Cookie Preference Centre

Your Privacy
Strictly Necessary Cookies
Performance Cookies
Functional Cookies
Targeting Cookies

Your Privacy

When you visit any web site, it may store or retrieve information on your browser, mostly in the form of cookies. This information might be about you, your preferences, your device or used to make the site work as you expect it to. The information does not usually identify you directly, but it can give you a more personalized web experience. You can choose not to allow some types of cookies. Click on the different category headings to find out more and change our default settings. However, you should know that blocking some types of cookies may impact your experience on the site and the services we are able to offer.

Strictly Necessary Cookies

These cookies are necessary for the website to function and cannot be switched off in our systems. They are usually only set in response to actions made by you which amount to a request for services, such as setting your privacy preferences, logging in or filling in forms. You can set your browser to block or alert you about these cookies, but some parts of the site may not work then.

Cookies used

Performance Cookies

These cookies allow us to count visits and traffic sources, so we can measure and improve the performance of our site. They help us know which pages are the most and least popular and see how visitors move around the site. All information these cookies collect is aggregated and therefore anonymous. If you do not allow these cookies, we will not know when you have visited our site.

Cookies used

Google Analytics

Functional Cookies

These cookies allow the provision of enhance functionality and personalization, such as videos and live chats. They may be set by us or by third party providers whose services we have added to our pages. If you do not allow these cookies, then some or all of these functionalities may not function properly.

Cookies used




Targeting Cookies

These cookies are set through our site by our advertising partners. They may be used by those companies to build a profile of your interests and show you relevant ads on other sites. They work by uniquely identifying your browser and device. If you do not allow these cookies, you will not experience our targeted advertising across different websites.

Cookies used


This site uses cookies and other tracking technologies to assist with navigation and your ability to provide feedback, analyse your use of our products and services, assist with our promotional and marketing efforts, and provide content from third parties


Here are some suggested Connections for you! - Log in to start networking.

Article : Optimizing the Customer’s Speech Experience

According to Gartner analysts, 92 percent of customer interactions happen over the phone. That's a huge number for any organization and makes a strong case for more effective phone communication. Companies have continued to provide multiple methods to serve their customer needs, including the Web, but global cell phone subscriber growth continues to outpace Internet user growth, indicating that the phone service channel will continue to be an extremely important communication medium for the foreseeable future.

The vast majority of phone interactions begin in some type of automated application, which are deployed with the goal of enabling customers with low to medium complexity questions or needs to serve themselves. These systems almost always include DTMF, or touch-tone IVRs, and many companies are also deploying speech recognition technologies in their automated offerings as well. Having a complete understanding of customers' experiences in these automated applications is imperative—as they act as a "welcome mat" to the majority of customer experiences.

Dr. Tal Cohen, Ph.D
CEO and Co-founder

Recently deployed speech applications, as well as existing IVRs, are typically designed and deployed with clear understanding of the business objectives these systems are designed to achieve. For example, most organizations deploy these types of applications to save money—according to Forrester Research, the average cost of a live customer service agent phone call ranges from $5 to $15, compared to the majority of self-service interactions that typically cost no more than a dollar or two per transaction.

The challenge to date has been that user goals are much more difficult to identify, and information and reporting is often limited or unavailable to analyze the true user behavior in these systems. Most IVR Applications rely on high-level volumetric data, anecdotal evidence or incomplete user surveys for improvement answers, resulting in systems that are poorly designed and subsequently do not meet user goals. The ability to monitor and measure actual user behavior in interactive voice response (IVR) and speech-enabled applications enables identification and improvement of problem areas that cause user drop-off or confusion and additionally define areas where automation can be extended to better serve customer requests.

In order to optimize any system, organizations must identify how they are going to measure the success of the system. Self Service applications such as IVRs are no different in this regard. If cost reduction is one of the system objectives, then measuring how successful the system is in containing and self-serving customers inside the system—and avoiding the costly live agent phone call—should be one of the measures. In this case, understanding the user behavior inside the system is essential to knowing whether or not the user goals are aligned with the system objectives. Many IVR applications not only do not meet user goals, but also fail to deliver the expected cost reductions due to sub optimal design.

