Avaya Speech Analytics Reporting

Release 2.0.2
Issue 2.0
March 2013
© 2013 Avaya Inc. All Rights Reserved. part of its purchase of the Nortel Enterprise Solutions Business in December
2009. The Heritage Nortel Software currently available for license from Avaya
Notice is the software contained within the list of Heritage Nortel Products located at
While reasonable efforts have been made to ensure that the information in this http://support.avaya.com/LicenseInfo/ under the link “Heritage Nortel
document is complete and accurate at the time of printing, Avaya assumes no Products”. For Heritage Nortel Software, Avaya grants Customer a license to
liability for any errors. Avaya reserves the right to make changes and use Heritage Nortel Software provided hereunder solely to the extent of the
corrections to the information in this document without the obligation to notify authorized activation or authorized usage level, solely for the purpose specified
any person or organization of such changes. in the Documentation, and solely as embedded in, for execution on, or (in the
event the applicable Documentation permits installation on non-Avaya
Documentation disclaimer equipment) for communication with Avaya equipment. Charges for Heritage
“Documentation” means information published by Avaya in varying mediums Nortel Software may be based on extent of activation or use authorized as
which may include product information, operating instructions and performance specified in an order or invoice.
specifications that Avaya generally makes available to users of its products.
Documentation does not include marketing materials. Avaya shall not be Copyright
responsible for any modifications, additions, or deletions to the original Except where expressly stated otherwise, no use should be made of materials
published version of documentation unless such modifications, additions, or on this site, the Documentation, Software, or hardware provided by Avaya. All
deletions were performed by Avaya. End User agrees to indemnify and hold content on this site, the documentation and the Product provided by Avaya
harmless Avaya, Avaya's agents, servants and employees against all claims, including the selection, arrangement and design of the content is owned either
lawsuits, demands and judgments arising out of, or in connection with, by Avaya or its licensors and is protected by copyright and other intellectual
subsequent modifications, additions or deletions to this documentation, to the property laws including the sui generis rights relating to the protection of
extent made by End User. databases. You may not modify, copy, reproduce, republish, upload, post,
transmit or distribute in any way any content, in whole or in part, including any
Link disclaimer code and software unless expressly authorized by Avaya. Unauthorized
Avaya is not responsible for the contents or reliability of any linked websites reproduction, transmission, dissemination, storage, and or use without the
referenced within this site or documentation provided by Avaya. Avaya is not express written consent of Avaya can be a criminal, as well as a civil offense
responsible for the accuracy of any information, statement or content provided under the applicable law.
on these sites and does not necessarily endorse the products, services, or
Third-party components
information described or offered within them. Avaya does not guarantee that
these links will work all the time and has no control over the availability of the “Third Party Components” mean certain software programs or portions thereof
linked pages. included in the Software that may contain software (including open source
software) distributed under third party agreements (“Third Party Components”),
Warranty which contain terms regarding the rights to use certain portions of the Software
Avaya provides a limited warranty on its hardware and Software (“Product(s)”). (“Third Party Terms”). Information regarding distributed Linux OS source code
Refer to your sales agreement to establish the terms of the limited warranty. In (for those Products that have distributed Linux OS source code) and identifying
addition, Avaya’s standard warranty language, as well as information regarding the copyright holders of the Third Party Components and the Third Party Terms
support for this Product while under warranty is available to Avaya customers that apply is available in the Documentation or on Avaya’s website at:
and other parties through the Avaya Support website: http://support.avaya.com/ThirdPartyLicense/. You agree to the Third Party
http://www.avaya.com/support Terms for any such Third Party Components.
Please note that if you acquired the Product(s) from an authorized Avaya Preventing Toll Fraud
reseller outside of the United States and Canada, the warranty is provided to
