Right-sizing your call centre for today’s market

Call centres that use telephony to route customer calls to anonymous agents armed with scripts are a ubiquitous feature of modern business, and some experts say virtually every company should have a call centre. Others say none should. Why? Because the model is broken. Nevertheless, technology vendors are adding hosting options, home-based agency systems, social networking and video-based interactive voice response (IVR) to the dizzying array of features and models already available. What factors should companies evaluate to determine what they really need? Start with basic business fundamentals Most companies need some type of call centre function, even small companies, says Keith Smithers, practice leader for contact centres at Avaya Canada. “Even if you only have five agents, you still need coverage for lunches and breaks,” he says. At a basic level, call centres use technology to take customer calls, then queue and route them to an area based on business rules. Direct dialling, by contrast, directs calls to a specific person. “Customers are looking for knowledge, not a specific person,” he says. An array of business factors need to be considered in setting up a call centre, be it an in-house or outsourced arrangement, says Gary Audin, president of Delphi Inc., an Arlington, VA.-based telecom consultancy. The types of coverage needed – weekends and after-hours, different time zones, and languages needed to service customers – are key factors. The volume of calls is also important, as is duration, since diagnostic or tech support calls may be low in volume but lengthy. But long distance charges are a minor factor nowadays, and even less so if VoIP is used. “Unlike the olden days, call centres don’t need to be near population centres anymore.” Services that can be easily commoditized and scripted such as order-taking are best suited for call centre operations, says Audin. “If scripting is hard to do, then outsourcing is not suitable.” Agent turnover rate is another critical factor to evaluate. “If it’s high, training costs may negate all the savings in cheaper labour. Some call centres do so poorly, the turnover rate is 100% annually.” Some industries have seasonal fluctuations in calls, so they need to accommodate that, says Audin. “LL Bean, for example, has peak traffic from November to January, but doesn’t need 75% of its lines the rest of the year. So it has a deal with its carrier to turn off – but not disconnect – its call centre lines, then bring them back up again next year.” If the retailer had a VoIP system, it might not need the lines, but then it would need extra bandwidth, he adds. “Either way, you end up with more capacity than you need.” Government entities have privacy and regulatory issues in addition to the usual business issues to consider in outsourced arrangements, he says. “But there are some cases where special groups do a particular function for government. 911 calls, for example, are outsourced to call centres, but agents deal exclusively with these calls and nothing else.” Snafus in call centre implementations typically revolve around human elements, not the technology, says Smithers. “Issues come up when agents aren’t trained for the types of calls coming in, or are covering too many product sets and have to hunt others. Predicting volume is another one we still see even in large enterprises – often the marketing department runs a campaign that increases calls without warning the call centre.” How to avoid common pitfalls While these are the business factors typically evaluated in setting up call centres, there are some major holes, says Tripp Babbitt, president of Bryce Harrison Inc., a quality assurance consultancy based in Indianapolis. Most companies don’t consider failure-demand – customer calls about a product or service that’s failed to deliver in some way – or distinguish it from value-demand calls to place orders, open accounts, and other revenue-generating functions. “Failure-demand is about 25 to 75% of calls in most service organizations. By not knowing what that is, many organizations are in essence outsourcing their failure-demand to another organization when they set up their call centres,” says Babbitt. While the original idea of consolidating service to gain economies of scale was right in intention, this has typically been translated to mean delivering standardized service, he says. “This makes service worse, because the system isn’t capable of absorbing the variety of demand customers need, which increases failure-demand. The customer must call back repeatedly to get a request understood.” Another mistaken assumption is the belief that agents can be held accountable for performance. “The way the work flow is usually designed, the system is responsible for 95% of performance, and only 5% is attributable to individuals. They have no control over the scripting if it doesn’t match customer demand, or the system – some have to move in and out of five or six applications to get information.“ Scripting is typically done in a top-down fashion by layers of consultants and management, and it’s difficult to change the system to respond to new issues, he says. “Organizations have to ignore the standard mantra about using tracking measurements such as average hold time. Agents get graded on these metrics instead of being participants in the system. What we’re really talking about is making agents smarter, and not dumbing them down with scripts and technology.” Babbitt points out another mantra recited since time immemorial is that organizations should focus obsessively on customer service. But he says he’s not necessarily advocating that companies spend more in this area. “What we’ve found is that, paradoxically, call centre costs go down if you smarten up agents. If you spend extra minutes on the phone deciding what the customer really wants, you don’t get second or third calls later from upset customers – and you don’t need to spend time writing scripts.” In setting up their call centres, companies should start by distinguishing between value- and failure-demand and how to respond appropriately, and providing training that allows agents to do one-stop resolution. “Some companies in Europe are starting to get it, and they’re giving higher levels of service than in North America. The US got its butt kicked in manufacturing, and it will get kicked on services if there’s no change.” The hottest trends for reducing opex costs The ferocious economy is driving many companies to move from a capex to an opex model, says Barb Rigel, senior marketing manager at Genesys, a call centre technology provider based in Daly City, CA. “This is the biggest trend in call centres, according to analysts,” she says. “Most companies set up capital expenditure budgets a year in advance so it’s hard to keep up with new technologies, but in an operating expense model, you pay as you go.” As a consequence, hosted