Once upon a VOIP

Once upon a VOIP

It's a truism. Technology change is the easy part. Changing people is the hard part. Or, as Housing New Zealand's Paul Duxfield puts it as he talks about the corporation's move to voice over IP: "Probably the biggest challenges we had weren't the technology but the organizational side around it."

It's such a common theme that you want to hammer it home: technology can be mastered; people need to be managed. Let's say it again: You can make all the changes you want but if you can't bring your clients with you, forget it. Managing change, at desktop level or in front of the board, is at the heart of any CIO's job.

Duxfield, network and user support manager at the corporation, has travelled a long way down the change management road in his role as one of Housing's key figures in its drive towards VOIP. It's a journey that took about three years of preparation as the key participants faced up to the challenges and pitfalls and formulated their business case.

After all, why change something that is working just fine as it is? Why oust the old technologies, the phones, the PABXs and the Centrex-based systems, at great expense when -- for the most part -- they are still functioning just fine? The board wouldn't really care much about the fact that there was a new paradigm in telecommunications awaiting fulfillment. What it wanted was a business case. And the users would need to be convinced.

"Before you embark on VOIP there are several questions you should be asking," says Duxfield. "We knew what VOIP meant but we didn't quite understand it. It wasn't quite what we thought it was. The big question was: Will VOIP replace a functioning PABX? When you take your business case to the executive, this is one question you will have to answer. What's the bottom line in terms of budget and service delivery? Can I migrate to VOIP or do I have to do a major upgrade?"

Well, no path is ever simple or straightforward. During a year of proofing the concept, Duxfield came to the inescapable conclusion that VOIP would mean -- in his own words -- "a major, major upgrade". Some of the network technology was already several years old. It worked for current needs but meeting the requirements of VOIP would require a big project.

Duxfield likens VOIP to a shift from electronic typewriter to a PC. "Neither a PC nor Windows can replace an electronic typewriter, and Notepad does not make a very good word processor. Also, you have to have some way to print."

Add Windows to your PC and attach a word processor and things start to look different. Now the PC can do word processing (with spelling and grammar checker) and much, much more. The user has an infrastructure that can be used for all sorts of applications. Now you can add a spreadsheet or a database. The PC doesn't replace the typewriter -- it improves on it and supersedes it. "VOIP is not voice, or IP, or phones connected to a PABX via IP. It's a different paradigm."

More to the point, it's a paradigm where VOIP phones and IP voice gateways running on a converged voice and data WAN, and integrated with other office productivity tools, can improve on the PABX.

Before the upgrade, Housing had a traditional telephony setup. Fifty PABXs were scattered around the country. There were also numerous Centrex systems. A separate network supplied data. As many as 700 staff were fortunate enough to have cell phones. Each system was separate, which meant that ownership and management lay with different departments. The traditional voice services were under the control of the finance group. Data fell within the domain of IT. No one really understood or managed mobile. As Duxfield says, "It was all a bit of a hodgepodge."

Nor was there any national dial plan. A contact center was implemented in 2003 with VOIP. It simply happened to be the best solution. Clearly, the overall structure needed some sort of logic and control. New technologies such as VOIP would not be the answer in themselves. They could even make things more complicated.

For that reason the strategy that was eventually defined within Housing does not mention VOIP. Nor did Housing go to market for VOIP. What was documented was a list of what the organization wanted to achieve and the services it sought to deliver. "This is what the market came back to us with," says Duxfield. "And it worked."

Surprisingly, one vexing question within the organization was voicemail. It did not use it - or, at least, most people thought they didn't use it. It certainly wasn't available on the landlines. Mobiles, however, told a different story. Woe to anyone who suggested voicemail shouldn't be used on mobiles.

"Just imagine going to the executives and saying, 'I am going to implement this, this and this, including voicemail. The executives: 'Why are you going to implement voicemail?' Your answer: 'Because you already have it.' The response: 'No we don't.' Your answer: 'Well, you have Cellular Secretary.' 'Yes.' 'So you've got voicemail.' The response: 'Oh, that doesn't count.' So you say: 'Can I turn it off?' The executives: 'No. I need that.'"