“Toll Fraud” is the unauthorized use of your telecommunications system by an
you by said Avaya reseller and not by Avaya. “Software” means computer
unauthorized party (for example, a person who is not a corporate employee,
programs in object code, provided by Avaya or an Avaya Channel Partner,
agent, subcontractor, or is not working on your company's behalf). Be aware
whether as stand-alone products or pre-installed on hardware products, and
that there can be a risk of Toll Fraud associated with your system and that, if
any upgrades, updates, bug fixes, or modified versions thereto.
Toll Fraud occurs, it can result in substantial additional charges for your
Licenses telecommunications services.
THE SOFTWARE LICENSE TERMS AVAILABLE ON THE AVAYA WEBSITE, Avaya Toll Fraud intervention
http://support.avaya.com/LicenseInfo/ ARE APPLICABLE TO ANYONE WHO
If you suspect that you are being victimized by Toll Fraud and you need
DOWNLOADS, USES AND/OR INSTALLS AVAYA SOFTWARE,
technical assistance or support, call Technical Service Center Toll Fraud
PURCHASED FROM AVAYA INC., ANY AVAYA AFFILIATE, OR AN
Intervention Hotline at +1-800-643-2353 for the United States and Canada. For
AUTHORIZED AVAYA RESELLER (AS APPLICABLE) UNDER A
additional support telephone numbers, see the Avaya Support website:
COMMERCIAL AGREEMENT WITH AVAYA OR AN AUTHORIZED AVAYA
http://www.avaya.com/support.Suspected security vulnerabilities with Avaya
RESELLER. UNLESS OTHERWISE AGREED TO BY AVAYA IN WRITING,
products should be reported to Avaya by sending mail to:
AVAYA DOES NOT EXTEND THIS LICENSE IF THE SOFTWARE WAS
securityalerts@avaya.com.
OBTAINED FROM ANYONE OTHER THAN AVAYA, AN AVAYA AFFILIATE
OR AN AVAYA AUTHORIZED RESELLER; AVAYA RESERVES THE RIGHT Trademarks
TO TAKE LEGAL ACTION AGAINST YOU AND ANYONE ELSE USING OR
The trademarks, logos and service marks (“Marks”) displayed in this site, the
SELLING THE SOFTWARE WITHOUT A LICENSE. BY INSTALLING,
Documentation and Product(s) provided by Avaya are the registered or
DOWNLOADING OR USING THE SOFTWARE, OR AUTHORIZING OTHERS
unregistered Marks of Avaya, its affiliates, or other third parties. Users are not
TO DO SO, YOU, ON BEHALF OF YOURSELF AND THE ENTITY FOR
permitted to use such Marks without prior written consent from Avaya or such
WHOM YOU ARE INSTALLING, DOWNLOADING OR USING THE
third party which may own the Mark. Nothing contained in this site, the
SOFTWARE (HEREINAFTER REFERRED TO INTERCHANGEABLY AS
Documentation and Product(s) should be construed as granting, by implication,
“YOU” AND “END USER”), AGREE TO THESE TERMS AND CONDITIONS
estoppel, or otherwise, any license or right in and to the Marks without the
AND CREATE A BINDING CONTRACT BETWEEN YOU AND AVAYA INC. OR
express written permission of Avaya or the applicable third party.
THE APPLICABLE AVAYA AFFILIATE ( “AVAYA”).
Avaya and the Avaya logo, Avaya Contact Flow Analytics, Avaya Aura
Avaya grants you a license within the scope of the license types described
Performance Center, IQ, CMS, Application Enablement Services, Avaya Aura
below, with the exception of Heritage Nortel Software, for which the scope of
Performance Analytics are either registered trademarks or trademarks of
the license is detailed below. Where the order documentation does not
Avaya Inc. in the United States of America and/or other jurisdictions.
expressly identify a license type, the applicable license will be a Designated
System License. The applicable number of licenses and units of capacity for All non-Avaya trademarks are the property of their respective owners, and
which the license is granted will be one (1), unless a different number of “Linux” is a registered trademark of Linus Torvalds.
licenses or units of capacity is specified in the documentation or other materials
Downloading documents
available to you. “Designated Processor” means a single stand-alone
computing device. “Server” means a Designated Processor that hosts a For the most current versions of documentation, see the Avaya Support
software application to be accessed by multiple users. website:
http://www.avaya.com/support
License type(s)
Database License (DL). End User may install and use each copy of the Contact Avaya Support
Software on one Server or on multiple Servers provided that each of the See the Avaya Support website: http://support.avaya.com for product notices
Servers on which the Software is installed communicates with no more than a and articles, or to report a problem with your Avaya product.
single instance of the same database. For a list of support telephone numbers and contact addresses, go to the
Avaya Support website: http://support.avaya.com, scroll to the bottom of the
Heritage Nortel Software
page, and select Contact Avaya Support.
“Heritage Nortel Software” means the software that was acquired by Avaya as
 Contents

Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
About this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
About Speech Analytics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Intended audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Related documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Chapter 2: Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
About Speech Analytics reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Functional aspects of Speech Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Speech Analytics Data Administration Console . . . . . . . . . . . . . . . . . . . . . . 8
Speech Analytics reporting terminology. . . . . . . . . . . . . . . . . . . . . . . . . . 8
Dashboards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Chapter 3: Dashboards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Driver Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Agent Compare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Outcome/Agent Disposition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Ranking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Agent Compare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Agent Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Policy Adherence Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Ranking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Agent Compare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Agent Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Levels by Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Feedback Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Ranking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Agent Compare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Agent Focus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Speech Analytics Reporting March 2013 1

 Levels by Item . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Information required for support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Support limitations and restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Remote access for technical support . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Support contacts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Contacting Avaya Technical Support . . . . . . . . . . . . . . . . . . . . . . . . . 37
Contacting Avaya Professional Services . . . . . . . . . . . . . . . . . . . . . . . 38
Escalating a technical support issue. . . . . . . . . . . . . . . . . . . . . . . . . . 38

Speech Analytics Reporting March 2013 2

Speech Analytics Reporting March 2013 3
4 Speech Analytics Reporting March 2013
Chapter 1: Introduction

About this document

Avaya Speech Analytics is part of the Avaya Aura ® Performance Analytics suite of softwares.All
softwares from the Avaya Aura ® Performance Analytics suite function in the Avaya Aura ®
Performance Center environment. This document explains the various Speech Analytics
reporting features and measures and provides information on how to use the reporting features.

About Speech Analytics

Avaya Speech Analytics is a speech search and analytics solution based on phonetic
technology. Data analysts and executives use Speech Analytics to rapidly and efficiently
analyze sets of call recordings for quality monitoring, compliance and business intelligence.
Unlike competitive solutions, Speech Analytics is based on phonetic technology and provides
speaker independent search and analytics that is not limited to a pre-defined dictionary of
keywords or terms, providing greater flexibility, significantly increased performance and reduced
total cost of ownership.

Intended audience
This document is meant for contact center staff who perform analytics and quality improvement
roles, such as the following:
● Business Analysts
● Contact Center Business Intelligence or Business Reporting members
● Line Managers

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 5

Chapter 1: Introduction

Related documents
This document is part of the Speech Analytics documentation set that also includes the
following documents:
● Administering Speech Analytics
● Implementing Speech Analytics

6 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

Chapter 2: Overview

About Speech Analytics reporting

Speech Analytics is an application using which you can capture information from the
voice-based channels that the customer uses to obtain information about your company and
your products. Using Speech Analytics, you can identify the reasons why customers call your
company and what causes customer dissatisfaction. With Speech Analytics, you can also
identify training opportunities to improve agent performance at your contact center.
The Speech Analytics phonetic audio search engine provides an overview of the entire contact
center operations. This search engine can handle much larger volumes of audio data than
speech-to-text systems that can only monitor a small sample of recorded calls. You can also
inspect the individual call recordings in detail and obtain valuable insights into individual agent
performance and compliance issues.
Because Speech Analytics reports indicate potential compliance issues, you can take proactive
action from the perspective of an agent and a customer. Supervisors can allocate resources
according to the skill sets of individual agents, resulting in improved service.
Speech Analytics includes four detailed dashboards that provide information necessary to
understand call center operations and performance. For more information, see Dashboards on
page 11.