models are becoming more popular than buying in-house infrastructure. The main factor to consider is the number of years companies believe the technology will serve, says Rigel. “If it’s about three to four years, it’s typically less expensive to go with hosting.” This has to be balanced against a loss of control over call centre operations, says Smithers. “It could take weeks or months to make changes to systems, whereas they can be made on the fly to respond to changing business conditions with an on-premise solution.” But Rigel counters that hosted solutions are faster to implement, and can be ready to go to market in weeks. Home-based call centre agents is another big trend, and this type of distributed model is easier to manage with a hosted solution, says Rigel. “There were about 47,000 home-based US agents in 2008, and this is expected to grow to about 225,000 by 2012.” Use of home-based agents is on the rise as it’s a cheaper option – companies don’t have to worry about real estate to house their agents, and their operations aren’t limited to any one region. Pure and hybrid hosted arrangements in general are increasing as companies grow more comfortable with the concept of software as a service (SaaS), she says. “When companies first started outsourcing their call centre operations to India and other far-flung places, hosting wasn’t really seen as a lower cost option. As it gets more popular, it will probably negate the need for offshore call centres.” The number of hosted call centre agents is expected to grow from 357,000 in 2008 to 1.6 million by 2012, she adds. Demand from SMB companies has been a big driver in developing more robust hosting solutions. By 2014, about 80% of all hosted spending will be coming from SMB companies with less than 500 employees, says Rigel, citing Frost and Sullivan figures. The market for small companies with less than 100 employees is particularly lucrative, with forecasts of about a billion dollars spent on hosted applications. “Smaller companies can’t afford to spend millions on technology, but they have to look larger to remain competitive, so they need those hosted apps to survive.” To outsource or not? Clearly Contacts, a Vancouver-based company that sells discount prescription contact lens and eyeglasses online, recently set up a call centre. The company has experienced significant growth, expanding over the last decade from a two-man operation to 150 employees, of which 50 are full or part-time client service representatives (CSRs), says Kam Dhillon, director of customer service. “We were looking for a cost-effective way to offer our products in more time zones and in different languages – we needed French for Canada, Spanish for the US, and we also have a lot of Japanese customers,” she says. Clearly Contacts considered, but rejected, an outsourcing arrangement. “Outsourcers charge premiums on everything: agents, telephony, long distance charges,” says Dhillon. When the costs of flying trainers out to India or Philippines and various premiums were factored in, the outsourcing costs were more than using the company’s existing CSRs with an on-premise system. “I would say you need about 100 agents before you break even by outsourcing your call centre.” The company opted for a system developed by Nortel Networks, which provided the technical support needed to set up the system. But setting up the scripts and deciding which features to select from the dizzying array available were Dhillon’s responsibility – and biggest headaches. “There’s all this stuff in the background we had to think about: how calls are routed, how many agents in each queue, how long do customers wait in one queue before you bump them into another. And all the features: do we want all our calls recorded, what kind of reports do we want, and so on.” The guiding principle Dhillon used to decide was finding a system that was the right size for the company and its near-term growth. “I believe this system meets our current needs and for another three years. But my original pick with another vendor a few years ago didn’t keep up – we didn’t get enough support.” Social media and other new bells and whistles Some new social media-based models are emerging to complement call centres. Wetpaint, a social media platform provider based in Seattle, has developed sites for major technology companies such as T-Mobile, HTC and Dell Computers. These Wetpaint sites are typically set up for tech support, says marketing director Devin Hughes. “Companies want a destination their users can go to find answers to questions they might potentially call tech support about. The user-generated content is written in the same language users would be using to search, so they can often find the answer right away instead of spending hours waiting in a call centre.” While most of these companies already have tech support sites, these are difficult to keep up-to-date, and are often not very friendly, he says. “Companies that have lots of products don’t want to spend time writing out solutions to every single problem. If someone bought a T-Mobile phone five years ago, it may be impossible to find relevant support for it today. But if you set up a social media site, the community figures out all the fixes and you get a never-ending stream of user-generated content for every product.” HTC, which has the largest Wetpaint community of 21,000 members who’ve generated 17,000 pages of content, has seen decreases in tech support calls, says Hughes. While hard numbers aren’t available yet, he says 81% of the site’s traffic is driven from search engines, which means these visitors are new to the site and have arrived via a question or issue they've typed into Google. “A conservative assumption would be that 10% of the visitors to the site find a solution and therefore don't move on to call HTC's support centre.” Mainstream call centre providers are also eyeing the social media space. Avaya is working to incorporate social networking capabilities in its call centre technology, says Smithers. “For example, agents could use Facebook to understand who’s calling in. Customer service could be improved if you can get background on customers by pulling up their profiles. While this might not work with older people, who tend to keep their profiles private, it may be useful with younger people, who tend to have public profiles.” Another new development is video IVR, which already has some working prototypes, says Rigel. “This isn't about showing a video of the agent, but video capabilities that allow customers to show the product, or if they're sitting in a queue, the company could show a new product on-screen or a funny video to keep their attention instead of playing elevator music.”

Right-sizing your call centre for today’s market

News | 03/24/2009 3:34 pm EDT

To keep reading, get a free trial.

Already a Subscriber?

Reuse & Permissions

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us