What is meant by the word “persona?”
“Persona” in the IVR world is the personality of the system itself. In the web world, it's used to refer to users.

Too often in the world of voice interface design, persona is used as a synonym for extreme personality. Whether you design it or not, all systems have a persona. The debate is, how much effort do you put into the description and definition of the persona ahead of time, and how strong a persona should the system have.

Explicitly define the qualities you want the persona to have
Almost all designers will agree on friendly and professional, although qualities like playful, reserved, hip, or nerdy will be used more sporadically in accordance with brand (see below).

Keep the style consistent throughout the system
A formal style guide can be helpful with this, especially when more than one designer is writing prompts for the system.

Make sure the persona is consistent with the brand
Brand relevance begins with engagement between a brand and its consumers/potential consumers. It is a key objective of a brand marketing effort. In general, the way a brand connects to its consumer is via a range of “touchpoints” – that is, a sequence or list of potential ways the brand makes contact with the individual. Examples include retail environments, advertising, word of mouth, online, and the product/service itself and of course the call center.

Maintaining the brand connection is an ongoing process. Ensuring all touchpoints feel consistent and on brand is critical in order to optimize the relationship that has been established with the organization.

Phone interaction still remains one of the leading touchpoints and has a major brand impact. Many customers receive the information they need to resolve their issue directly from the IVR, which increases customer satisfaction, reduces client cost and enables support professionals to focus on more complex inquiries, builds brand loyalty and promote up-sell/cross-sell opportunities.

It’s OK for applications to refer to themselves in the first person singular Although anthropomorphism in IVRs has been controversial among researchers in human–computer interaction (Gardner-Bonneau, 1999), it is currently common to script applications using the personal pronouns “I” and “me,” (especially when doing so allows more concise, conversational scripting than would be possible without the personal pronouns). There is some evidence that callers prefer applications that use personal pronouns even though they do not explicitly notice their use (Boyce, 2008). Even so, it is important to keep the user in focus when designing dialogs, avoiding applications that are excessively self-referential, which can be distracting to callers (Balentine, 2007).

Be careful in the use of named personae – most IVRs do not need a name There is general agreement among VUI designers that applications that speak to users elicit automatic reactions that lead them to imply personality characteristics of the artificial speaker (Dahl, 2006; Larson, 2005). Most IVRs designed to provide service to callers should have a voice that is generally friendly and moderately enthusiastic without going “over the top,” although there will be exceptions (for example, imagine an application designed for children to call and interact with while waiting in line at a theme park). In accordance with the appropriate social role of an IVR and the known characteristics of expert customer service representatives (Balentine & Morgan, 2001; Suhm, 2008), IVRs should assume the caller is busy, be efficient when communication is good, be helpful when progress is slower, be polite, rarely apologize, and never blame the caller.

Is there a need to go beyond this to create a more compelling and successful caller experience? This question has been at the heart of the debate over the use of personae in VUI design for more than 10 years (Balentine, 2010). As Balentine & Morgan stated in 2001 (p. 251):

A recent trend in the speech recognition community emphasizes what is called the persona as an essential part of a telephony application. A persona is distinguished from a personality in its emphasis on the fictional nature of the character. Promoters of this trend specify the persona thoroughly and painstakingly before approaching other details of the application—implying a belief that the persona is the foundation for every other design element. In many cases, considerable time and expense are devoted to the persona, including focus groups, psychological profiles, and voice talent audition and selection.on and selection. One of the arguments in favor of developing a persona is corporate branding (Kotelly, 2003) – which usually involves giving the persona a name. Another is to help designers produce a consistent caller experience by drawing on the characteristics of the persona to guide design elements such as tone, accent, specific wording choices, and confirmation styles (Cohen et al., 2004).