Functional aspects of Speech Analytics

Speech Analytics operates in a two-fold manner. The two key components are the Speech
Analytics desktop client called Speech Analytics Desktop Client (SADC) and the Speech
Analytics server. SADC is deployed on the workstation of the reporting analyst reponsible for
listening to calls and tagging the calls. You can use the desktop client to create Speech
Analytics projects and perform the following functions:
● Create and modify projects.
● Generate media library.
● Apply tags to the media files.
● Search and view results.
● Filter data.
● Use tags to categorize media files.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 7

Chapter 2: Overview

The server consists of a webservice, a datamart, and a business intelligence component. You
must become proficient with SADC as the quality of reports that you can view in the form of
dashboard depends on accurate tagging and categorizing of call recordings. You can use
Speech Analytics Data Administration Console, the administrative console of Speech Analytics,
to create categories and perform other administrative tasks.
The SADC connector handles communication between Speech Analytics Desktop Client and
the Speech Analytics server. After the business intelligence server applies algorithms and
processes the data, you can view the information as dashboards and reports.

! Important:
Important: To generate accurate reports, ensure that all VDNs, Skills, and Agents on Avaya
Aura ® Communication Manager or the PBX are set to be measured.

Speech Analytics Data Administration Console

Speech Analytics Data Administration Console (SADAC) is the administrative interface that you
can use to administer and configure Speech Analytics. SADAC is a crucial aspect of Speech
Analytics, and you must have a thorough understanding of SADAC to make Speech Analytics
peform optimally. Using SADAC, you can manage:
● Agents
● Business categories
● Call components
● Call drivers
● Call outcomes
● Adherence profiles
● Customer feedback
● Non mineable reasons
● Teams
For more information on how to administer Speech Analytics, see Administering Speech
Analytics.

Speech Analytics reporting terminology

The following terms are frequently used in relation to Speech Analytics:

8 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Dashboards

● Driver: The main reason for the customer calling your contact center.
● Business category: Business impact areas grouped as categories.
● Call outcome: The result of the call from the perspective of a customer.
● Call component: Pre-defined sections of the calls linked to business profiles. For
example, greeting, security check, and closing.
● Adherence profile: A business process that consists of call components.
● Non-mineable calls: Calls that could not be tagged for reasons such as poor voice
quality.
● Feedback class: Groups of feedback items that might include agent evaluation.
● Business impact: The positive, netural, or negative impact that the call has on the
business.
● Feedback item: The specific item for which the customer provides feedback, such as
advertising or pricing.

Dashboards
Speech Analytics includes the following four dashboards that provide insights on the respective
subject areas:
● Driver Classification: Provides information on the drivers. Common drivers or the
reasons for the customer to call your contact center include queries, billing, sales, and
complaints. You can define the drivers based on your business scenario from SADAC.
● Outcome/Agent Disposition: Provides information on the overall result of the calls.
Commont outcomes include unsuccessful, successful, or pending depending on whether
the reason your customer called was addressed or not.
● Policy Adherence Monitoring: Provides information on whether your contact center
agents adhere to the defined policies and procedures. You can define the adherence
profiles based on your business scenario from SADAC.
● Feedback Assessment: Provides feedback from the customer-agent conversation. You
can assign calls according to a combination of feedback items, on the basis of which the
report shows the Promote rate, Passive rate, and Detract rate. Each of these categories
indicate the impression that the customer has and the likely future behavaior that the
customer will exhibit. You can define the business categories and the feedback class
based you on your business scenario from SADAC.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 9

Chapter 2: Overview

10 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

Chapter 3: Dashboards

Speech Analytics has four standard dashboards. The dashboards, which are a combination of
several reports, provide you with detailed information on Driver Classification, Outcome/Agent
Disposition, Policy Adherence, and Feedback Analysis. Each dashboard graphically represents
information using pie charts, bar graphs, tables, and pivot tables. Speech Analytics divides each
dashboard into tabs, which provide different perspectives on the selected report area. You can
export and print each report.
Note:
Note: Speech Analytics deletes data that more than an year old from the database
through an automatic purge job. If you want to retain data that is more than an
year old, disable the purge job from the ETL jobs.

