as this, the first thing to go is the space to think.    


Eric’s original reasoning behind the USR acquisition, in the context of the telecom market, was not wrong.  At a time when we were struggling against Cisco in the large enterprise market, a strong beachhead in the emerging multi-service telecom space based on USR technology could be a huge strategic advantage.  And, it could give us incumbency for a massive, emerging opportunity as significant portions of the PSTN transitioned to IP, and as wireless services took off.  But, good strategy is no match for bad timing or inadequate execution.  When MCI blew up, and dragged the entire telecom market down with it, all bets were off.  And, in the ensuing storm, our inability to deliver on the Millenium project took future promise and converted it into a large, current liability.  


For those of us who truly believed in the promise of the imminent telecom revolution, the downturn was a significant but temporary setback.  When I and my management team undertook the grisly task of laying off a thousand employees, we viewed it as a necessary step to save the organization.  And, as good corporate citizens, we did not want to be a burden on the rest of our 3Com family.  The downsizing, along with a number of other steps, including the outsourcing of a significant part of our legacy development to India, brought our business to cash neutral.  I truly believed that these drastic measures had bought us the time to weather the storm and continue with our quest to win a sizeable market share in the new, converged public network.  However, that was not the case and the severe pressure for additional downsizing continued.  Having lost the argument for more time, and therefore the ability to maintain critical mass for the organization, I left 3Com in September 2002.  In March 2003, CommWorks was sold to UT Starcom for $100 M; in essentially, a garage sale. 


Had the telecom market stayed intact for another year, I believe CommWorks would be a public company, with an initially significant valuation.  Even with the subsequent downturn in the telecom market, CommWorks would have just as much of a chance for success in the public market, as Sonus Networks and Starent Networks.  In the process, it could’ve provided significant financial return to 3Com.  In which case, perhaps the judgement on Eric regarding the USR acquisition would be slightly different.




bottom of page