Relatively well-known named voice personae are Amtrak’s Julie, Sprint’s Claire, and Bell Canada’s Emily. As reported in The New York Times (Hafner, 2004):

She’s adventurous and well educated, friendly but not cloying, and always there to take your call. Meet Emily, the automated agent that answers customer-service calls for Bell Canada. So eager was Bell Canada to infuse the system with a persona that when Emily was introduced to customers nearly two years ago, she came complete with small-town roots and a history degree from Carleton University in Ottawa. Her biography said she played volleyball and had backpacked around Asia after college.

Providing this level of detail in the biography of a persona raises the question of exactly how playing volleyball or being a history major at Carleton can possibly affect IVR scripting decisions (Balentine, 2007; Balentine & Morgan, 2001; Dahl, 2006; Klie, 2007). Note that Julie is still at Amtrak, but Emily is no longer with Bell Canada (and Claire is no longer with Sprint). Other arguments against designs focused on personae are:

• Investments in VUI should be on the user and efficient task completion rather than a focus on crafting the persona—“The enterprise exhibits shock and dismay when callers reject the solution because no one could afford a simple usability test, and yet $100,000 and more went to a detailed analysis of her clothes, her marital status, and whether there’s a mole on her left arm” (Balentine, 2007, p. 102).
• When things go wrong in the IVR, having a branded name can give callers “something to latch onto … and that personality becomes a negative reflection on the entire company, not just the IVR” (Klie, 2007, p. 24).
• It might look fine on paper, but actually experiencing an “overly animated, artificially enthused IVR” (Rolandi, 2003, p. 29) can quickly become annoying to callers.

Walter Rolandi, one of the most outspoken writers against excessive investment in persona development, noted in his article “The Persona Craze Nears an End” that inappropriate, overly emotive personae can establish unrealistic user expectations, distract callers from their tasks, and annoy callers. Even appropriate personae cannot “make a dysfunctional application functional, add value to a questionable value proposition, or make an unusable application more usable” (Rolandi, 2007, p. 9). For practical VUI design, the following advice from Polkosky (2005b, p. 25) seems appropriate:

“Be concerned about persona, but not too much. While it is true that the speech and linguistic characteristics of an interface are important to user perceptions, persona should have the appropriate perspective: it is a secondary consideration after catering to users’ needs. Undoubtedly, the friendliness and naturalness of the system voice are important characteristics that need to be controlled and the prompts should convey helpfulness and politeness; but don’t let a blind focus on these design issues lead you to neglect the more important design decisions that enable a clear, simple, and efficient UI based on user goals.”

Apply branding cautiously
Service providers use IVRs to provide services to end users (who may be customers, employees, or other businesses). Service providers want their IVRs to save them money and to simultaneously maintain customer contact and satisfaction. Thus, they are concerned with their corporate image and how their IVRs fit into their branding strategies.

Keep brand in mind while choosing voice talent
One aspect of branding in an IVR is the selection of the voice talent. If an enterprise has an existing spokesperson or actor, it may be a branding-consistent choice to use that or a similar voice for the IVR.

Voice talent agencies want to work with designers and clients to help them select a good voice for their applications (one consistent with the enterprise's branding) (Graham, 2005, 2010).

A typical method is to provide samples of three or four voices to clients so they can choose one that appropriately represents their company. We recommend making sure that the top-level decision maker for this aspect of the IVR participates. As Lewis (2011, pp. 103-104) writes, “Trust me—you do not want to be in a meeting where you’re presenting the working version of the application (including all professional recordings) to the senior vice-president in charge of customer care who, upon hearing the voice for the first time, says, 'I hate it. We need a different voice.'”

Use branding tones only if they are clearly identified with the brand and short enough to not aggravate callers
Another aspect of branding is the use of branding tones or audio logos (think of the Sprint pin dropping sound). As with the use of branded voices, if an enterprise has an existing audio logo, it makes sense to use it to brand the IVR. One thing to keep in mind, though, is the importance of keeping audio logos used in IVRs as short as possible. Audio logos used in advertisements do not typically have this constraint, and can sometimes last several seconds longer than necessary to convey the brand impression. If it is possible to create a short version of a branding tone, consider doing so because branding tones will typically play as part of an IVR's introduction, where it is important to be as crisp and concise as possible to keep costs down and to create the most favorable caller experience.

