Best Practice Spotlight Developing Voice…

              Best Practice Spotlight
              Developing Voice Self-Service
              Solutions Customers Love




White Paper




              www.goldsys.com
    Most environments have already adopted, or are in the process of adopting, speech/voice
    self-service solutions and abandoning their touch-tone applications. Most recognize the revenue
    generation and efficiencies gained from automating services with a speech solution that allows callers
    to get what they need without speaking to a live agent. However, many continue to struggle with the
    overall usage, performance and caller satisfaction of their voice self-service solutions.
    Why? There could be several reasons but one overarching reason that continues to plague the caller
    satisfaction of a voice, self-service solution is the fact that caller experiences are subjective and can
    vary greatly. One voice self-service solution may evoke an extremely positive experience in one caller's
    mind and might evoke the exact opposite in another.
    So, how do you marry the operational convenience and cost savings of a voice self-service solution
    with a satisfied, positive caller experience? By understanding that common self-service caller "stereo-
    types" exist (i.e., Mary and Bob- more on them later) and learning the design/implementation best
    practices of companies who have convinced customers to embrace and even love, using their voice
    self-service solutions.
    In this white paper you'll learn:
    · The characteristics of two common voice self-service users.
    · 6 best practices to deploy voice self-service solutions that customers love.
    · Tips on finding the right speech solution partner to fit your needs.

    Bob vs. Mary ­ Understanding Common Self- Service Users
    Companies wanting to use the power of speech in their environment are struggling to
    find that "teeter-totter" balance of functionality and usability with their self-service solutions. Here
    is where the saying, "Know your Audience" is going to become critical to your call center solution's
    success. If you do your homework and learn from those who have gotten it right, you'll implement a
    solution your customers actually use and enjoy.
    The speech industry has spent tremendous amounts of time, money and research on studying
    common caller experiences. This research has helped define two common caller groups: those who
    embrace the idea of using an application to speak their service/informational request and those who
    feel that "talking to a machine" is annoying and difficult to navigate.
    For the purposes of this discussion, we'll give our two aforementioned self-service caller groups the
    names Bob & Mary.
    Bob is a caller who generally hates self-service telephone applications and for good reason. He's
    encountered too many bad touch-tone applications where he has entered the information he was
    asked and didn't get the result he was looking for. He is the caller who instantly "zeros out" to speak
    to a live agent. His perceptions of voice self-service solutions are that they were developed "just to
    annoy him in a new, high-tech kind of way."
    Mary, on the other hand, enjoys the simplicity of speaking her request and receiving the exact infor-
    mation she needs. She sees the convenience in not having to wait to speak to a live agent to conduct
    her business. She sees this technology as helpful and knows that if she has a more difficult question
    or transaction, she can ask to speak to a live agent to assist her in those instances. For the most part,
1   Mary recognizes that technology is not perfect but her perception of voice self-service is positive.
                                                                                         White Paper