Driver Classification
The driver classification dashboard provides information on the drivers, or the reasons for the
customer calling your contact center. Speech Analytics divides the dashboard into three tabs,
Overview, Trend, and Agent Compare. You can apply the following common filters to generate
the dashboard:
● Dataset: Identification of the specific imported data.
● Call time (AM/PM): The time of the call in terms of AM or PM.
● Direction: The direction of the call: inbound, outbound, or internal.
● Driver: The main reason for the customer calling your contact center.
● Call Date Time: Two dates between which you want to generate the dashboard.

Overview
The Overview tab provides a picture of why customers call your contact center. The Overview
tab first provides a snapshot of the crucial information and then uses a combination of pie charts
and pivot tables to go into the details. The overview tab contains the following information:
● Calls in Period: Shows the total number of calls that are present for analysis in the
specified date range.
● Classified Calls: Shows the number of calls that have a driver associated to the call in the
specified date range.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 11

Chapter 3: Dashboards

● Calls Under Analysis: Shows the number of calls in the specified date range that meet
the specified filter or prompted criteria.The graph presents information over a period of
time.
Note:
Note: If none driver is left in the prompted set, the number of Calls Under Analysis can
exceed Classified Calls.
● Non-mineable rate %: Shows the percentage of calls on which Speech Analytics could
not perform data mining. The report shows the reasons for the inability to mine the calls in
a separate color-coded pie chart and a pivot table.
- Non-mineable reason ratio: The pie chart represents the reasons for the inability to
mine calls as a ratio of the overall percentage of non-mineable calls.
- Non-mineable calls level: The bar graph shows the number of calls on which Speech
Analytics could not perform data mining in relation to the reasons for the inability to
mine the calls.
- Non-Mineable Call Levels and Ratio: Shows proportions and levels of calls
according to the reasons that the call could not be searched.
● Non-Mineable / Unclassified / Classified Ratio: The pie chart shows the proportion of
calls for classified calls, unclassified calls, or calls on which Speech Analytics could not
perform data mining.
Note:
Note: Total calls is the sum of non-mineable, classified, and unclassified calls.The
graph presents information over a period of time. So, mineable calls = classified
calls + unclassified calls.
● Unclassified rate %: Shows the percentage of calls that were without an identified driver
and hence could not be classified by driver. Speech Analytics presents the unclassified
rate % along with other relevant information in a pie chart.
● Classification rate %: Shows the proportion of all calls according to the primary driver for
the call. Speech Analytics presents the classification rate % along with other relevant
information in a pie chart.
- Driver Ratio: The pie chart shows the proportions of all calls according to the primary
driver for the calls.
- Driver Levels: The bar graph shows the number of classified calls in relation to the
call drivers.
- Driver Levels and Ratio: The pivot table shows the number of classified calls in
relation to the call drivers.

12 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Driver Classification

Trend
The Trend tab contains three graphs: two focus on the call drivers and one on the non-mineable
call trend. The graphs show proportions of calls over time according to the primary driver for the
call. The Time dimension is by Week in Week and Year format where the week runs Monday
through Sunday, with drill path to Call Date. Organization is by Company with drill path through
the organizational structure to Agent.
The graphs provide a visual view of the information. Using the graphs in the Trend tab, you can
quickly gain an understanding of the factors that drive customers to call your contact center. The
trend analysis graphs are especially useful when you have a large number of calls as part of the
report, and you cannot analyze individual calls. For example, by looking at the Driver Trend %
graph, you can notice any significant change in the number of complaints or decline in sales
calls in one glance. Based on your analysis, you can take action to reverse a negative trend or
boost a positive trend. You can also click on any week to view the trend for the selected week.