Place branding content appropriately
“Poorly placed advertisements that inappropriately take up the caller’s time or offer directions that are unlikely to be followed, are a hit against the brand—unless the brand stands for slow, thoughtless service. This behavior frustrates the caller and wastes money” (Kotelly, 2006, p. 62).

As with any other content that is not central to the caller's situation, avoid putting that content in the caller's way. Designers often place concise branding content in the welcome statement of the introduction (e.g., “Welcome to the XYZ help line”) and in the closing (e.g., “Thanks for calling XYZ). Inside the IVR, excessive reference to the enterprise runs the risk of damaging rather than enhancing the brand.

After callers have finished their primary task and have indicated that they're ready to end the call, it's appropriate to present an up-selling or other personalized branding message.

Balentine, B. (2007). It’s better to be a good machine than a bad person. Annapolis, MD: ICMI Press.

Balentine, B. (2010). Next-generation IVR avoids first-generation user interface mistakes. In W. Meisel (Ed.), Speech in the user interface: Lessons from experience (pp. 71–74). Victoria, Canada: TMA Associates.

Balentine, B., & Morgan, D. P. (2001). How to build a speech recognition application: A style guide for telephony dialogues, 2nd edition. San Ramon, CA: EIG Press.

Boyce, S. J. (2008). User interface design for natural language systems: From research to reality. In D. Gardner-Bonneau & H. E. Blanchard (Eds.), Human factors and voice interactive systems (2nd ed.) (pp. 43–80). New York, NY: Springer.

Cohen, M. H., Giangola, J. P., & Balogh, J. (2004). Voice user interface design. Boston, MA: Addison-Wesley.

Dahl, D. (2006). Point/counter point on personas. Speech Technology, 11(1), 18–21.

Gardner-Bonneau, D. (1999). Guidelines for speech-enabled IVR application design. In D. Gardner-Bonneau (Ed.), Human factors and voice interactive systems (pp. 147–162). Boston, MA: Kluwer Academic Publishers.

Graham, G. M. (2005). Voice branding in America. Alpharetta, GA: Vivid Voices.

Graham, G. M. (2010). Speech recognition, the brand and the voice: How to choose a voice for your application. In W. Meisel (Ed.), Speech in the user interface: Lessons from experience (pp. 93–98). Victoria, Canada: TMA Associates.

Hafner, K. (2004, Sept. 9). A voice with personality, just trying to help. The New York Times. Retrieved from www.nytimes.com/2004/09/09/technology/circuits/09emil.html.

Klie, L. (2007). It’s a persona, not a personality. Speech Technology, 12(5), 22–26.

Kotelly, B. (2003). The art and business of speech recognition: Creating the noble voice. Boston, MA: Pearson Education.

Kotelly, B. (2006). Six tips for better branding. In W. Meisel (Ed.), VUI Visions: Expert Views on Effective Voice User Interface Design (pp. 61-64). Victoria, Canada: TMA Associates.

Larson, J. A. (2005). Ten guidelines for designing a successful voice user interface. Speech Technology, 10(1), 51–53.

Lewis, J. R. (2011). Practical speech user interface design. Boca Raton, FL: CRC Press, Taylor & Francis Group.

Polkosky, M. D. (2005b). What is speech usability, anyway? Speech Technology, 10(9), 22–25.

Rolandi, W. (2003). The common causes of VUI infirmities. Speech Technology, 8(6), 29.

Rolandi, W. (2007). The persona craze nears an end. Speech Technology, 12(5), 9.

Suhm, B. (2008). IVR usability engineering using guidelines and analyses of end-to-end calls. In D. Gardner-Bonneau & H. E. Blanchard (Eds.), Human factors and voice interactive systems (2nd ed.) (pp. 1–41). New York, NY: Springer.