![]() |
![]() |
|
|||||||||||||||||
|
Wednesday, February 8, 2006 EIG's Rex Stringham on Hosted Speech Rec:Enterprise Integration Group, Inc. (EIG) is a consulting firm that helps companies build better IVRs. Their mission: "EIG is the premier independent professional services firm specializing in simultaneously improving IVR utilization and customer satisfaction." EIG helps plan, design, and test IVR and automated speech recognition applications, among other services. We asked EIG's president and co-founder Rex Stringham a few questions about hosted speech recognition services, and he responded via e-mail. Call Center Magazine: What are the advantages of hosted speech recognition service over premise-based applications? Is there a trend toward hosting? Rex Stringham: The first advantage is protection from technology obsolescence. By purchasing capacity from a service provider, it becomes the service provider's responsibility to keep up with new releases of the underlying speech recognition platform. Second, a hosted solution can be especially cost-effective if call demand fluctuates dramatically. For example, an electric utility may have spikes in demand during a power outage. Rather than purchasing a significant amount of premise-based capacity that sits idle except for these rare events, the hosted solution allows an enterprise to only pay for what they use, when they use it. Third, the hosted solution eliminates or at least reduces the need to ramp up and maintain expertise on speech recognition. While many enterprises have resources and expertise for touch-tone-based interactive voice response (IVR), speech recognition requires a significant and potentially expensive "re-tooling" of internal resources that can be avoided through hosting. There does appear to be a trend toward hosting, and it appears to be largely driven by FUD (fear, uncertainty, and doubt) about speech recognition. Speech recognition is significantly more expensive than touch-tone. By shifting responsibility for optimizing speech recognition to the service provider, the initial investment and risk with speech recognition is minimized. Call Center Magazine: What should call centers keep in mind when choosing a vendor and designing an application? Rex Stringham: Insist on having your vendor commit to service level agreements (SLA). That is, the vendor should only get paid if they meet mutually-agreed performance objectives. If the vendor is not willing to "put their money where their mouth is" then they don't really have confidence in their ability to deliver. If they are willing, then you "lock in" the return on investment (ROI). The biggest mistake we see being made is the wholesale replacement of touch-tone with speech recognition, under the assumption that pressing buttons is why people hate IVR. Don't do this. Many callers actually prefer touch-tone to speech recognition for privacy, speed, or when there is background noise. So allow customers to choose the user interface that best meets their needs. Furthermore, implementing speech recognition is not rocket science. There are simple, practical, and cost-effective solutions to today's business problems. Don't be sucked into the Hollywood hype. Focus on your business goals and look for ways to use speech recognition to improve customer service, especially for things that can't be done well with touch-tone. Call Center Magazine: Paul English has gone on an anti-IVR media rampage and now has a new website (gethuman.com). Why have speech and IVRs gotten such bad press? Rex Stringham: Paul English only addressed one common problem with IVR and that is, that many enterprises try to "lock" callers into an IVR by hiding the operator option. This has never been a user interface design best practice. Well-designed IVR and speech recognition systems always have a straightforward option to reach live answer. Quite simply, there are not enough operators in the world to handle calls if IVR and speech recognition technology is replaced with humans as Paul English recommends. And I don't see Paul English volunteering to pay all the extra operator salaries. There are good reasons to hate IVR and speech recognition, but it's not the technology's fault. There are dozens of ways to make IVR more efficient, effective and satisfying for customers. The good news is that well-designed IVR systems are actually preferred by customers over live answer, and customers complain when well-designed IVR systems are taken away. The IVR and speech recognition industry needs to focus on good user interface design to avert customer backlash. Paul English offers no real solutions. He's had his 15 minutes of fame and now he needs to sit down and shut up. Posted by Harry Sheff on Wednesday, February 8, 2006 at 1:37 PM |
Free CallCenter Insider Newsletter
|
|||||||||||||||||||||||||||