A Winning Ticket

Should the president of Shell Oil Canada wish to spend a week working from a remote cabin in the wilderness, no matter how many thousands of kilometres from his Calgary head office, Canadian telco Telus Corporation wants to know all about it.

Not only that, it wants to be able to assure the executive the same level of service he gets in his office. It wants to know exactly how any service interruption impacts on the executive and his business. And it wants to know exactly how a disruption to his services would affect its own business, by associating a "lifetime value"with the customer.

It may sound a tall order, but the British Columbia-based company is getting close to being the first Canadian service provider to do just that, according to Leona LaChance, director shared services, Telus technology & operations. The key lies with an intelligent service assurance solution that is helping Telus proactively prevent outages. The solution helps the company to distinguish its offerings from competitors' by providing tangible proof to customers that contracted obligations have been met and letting it rebate them for data services that were not delivered.

Now LaChance wants to go further.

"We want to know that what might seem like a very minute service - such as a private connection to someone's cabin - is in fact the connection the president of Shell Oil in Canada uses so he can work while he is on vacation. We should know how that would profoundly affect our business,"she says.

Taking on the Big Boys

Perhaps more than any other industry, a service provider lives and dies by fault management, says La Chance. And when that service provider is competing with a telecom giant, that fault management had better be very effective indeed.

Telus connects organisations coast-to-coast across Canada and provides a full range of telecommunication products and services connecting Canadians to the world. The second-largest telco in Canada, as well as the largest mobile phone service provider in Canada, the organisation has almost 30,000 employees.

The company provides data, Internet protocol, voice and wireless services to Western, Central and Eastern Canada. It is an Internet service provider, an application service provider, and provider of Web hosting and DSL services to both its own customers and other ISPs.

Since deregulation, Telus as the incumbent carrier in Alberta and British Columbia in the West has been slogging it out for market share against Bell Canada in the East. And since it knows how unrealistic it is to attempt to compete against a giant like Bell on price, it is left to compete for market share on stickiness.

"Stickiness,"Davi Ghose, an analyst with CIBC World Markets told ITWorld Canada last year, "is Telus providing its business customers with a whole range of networking services rather than squaring off with Bell on price where it will not be able to compete."

And stickiness is just what the company is achieving, with a spectacular renewal rate for outsourcing contracts of around 90 per cent in an industry where customers commonly bounce from one provider to another, then back to in-house because of bad experiences with suppliers.

The secret lies in superior service. When Canada's provincial telephone companies decided to get into the IP and ATM data offerings, they all spun off "advanced communications"companies in each market. Reacting to the intense competition, Telus decided from inception to build its business around service management and offer distinct service level agreements to the marketplace. The company was determined to take advantage of the availability of new, smarter software. It is a determination that is ongoing and keen.

Page Break

Telus found success with Spectrum Event Manager, used by the company to automate service response procedures while guaranteeing individual service level agreements. LaChance says Telus is: delivering proactive notification to customers before they are impacted by a degradation of service; guaranteeing network availability of 99.7 per cent per site, per service, per month; and providing scalable infrastructure management that can support multi-vendor devices.

The system, purchased in 1996, continuously monitors the network for traps indicating a service interruption. Details of any problems are automatically forwarded using the Trouble Event Forwarder (TEF) to Telus' trouble-ticketing system. The trouble-ticketing system discovers what services have been interrupted by looking at the network element to service through correlation with an Oracle database. At three regional network operating centres, the system is continually communicating with more than 13,000 connected nodes nationally, approximately 6000 Cisco routers, 300-plus Catalyst 5500s and more than 1000 Sun servers.

In the highly-competitive telco environment, a service provider's success or failure depends on offering continued, reliable services to its customer base. And better still from its customers' point of view, the solution lets Telus define and monitor services per site, per service, per month, with no averaging.

"A trouble ticket is automatically generated and the operations person is paged,"LaChance says. "What's important to note is that all the information that the operator requires is readily available in the ticket. Therefore Â'OOIR' or Â'Ooh, Ooh, It's Red' is not practised at Telus,"she says.

"All information automatically generated into the ticket is real-time from the central database, providing customer information, affected device and description of the problem,"she says.

LaChance says Telus beats all competitors on "speed to notify", giving it a major competitive edge. "If we didn't have this we would be fighting on price points,"she says. "This lets us avoid that for the most part. Customers will still try to get a good price, but this gives a lot of value add."

Telus can also grow the network and increase the network and service base without adding much in the way of headcount for operational support. "In an environment where people are staring at screens, if you put a lot more devices on the network, you would have to put in a lot more people to look at the screens, to keep an eye on things. But because all of this is automatic we have not had to put on the headcount to look after the network as it grows."

And the solution is allowing Telus to really put some "skin in the game", LaChance says. "We guarantee 99.7 per cent availability, per site, per service, per month, or we start rebating the bill, starting at 20 per cent. We will rebate as much as 80 per cent of the bill for that month - and we don't average.