Agent Compare
The Agent Compare tab provides information on the ratio of calls that individual agents handle
in relation to call drivers. Using the information, you can ensure that the agents who are trained
in a particular skill are handling calls that best suit those skills. You can also get an idea about
the distribution pattern of calls between agents. The Agent Compare tab provides comparative
information through the following six columns:
● Center: Calls based on the physical location of the contact center.
● Team: Calls based on the teams within one center.
● Agent: Calls based on agents from one team.
● Driver: Calls based on the call driver for an individual agent.
● % Calls (Classified): Calls based on a set of call drivers for a set of agents.
● Calls (Classified): The number of calls per driver for an individual agent.

Measures
The following table describes the measures used for computing the Driver Classification
dashboard:

Table 1: Driver Classification measures

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 13

Chapter 3: Dashboards

Measure Measure Description

folder

% Classified (group by Measures The percentage of filtered calls that the

agent, driver) system can classify, totaled across all
drivers, and grouped by each agent.
% Classified (group by Measures The percentage of filtered calls that the
driver) system can classify, totaled across all
drivers.
% Classified (relative) Measures The percentage of filtered calls that the
system can classify, basically a straight
percentage of classified calls taken from all
calls without totaling across hierarchical
levels.
% Mineable (relative) Measures The percentage of filtered calls that the
system can mine, basically a straight
percentage of mineable calls taken from all
calls without totaling across hierarchical
levels.
% Non-Mineable Measures The percentage of filtered calls that the
(relative) system cannot mine, basically a straight
percentage of non-mineable calls taken from
all calls without totaling across hierarchical
levels.
% Unclassified (relative) Measures The percentage of filtered calls without a
driver that the system cannot classify,
basically a straight percentage of
unclassified calls taken from all calls without
totaling across hierarchical levels.
All Calls Measures All calls that the system captures in the data
mart. Typically, the system represents the All
Calls measure in reports as Calls in Period,
where the system applies the date prompt.
Classified Calls Measures Mineable calls that have a primary driver.
Cloud Based Analytics (CBA) supports only
a single, primary driver. Note that
Non-Mineable calls + Classified calls +
Unclassified calls = All Calls.
Classified Calls (agent) Measures Mineable calls that contain a driver, summed
across an agent. For example, if a call could
be further filtered by Agent by Driver, the
classified calls in this case will be across all
drivers.

14 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Driver Classification

Measure Measure Description

folder

Classified Calls (agent, Measures Mineable calls that contain a driver, summed
drivers totals) across all drivers and grouped by each
agent. For example, this measure will be all
classified calls, regardless of driver, for each
agent.
Classified Calls (center, Measures Mineable calls that contain a driver, summed
drivers totals) across all drivers and grouped by each
center. For example, this measure will be all
classified calls, regardless of driver, for each
center.
Classified Calls (drivers Measures Mineable calls that contain a driver, summed
totals) across all drivers.
Classified Calls (group Measures Mineable calls that contain a driver, summed
by company, week) across all drivers and grouped by each
company. For example, this measure will be
all classified calls, regardless of driver, for
each company.
Classified Calls (group Measures Mineable calls that contain a driver, summed
by Team) across all drivers and grouped by each
team. For example, this measure will be all
classified calls, regardless of driver, for each
team.
Mineable Calls Measures Calls that the system can mine for additional
analysis. A call that is mineable might be
classified with a driver. Any given call is only
counted once when the system categorizes
the call as either non-mineable, unclassified,
or classified. Note that Non-Mineable calls +
Classified calls + Unclassified calls = All
Calls.
Non-Mineable Calls Measures Calls that the system cannot mine for
additional analysis. A call that is
non-mineable includes the reason the call is
not mineable, for example, the call was too
short, poor speech quality, not handled by
an agent, or no speech to analyze. Any
given call is only counted once when the call
is categorized as either non-mineable,
unclassified, or classified.
Total Non-Mineable Measures Non-mineable calls totalled across all
Calls relevant dimensions.
Unclassified Calls Measures Mineable calls that do not contain a driver.
Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 15