Why? Because today, many IVR systems are modified on gut feelings, subjective guesses or anecdotal, random information about what customers want—not with hard data about the users' behavior and desires. Then companies spend millions of dollars on new technology to improve the IVR system, such as speech recognition, in an attempt to "fix" the problem. However, the root cause is still unresolved, and since companies do not have a way of factoring in actual users' behavior systematically, the new technology falls short of its potential just as its' predecessor.

A new approach to IVR/Speech design and modification is necessary to enable companies to manage IVR business performance. Understanding user behavior is critical in order to match business objectives to those of the customers. For example, if 50,000 people call the IVR/Speech system each day, but 10,000 of them leave the system without completing their purpose, companies need to know why.

The best way to align these factors is by focusing on user behavior—understanding how customers are acting once inside your IVR/Speech system, and subsequently incorporating that insight with existing customer and transactional information. Leveraging these knowledge bases to drive user-behavior based IVR/Speech improvements.

Customer Behavior Intelligence technology can help organizations monitor and measure user actions in interactive systems, including speech recognition applications. Speech applications are unique in their implementation, in that they typically involve a lengthy tuning process to optimize the success of the application. Customer Behavior Intelligence can improve the effectiveness, speed the time-to-benefit and increase the associated ROI for Speech deployments. Specifically, Customer Behavior Intelligence can be leveraged to achieve the following items:

  • Ensure smooth DTMF-to-Speech migration
    By analyzing the user behavior in existing DTMF-based applications, a baseline assessment can help design a more effective voice interface that delivers optimal results.

  • Align call flow and business logic with user needs and wants
    A thorough analysis of overall user flows through the system shows immediately where users might be experiencing problems with context or overall experience that might suggest an alternate sequence of options or functionality to improve user task completion.

  • Link utterances to behavior to optimize grammar and recognition
    Analyzing user behavior will uncover the actual user utterances in context of the user behavior experienced. Having a complete understanding of aggregate user behavior with the associated utterances can help optimize flow, grammar and recognition for maximum effectiveness.

  • Optimize placement of speech recognition technologies
    User behavior analysis can pinpoint the optimal areas for speech recognition placement in the overall self-service application, increasing the transaction throughput and conversion rates.

  • Maximize tuning effectiveness
    Through monitoring and measuring user behavior, identifying and improving speech applications automates the entire tuning process, increasing the time-to-benefit and improving the ROI.

Customer Behavior Intelligence can be deployed on current DTMF IVRs before speech applications—that way managers can look at existing systems and get an understanding of current behavior and identify areas where deployment of speech applications would be most effective. For example, it might make sense to deploy speech within an application with low throughput resulting from a complex set of steps. If speech applications are already deployed, Customer Behavior Intelligence can specifically help organizations understand how customers are utilizing Speech applications, highlighting areas of sub optimal performance in three specific areas:

At each point within a speech system, whether it's natural speech or directed dialogue, the system can recognize customers utterances so long as they are included in a pre-programmed set of words defined as "grammar." For example, if a customer gets to a main menu and says "billing options" that phrase should be "in grammar" at that particular point in the menu. If a customer attempts to say something that's not "in grammar," the system won't recognize it. The list of words recognizable by the application is called the grammar list.

Customer Behavior Intelligence helps organizations identify points in their system in which customers are being routed to customer service representatives. For example, if a customer calls his or her local phone company and utters "state-to-state calling" and the application does not recognize this utterance as being "in grammar," the customer will likely end up pressing zero to talk with a customer service agent. By utilizing Customer Behavior Intelligence along with a transcription service, organizations can identify bottlenecks in their system and also words and phrases that might need to be added to the system.

In addition to translating end user utterances into "in grammar" phrases, speech technology assigns a confidence level to the utterance. If a customer says "billing options" and it matches to a certain grammar, the system might assign it a 70 percent confidence—meaning it is 70 percent sure it has made a correct match. Managers can set a threshold by allowing customers above a 70 or 80 percent confidence level to be routed to the next step in the system.