Duxfield was surprised, to say the least. "There was a perception we didn't have voicemail." The reality was that those who didn't have mobiles didn't have voicemail.

Returning to strategy issues, Duxfield says Housing wanted to create a voice infrastructure that mirrored the data network. Staff wanted to be able to add a new phone, PC or printer as the need arose. What they didn't want was to have to go out and contract a PABX provider to come in, configure a card, get a phone and have to wire it into the system. Basically, the support team wanted to be able to plug in new phones or whatever as the need arose. As already stated, the vision statement made no mention of technology or VOIP. It simply said Housing required "a seamless network and related services as a platform to meet all voice and data requirements".

As long as the different solutions lay within different silos or empires within Housing, change could not happen. "Finance controlled the phones. IT had its little bit. And everybody wanted mobiles. Then IT came along and said it was going to do the whole lot. We were trying to look at it as telecommunications."

Meanwhile, Housing itself had been going undergoing change at the hands of its political masters. After being split apart, its structure was once again being unified. At that early stage it wasn't a good time to add even more change, and so VOIP went on the backburner for a while until things settled. "We were doing things differently," says Duxfield. "We needed to be more flexible and we needed to be able to place the workers out in the field a lot more. At the same time, demands were increasing. We needed a way to do things quickly and efficiently. We wanted to be able to provide the business with a flexible infrastructure that would support it into the future and be reasonably cost-effective now."

The starting point for real change lay within the control structure. Contracts needed to become available for renewal at the same time so that unified services could be established. This would be a two-year process. At the same time the telecommunications vision and strategy needed buy-in from all stakeholders. Then, if the vision was to be achieved, it needed confirmation from the telecommunications and networking industry that it could indeed be done. That part of the process required a strategy on how to approach the market. Duxfield had a solid workload ahead.

Handling of the various contracts was done in a simple enough matter. As each one came up for renewal, a single ending date was established. By the end of 2000, that task was done. Interestingly, earlier attempts to talk to Telecom's competitors about the separate services had not been received with a great deal of enthusiasm. Yes, they said, we can go for it but Telecom will be able to undercut us.

Internal buy-in at Housing was based entirely on a business benefits approach. The stakeholders -- finance, IT and other interested parties -- were involved in creating the vision and strategy. The fact that the focus was on the business aspects made the job a whole lot easier and eased the migration path for IT to take responsibility.

Nagging questions remained: Will it work? How do we know we can deliver? To answer those questions, Duxfield's team booked a room and invited Telecom, Telstra, Vodafone, a PABX supplier and a data network supplier. "We presented our strategy and asked them: 'Can we achieve it?'" Each player was given five minutes to present their spiel. The conclusion: What housing wanted "is here, it is now, and we can do it".

The debate continued. "Can we do it in today's environment?" Housing asked. "Is it going to be easy?" The feedback was good. "Yes, we can do it. Here is what we can do. Here is how we have done it. Here are our strengths."

Based on the vendor feedback, Housing formulated a strategy that was somewhat different from what many might have expected. Housing decided a consortium would best answer its needs. "Some of the vendors weren't that happy. Some of them threw the toys out of the cot. That's life. We wanted an even playing field. We didn't want to go to market with something only Telecom or TelstraClear could provide."

The consortium lead partner would provide solution integration and project management services. The consortium would provide fixed and mobile data and voice services. It would also provide data and voice equipment. In all of this there was no mention of a specific technology such as VOIP. Requirements could be fulfilled in a number of ways.

"It wasn't just about cherry-picking, although we did want to pick the strengths from here and here and here... We just didn't find it possible to get the best deal by placing all the business with one player. Mixing and matching gave us a better solution and we could spread the total cost of ownership return over three years. We felt this was the best way to fulfill our new structure."