Chapter 3: Dashboards

Outcome/Agent Disposition
The Outcome/Agent Disposition dashboard provides information on the outcome of a call
according to the impact the call has on your business. One of the key measures for any contact
center is customer satisfaction, that is, ability to deliver the service that the customer expects
from your contact center. A customer might call your contact center for various reasons. The
outcome of a call is crucial in ensuring customer satisfaction, promoting your brand, and
minimizing repeat calls. You can classify calls as Successful, Pending, or Unsuccessful.
Outcomes can vary from the customer taking an immediate decision to buy a product to an
agent successfully handling a general billing query. Drivers are crucial in understanding the
outcome of calls. For example, ensuring that a call to make a complaint is Successful is
important. If the outcome of a large number of complaint calls is Pending or Unsuccessful, you
must investigate the reasons for the inability of agents to resolve customer complaints. Speech
Analytics divides the Outcome/Agent Disposition dashboard into five tabs: Overview, Trend,
Ranking, Agent Compare, and Agent Focus. You can apply the following filters to generate the
dashboard:
● Dataset: Identification of the specific imported data.
● Call time (AM/PM): The time of the call in terms of AM or PM.
● Direction: The direction of the call: inbound, outbound, or internal.
● Driver: The main reason for the customer calling your contact center.
● Outcome Class: The call outcome that you define using SADAC.

Overview
The Overview tab provides the outcome levels of the calls that your contact center agents
handle. The Overview tab first provides a snapshot of the crucial information about the outcome
or agent disposition of calls and uses a combination of pie charts, tables, and pivot tables to go
into the details. The overview tab contains the following information:
● Calls in Period: The total number of calls that are present for analysis in the specified
date range.
● Calls With Outcome: The number of calls that have a call outcome associated to the call
in the specified date range.
● Calls Under Analysis: The number of calls that are included as part of the analysis in the
specified date range.
● Successful Rate: The percentage of calls that the agents resolved successfully. Speech
Analytics reports the success ratio, success level by outcome class, and success levels in
a separate color-coded pie chart, a bar graph, and a table.

16 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Outcome/Agent Disposition

- Outcome Ratio: The pie chart shows the proportions of all Outcomes according to
their impact on the business.
- Outcome Ratio by Driver: The bar graph shows the proportions of all Outcomes in
relation to call drivers.
- Outcome Levels by Outcome Class: The bar graph shows the level of calls
according to the impact on the business of their associated Outcomes.
- Outcome Levels by Outcome: The table shows the percentage and the number of
calls in relation to the outcome, classified as the following:
● Company
● Outcome Class
● Call Outcome
● Impact on business: Positive, Neutral, and Negative
● Percentage of calls under each of the three impact categories
● Pending Rate: Shows the percentage of calls that the agents could not resolve for a given
driver. Pending outcomes are for calls that had at least one neutral outcome. Pending calls
include all instances where a future action is pending on behalf of the agent or the
customer.
● Unsuccessful Rate: Shows the percentage of calls that were unsuccessful from the point
of view of the customer for the selected drivers.

Trend
The Trend tab contains a color-coded graph, which focuses on the outcome of all the calls
under analysis. The graphs provide a visual view of the information. The graph presents
information over a period of time representing a certain trend or behavior pattern. For example,
the trend report might include a period of one month, during which your company had initiated a
promotional exercise. From the Trend graph, you can understand if the promotional exercise
made a difference to the Outcome/Agent Disposition trend, such as sales. The trend analysis
graphs are especially useful when you have a large number of calls as part of the report, and
you cannot analyze individual calls. For example, by looking at the Success Trend graph, you
can discern a significant change in the outcome of calls at a glance. Based on your analysis,
you can take action to reverse a negative trend or boost a positive trend.