Quite often the speech application recognizes what a customer says, but does not with the level of confidence needed to route the customer. Customer Behavior Intelligence allows organizations to combine the visibility of what the caller said, as well as the confidence level assigned by the system, enabling visibility into confidence problems.

The area ignored by tuning technologies and arguably the most important aspect to success, Dialog is described as the "customer experience," or flow through the application. Customers often move from the main menu to a "disambiguous state to further segment their need." Or, a customer might be funneled into one of four main menu categories based on their selection or utterance. Dialog will be very similar whether you're working with a DTMF or a speech application—it's a set of steps needed to accomplish a task.

Customer Behavior Intelligence will provide organizations with a map of how, in aggregate, customers navigate through an IVR or speech application. The map will show deviations from the expected flow, and the implications of these deviations. For example, if customers continue to get caught in a disambiguous state, rather than efficiently moving through the billing application and on to pay their bill, this deviation and its' impacts will be clearly understood with customer behavior intelligence.

As self-service options continue to proliferate the consumer experience, companies are finding it essential to create interactive user experiences that are both helpful and meaningful. At the same time, these experiences must meet bottom line business objectives – cost savings, revenue generation and customer retention. As companies respond to these opportunities, they are finding a variety of reasons to reinvigorate the IVR system.

With behavioral insight, companies can better balance and meet business objectives. The key to optimizing business performance is to turn meaningful observation and analysis into new, value-driving action.

About Dr. Tal Cohen, Ph.D:
Dr. Cohen is president, CEO and co-founder of ClickFox. ClickFox's software modeling solutions enable its customers to translate complex customer interactions across multiple self-service channels—such as IVR, speech recognition, Web sites, kiosks and CRM systems—into fact-based decisions that optimize channel and cross-channel business performance. Dr. Cohen holds a Ph.D. in Mechanical Engineering and a Master of Science degree in Computer Science. A data modeling expert and a leading research faculty member and educator at the Georgia Institute of Technology, Dr. Cohen alongside Nissim Harel, conceived the original concept of ClickFox. Through his vision and leadership ClickFox is growing from an idea to counting among its customers several Fortune 100 companies.

Today's Tip of the Day - Motivation & Charity Work

Read today's tip or listen to it on podcast.

Published: Wednesday, August 4, 2004

Printer Friendly Version Printer friendly version

2023 Buyers Guide Surveys

Nuxiba Technologies

Provide an unparalleled, engaging user experience for the creation of dynamic voice broadcasting campaigns. Its block-based interface enables you to develop easy to complex call flows aligned with given strategic business targets by simple drag and drop operations.

Mass Automated Telephone Surveys are based on a key-pad answering system that enables you to poll large audiences with your own custom questions. They are great for:

- Building Customer Loyalty
- Lead Generation
- Enhancement of Current Service Portfolios

Manage the platform with great ease and assess your results. Schedule your demo or request a quote today!


Pointel Survey360 is a closed-loop feedback solution designed for business users to create and manage surveys. It is equipped with the ability to capture and analyze data in real-time.

Survey360 provides a competitive edge that improves our customers' business strategy with actionable intelligence. In addition to the details in the surveys themselves, Survey360 enables users to generate interactive reports on demand even without the hassle of coding to interpret their customers' feedback and translate survey results into improvement recommendations.

About us - in 60 seconds!

Submit Event

Upcoming Events

Europe's leading call & contact center event is now arriving at the U.S., showcasing the latest and most effective technologies, strategies and advancements to industry professionals who are looking to excel in the customer engagement world!

Disco... Read More...

Newsletter Registration

Please check to agree to be placed on the eNewsletter mailing list.

Latest Americas Newsletter
both ids empty
session userid =
session UserTempID =
session adminlevel =
session blnTempHelpChatShow =
session cookie set = True
session page-view-total = 1
session page-view-total = 1
applicaiton blnAwardsClosed =
session blnCompletedAwardInterestPopup =
session blnCheckNewsletterInterestPopup =
session blnCompletedNewsletterInterestPopup =