"In other words, if a customer buys services and they have 18 sites, and one site goes down for 18 hours, we don't average that out and say: Â'Well, gee, that means it was only one hour if you average it out. We don't owe you any money.' If that site was down for 18 hours we will not be charging any money. The actual service for the entire month will be discounted at the 80 per cent mark."

That prospect of having to rebate money to its customers gives Telus a pretty powerful incentive to get the SLAs right, LaChance says, and also helps it better manage its internal business. "For instance, if the reason why we keep losing connectivity is we need to desperately upgrade a piece of equipment, from a financial perspective it's good leverage to say: 'We need to make sure we have the appropriate infrastructure in place. Can we please get the correct capital funds to look after that area?',"she says.

Page Break

LaChance's message is twofold. First, that delivering on SLAs is not a pipe dream - it can be done. And second, that it is no longer good enough in the modern environment for any service provider merely to do network management: monitoring the network and reacting when it hiccups. Indeed, in the future Telus wants to monitor the precise impact on customers of any service failure.

"I think that a lot of people have a hard time in distinguishing what the customer impact is when you talk about service interruption,"LaChance says. "What we're trying to do is not so much establish that there's an issue in the network, but [understand] the services that were interrupted. How did that impact our customer?"

Telus achieved something along those lines, albeit only on a manual basis, as service provider for the World Summer Games in Edmonton, Alberta, where by placing certain conditions around various services it was able to set severity levels. Now it wants to take the next step and assess the true impact of failures on the customer.

"In order to do that and be able to attribute what the lifetime value of that customer is, the whole corporation has to get behind this. It isn't just operations."For instance, she says, marketing should develop a rating system that ranks customers by priority of action. The customer care area has to understand what the numbers mean so that they can ensure valuable customers aren't inconvenienced, even from the most minor problems or faults.

"It's all about getting a competitive edge and giving the customer a full suite of connectivity and giving them one name to identify with when they go to the market,"LaChance says. "We're trying to give them all those options. In other words, whether they want high-speed lambda services or a simple DSL line to someone's cabin out on the lake, we want to be able to provide those services. But if there's any linkage between that DSL line to the cabin and that lambda service, we should know that. We should know that internally."

And Telus also wants to be able to escalate warnings based on business impact, rather than on the SLA signed with the customer, as is now the case. "In other words, if a trouble ticket gets cut and even if we have a certain amount of time before we are going to be in breach of the SLA, the escalations happen automatically in the trouble ticketing system. They'll start paging out up to the manager, up to the director level and so on, to make sure that all the possible help is provided to get those issues resolved."

From there Telus hopes to add new and exciting services for customers. "Companies have big appetites for bandwidth in North America, so we want to be able to do this type of SLA management for such things as in our optical and lambda services (see below left) as we do with all our ATM and Ethernet environments,"LaChance says. "We want to be able to do things like auto provisioning where possible, auto correlation, bandwidth on demand.

"And at the same time as we're expanding and being able to provide all these new and fun things for the customer, we always want to make sure we're able to still provide the absolute concrete monitoring that we have today, and reporting and SLA agreements."

LaChance says the organisation has been able to take two major lessons from the exercise.

First, the importance of always ensuring your technology is buffered from the customer. Customers want to purchase services, not technology, she says. In essence, most customers don't care if their service is delivered over tin can and strings or the latest, greatest optical equipment. "As long as they know that it's getting delivered and that that application is working for them, the technology should be transparent to them,"she says.

And second, the importance of communications back to the customer to keep them informed of your activities on their behalf. "I think a report card is one of the most important things that you can give back to the customer on how their service is doing. So they know you know about their business and you can report back to them on how well you're doing to assist them in their business,"LaChance says.

Page Break

SIDEBAR: Enterasys Sheds Aprisma

In mid-August Enterasys Networks announced it had sold network management software company Aprisma Management Technologies to Gores Technology Group.

Effective immediately, the terms of the deal were not disclosed. Gores Technology Group (GTG) is a Los Angeles-based acquisition and management company.

GTG got all of Aprisma's assets, customers, products and employees in the transaction. A spokesman said Aprisma is looking forward to continuing on as an independent software company, no longer directly connected to a hardware maker. Aprisma had planned to become independent of Enterasys earlier this year, but instead the company found itself cutting staff.

Aprisma Management Technologies develops and sells the Spectrum network management platform (Aprisma's formation was announced in June 1999). Until the sale, Aprisma remained part of Enterasys. In May Aprisma closed its local Asia Pacific operations, sacking 15 staff.

Like DWDM (dense wavelength division multiplexing), lambda switching uses small amounts of fibre-optic cable and differing light wavelengths (called lambdas) to transport many high-speed datastreams to their destinations. But unlike DWDM, lambda switching injects intelligence, built around a set of evolving industry standards, into optical network gear.

The key to lambda switching is the ability to automatically connect optical network endpoints. In the past, arranging optical net communications required tedious, expensive configuration of each device, fibre, lambda and even higher-level protocols such as SONET and routing.

Because lambda switching connections, or light paths, are set up on the fly, using a scheme that's integrated with upper-layer protocols, network configuration can be simplified. And because light paths streamline operation, highly redundant network designs, previously deemed impossible to manage, can be built.