So why not just outsource? No, says Duxfield. Housing wanted to maintain healthy competition between service providers and equipment providers. And, once the decisions were made, it wanted to continue that competitive approach. It wanted to change how telecommunications were implemented, not just try for a better discount on services.

Why not do the migration slowly? Once again Duxfield says no. "We were incurring all the costs upfront and we wouldn't be getting any cost-savings until the job was finished. A slow migration didn't stack up and staying where we were wasn't an option."

As for outsourcing direct to Telecom, others had done it successfully. However, such a move would not answer one key requirement -- healthy competition between service and equipment providers.

Duxfield uses Centrex to highlight the issue. "You go to Telecom and they say Centrex is the deal for you if you have anything up to 25 phones. You go to a PABX provider and they say you need a PABX if you have more than five phones. The reality is that the real answer is something in between, depending on your business. You need to be able to get both views and you need to be able to pick and choose a bit."

Behind all of this remained the more practical objectives. Housing wanted to reduce its operating costs. A group discount might hook back 1 percent to 5 percent but by changing the way things were done a reduction of 10 percent, 20 percent, maybe even 30 percent might be possible. Now, that made sense.

The actual targeted objectives for savings were more mundane: reduce operating costs to the business. Housing also wanted to replace failing equipment such as cell phones, commodity items that were unlikely to last beyond two years. Then there was the obsolete equipment, reliable as it might be. Some of it, including the oldest Cisco routers, were as many as eight years old. They had provided a dream run but they would need replacing too.

A new structure could provide an option for more efficient -- and less costly -- payment for telecommunications. Why not do it electronically? That made sense -- and a reduced paper chase would reduce transaction costs.

Putting all the new technology aside, Housing wanted staff to feel they were working the same as they did before the changes. What went on in the bowels of the machines might be another matter. If a phone rang, staff should be able to pick it up and respond as they did before.

All 55 offices said yes to the proposals. That number might not sound like all that many but the offices housed 1251 phones, 1200 PCs and 700 mobiles. Most of the offices, except for the ones in Auckland and Wellington, were small. Most had fewer than 10 staff; many housed no more than three people. A PABX was overkill for an office staff of three, while a Centrex probably didn't the integration that was required.

So what was the best solution? You guess it: the market came back and said VOIP was the answer. But first it had to show it could deliver a good quality of service. That certainly should be possible over a single data pipe. After all, Telecom had been done something similar for years with its own services. "They split it out for our benefit and then charge us for the privilege."

The question of who could vouch for quality of service was vexing. "There are two philosophies here," says Duxfield. "People coming from the voice side say data people don't have a clue. People coming from the data side say voice people don't have a clue."

Housing decided to go for a Cisco solution -- from the data networking side, of course. While Cisco might not know everything about voice, at least it knew how to build a solid network.

Under the old system, Housing shared between 600-700 trunk lines among its 1200 users. Because the organization was so fragmented, economies of scale were hard to achieve. Now it is running around 180 trunk lines, considerably cheaper than those in offices with analogue lines. The VOIP phones are fed via gateways in Auckland, Wellington and Christchurch. Voicemail is a reality, despite the initial opposition and debate.

Mobile phones have been partly integrated into the new systems. It's not entirely what Duxfield wants but, he says, mobile carriers won't clip the ticket at every point. "The protocols are there, the standards are there, but for some reason they don't want to do it. We have had to work around that in various ways."

Housing's experience with carrier IP services are not as extensive as Duxfield would like them to be. They have been tried and tested but are seen as not being ready for mainstream use. The last thing Duxfield needed was uncertainty at a time when staff numbers were rising rapidly.

Despite his best attempts to the contrary, the changes nevertheless sprang a few surprises Duxfield's way. It was interesting, for example, to find out how phones were really used. The business teams had said what they needed; the reality was somewhat different. "This was something we should have put more effort into upfront," says Duxfield. "But it always came back to the policy thing, something we didn't want to get too involved with. And some things were different with VOIP. It's not a PABX."