6 Keys to Implementing a Solution BOTH Bob and Mary will Love
So how do you please both Bob and Mary? Realistically that may not be possible. You know that you
can't please everyone but the best way to mitigate negative user experiences is to know how to avoid
some common pitfalls.
Using the knowledge and best practices of speech veterans can help ease your solution's acceptance
and ultimately be responsible for its success. New technologies always see initial resistance from a few
"Bobs." But the more "Marys" you can convince to use your self-service application with a positive
result, the better chance for its overall acceptance.
Here are some "lessons learned" that are critical in developing a voice self-service solution that's
accepted, used, and even enjoyed by your customers.
1. Ask consistent questions for consistent answers.
   One of the keys to getting good self-service performance, and ultimately creating a positive user
   experience, is to ask questions that generate consistent responses. If you confuse the caller with a
   complex question ­ and remember they may be driving or eating their lunch and not always paying
   close attention ­ they will answer something like "ah, uh oh let's see, I think uh, yes my account
   number is, uh...CLICK." It is going to be a long time before speech recognition engines can get
   much out of that sort of response.
2. Make sure your design has passed the "human factors" test.
   Design a self-service application that callers want to use and they will use it. Make sure that your
   developers have considered accents, slang and other "human factors" in your solution design.
3. Test, test and re-test.
   There are no shortcuts when it comes to tuning and refining a self-service solution. As part of
   the self-service solution development and deployment process, plan on monitoring calls and using
   the resulting information to improve the application. Monitoring, tuning and usability testing was
   important with touch tone applications, though unfortunately some developers did not utilize
   best practices, resulting in poor user experiences. With speech recognition, monitoring, tuning
   and usability testing is crucial to a successful application with high rates of call completions.
4. Consider the entire self-service user experience.
   Do not settle for part of your self-service solution experience being great, while neglecting the rest
   of the caller's experience. From the moment they pick up the phone to the moment that they hang
   up with the information they need, you are responsible for the complete user experience.
  Take advantage of existing technology, including CTI (Computer Telephony Integration) and expected
  wait time announcements. With CTI, the self-service application can pass information collected
  from the caller to the live agent, as well as information about why the caller is transferring out.
  It is very irritating to callers to be prompted for a long account number by a machine, only to be
  asked to repeat the same account number to the live agent. A great speech recognition application
  surrounded by substandard legacy applications, poorly planned routing, insufficient live agent
  resources or unreliable connectivity to back-end systems will ultimately be perceived as a poorly
  performing application.
5. Don't insult callers who have problems with the self-service system.
   Instead of telling the caller that "their input is invalid", consider taking responsibility for the mistake
   and re-prompt with a little more help. For instance, saying "I'm sorry, I didn't get that. Please repeat
   your 10 digit account number, which can be found in the top right-hand corner of your statement,"
                                                                                                                 2
      will reassure the caller that the recognition problem was not their fault. A poorly designed system
      that attempts to keep callers "in the system" only results in angry and annoyed callers who will
      eventually take their frustrations out on your live agents, or worse, they will call your competition.
      And if the caller DOES want to talk to a real human, LET THEM!
    6. If you're not a voice self-service expert, consider finding one.
       Voice self-service technology is not rocket science but there are some insider secrets to developing
       solutions that your customers are going to want to use. Just as a new set of woodworking tools
       doesn't make a craftsman, the new tools available that make speech recognition development look
       easy will not result in great applications unless the development team understands the design issues
       that are particular to today's complex enterprise applications.

    Finding the Right Speech Partner for Your Needs
    Voice, self-service solutions are powerful and proven solutions that offer real improvement in customer
    service while simultaneously delivering cost savings and efficiency rates that will positively affect your
    bottom line. If you have a self-service solution that isn't reaching its potential, or are looking to imple-
    ment one, you may want to utilize the expertise of a speech application expert.
    Here are some things to keep in mind when choosing a provider:
    · Do they have a proven process with regard to design, interface and project management? Do they
      have vertical expertise to understand your industry's challenges? Choose a provider who has a
      repeatable and proven process that has developed results for similar companies as yours.
    · Have they invested in development and service creation tools? Do they re-use some portion of the
      solution code? Do they have off-the-shelf speech dialogues available for your use? If so, you'll be
      able to get your solution developed in a shorter time frame with the same level of expertise and
      functionality you desire.
    · Do they have technology integration experience? Most likely you'll need your voice self-service
      solution to integrate into your current telecommunications platform, which includes both your voice
      and data infrastructures. Also, your self-service solution will need to access several back-end data
      sources in order to get the information requested by your application user. Make sure that your
      provider has proven expertise in the complete technology solution.
    · Are they experts in developing solutions that make all self-service users happy? You want your
      self-service solution to be used, and you want it to create a positive experience with your end users.
      A speech application veteran will know all the "human factors" issues to test for will finely tune
      your solution to provide maximum results.
    · Who have they done work for? Ask to hear and see solution demos as well as speak with former
      and current clients. You will learn a lot on how the company performs, and more importantly
      supports, a solution once it has been developed.
    This white paper only scratches the surface of what it takes to build voice self-service solutions your
    customers want to use and even enjoy. Finding an experienced voice application provider who can
    make your self-service goals a reality will get you one step closer to reaching your ultimate customer
    satisfaction goals.

3
                                                                                     White Paper




About Gold Systems
Gold Systems' self-service, speech recognition and text-to-speech solutions allow businesses to
improve profitability while providing customers with a service interface they love to use. Using
our proven Solutions Discovery Process, Gold Systems helps call centers and enterprise businesses
identify business processes that can be optimized via a custom or packaged voice-driven solution.
Gold Systems is the premier developer of solutions for the financial, insurance, healthcare, telecom,
retail and utility industries.
To view our solution demos, a client list or get more information on how you can enhance your
service strategy with a self-service solution, go to www.goldsys.com.




     Stay tuned for our next Best Practice
     Spotlight! Next time we'll focus on
     Human Factors Testing Best Practices.




                                                                                                        4