There was also the fact that Cisco's core business was data. Simple telephony concepts such as hunt groups were foreign to the data people. To telephony people, a hunt group is a supplementary service in which a group of fixed or mobile telephones are defined. An incoming call will be passed from phone to phone until the call is accepted, so the definition goes. Duxfield says the next version of hunt groups will feature be closer to the telephony concept.

"There are a few little gotchas. Some of the services intrinsic to a PABX have to be constructed using components with VOIP solutions. It's an infrastructure thing. You have got to add your applications to make things like this work."

Another issue to be aware of is the need for training in VOIP phone usage. Duxfield found two people were needed onsite for training when an IP phone system went live. "We thought it was just a phone, right? Everyone knows how to make a phone call."

Then were things such as the menu systems. For Duxfield, a technology person, they were as easy as pressing a button. Users also have to log in each day. "We wanted the phones to be theirs for the day." Just log in.

Duxfield warns against using upgrades to enforce policy changes. If your staff have bad habits, policies need to be handled properly, otherwise conflict can center on the product itself. More to the point, new policies need to be understood before the technology is put in place. You can't simply hold up a project for six months while the opposing sides try to sort out their differences.

Other challenges on Duxfield's list include car kits. "Should we or shouldn't we?" The initial reasoning at Housing was that those who already had car kits should continue to be able to use them. Then someone introduced a paper pointing out that talking on phones in your car can be dangerous. Charges could be laid against corporates and individuals under two sections of the Land Transport Act.

Use of car phones could come under occupational health and safety in the workplace. "If someone has an accident when we have told them they must answer the phone, we could be charged. So we had a bit of a debate there ..."

The voicemail question is another one that needs to be worked out. Housing's philosophy was that when a customer called they should be able to talk directly to someone. Voicemail did not answer this policy. Anyone considering the issues needs to ask: "Is this for the benefit of staff or customers?"

Mobile phones raise particular issues. "My impression is that they are targeted at teenage girls," says Duxfield. "Fifty-plus males with fingers like sausages can't push the buttons. They can't read the screens and they can't hear the ring tones. It's hard to get a cell phone that meets their requirements. And phones are getting smaller. What I found is that users' main criteria are for a color screen and silver casing. Battery life, signal strength and voice quality don't even come into it."

Then there is the matter of customer service policy. It needs to be looked at as the changes are planned. Duxfield observes that many managers were reluctant to change. They had to, though, in order to meet the changes implemented under a rationalized telecommunications policy. An unfortunate side effect of this reluctance is that the change management leaders, including Duxfield, tend to become the focus of frustration. As Duxfield says, "You have to learn to roll with the punches."

Electronic billing, related as it is to external suppliers, raises special challenges. Duxfield says his team saw it as a great idea, one that could save everyone money. Unfortunately the carriers didn't quite see it that way. They wanted their pieces of paper, and they didn't want to change their procedures. Duxfield wanted to make it work but it has proven to be too hard to do.

"Telcos aren't great at billing," he says. "You have got to watch those invoices like a hawk. You've got to understand them. The people who do the billing have no idea how the services are built. And no one has actually read the contract you have signed ... Having someone who understands the bill can save you thousands."

If you are left wondering whether all the effort has been worth it, Duxfield's bottom line will leave you with few doubts. First up, he tells you that between February 2004 and February 2005 staff increased by 18 percent. At the same time cell phone numbers increased by 28 percent. Cell phone usage increased by 35 percent. Now comes the punchline: in that period actual telecommunications costs have decreased by 6 percent.

So, forget about technology. Forget about VOIP. What really matters here is the business. When you make your case, focus on the business. The outcome might well be that you will go to VOIP. But, hey, what really matters in the end is the bottom line - assuming, of course, that everything works as it should.

The road might be arduous at times. Do the job properly and you will certainly find it well worthwhile.

Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!

Error: Please check your email address.
Show Comments