Ranking
The Ranking tab contains the Outcome Levels (Ranked) table. The table shows the percentage
and the number of calls in relation to the outcome classified by the following:
● Company

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 17

Chapter 3: Dashboards

● Outcome Class
● Call Outcome: Positive, Neutral, or Negative
● Percentage of calls under each of the call outcome categories
Speech Analytics ranks the metrics with the largest share of positive calls first and the largest
share of negative calls last.

Agent Compare
The Agent Compare tab provides information on the outcome of calls in relation to individual
agents who handle the calls. Using the information, you can gain insights on the performance of
individual agents. You can use the metrics to ensure that you retain agents with a high success
rate in the assigned role whereas agents who do not perform up to the mark are trained further
or shifted to a different role. Apart from the filters described in the Outcome/Agent Disposition
description, you can apply the Team filter to narrow your report results. The Agent Compare tab
provides two analytical tables, Agent Overall Impact of Outcome and Agent Outcome by
Summary. The Agent Overall Impact of Outcome table uses the following six columns to group
information:
● Center: Calls based on the physical location of the contact center.
● Team: Calls based on the teams within one center.
● Agent: Calls based on an individual agent from one team.
● Unsuccessful rate (%): The percentage of unsuccessful calls for an individual agent.
● Successful rate (%): The percentage of successful calls for an individual agent.
● Pending rate (%): The percentage of pending calls for an individual agent.
The Agent Outcome by Summary table uses the following six columns to group information:
● Center: Calls based on the physical location of the contact center.
● Team: Calls based on the teams within one center.
● Agent: Calls based on an individual agent from one team.
● Call outcome:
- Impact On Business - Negative: The percentage and number of negative calls based
on the outcome of the call.
- Impact On Business - Neutral: The percentage and number of neutral calls based on
the outcome of the call.
- Impact On Business - Positive: The percentage and number of positive calls based
on the outcome of the call.

18 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Outcome/Agent Disposition

Agent Focus
The Agent Focus tab offers a different view of the information present in the Trend tab and the
Agent Compare tab. Apart from the filters described in the Outcome/Agent Disposition
description, you can apply the Agent filter to narrow your report results. Using the filters, you
can view information for an agent in relation to the outcome of the call. With the trend graph,
you can identify any significant changes in the outcome trend for a given agent.

Measures
The following table describes the measures used for computing the Outcome/Agent Disposition
dashboard:

Table 2: Outcome/Agent Disposition measures

Measure Measures Description

folder

All Calls Overview The number of all calls that occurred in the
Measures time period selected for a dashboard. This
measure is called Calls in Period in the
report.
Calls With Outcome Measures The number of classified calls that have an
outcome. Each call can only have one
outcome.
Classified Calls Measures Mineable calls that have a primary driver.
CBA supports only a single, primary driver.
Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.
Classified Calls (agent, Measures Classified calls summed across all outcome
outcome totals) classes and grouped by each agent. For
example, this measure will be the total of all
classified calls, regardless of outcome, for
each agent.
Classified Calls Measures Classified calls, summed across all outcome
(outcome totals) classes. For example, this measure will be
the total of all classified calls, regardless of
outcome, useful in denominators of
percentage calculations.
Pending Calls Measures The number of classified calls with a neutral
outcome, indicating the agent could not
resolve the call during that interaction.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 19

Chapter 3: Dashboards

Measure Measures Description

folder

Pending Rate (group by Measures The percentage of calls with neutral outcome
agent) from the level-based total set of outcomes
grouped by agent. This calculation is a
call-level calculation.
Pending Rate (relative) Measures The percentage of calls with neutral
outcome. This calculation is a call-level
calculation.
Pending Rate (totals) Measures The percentage of calls with neutral outcome
from the level-based total set of outcomes.
This calculation is a call-level calculation.
Successful Calls Measures The number of classified calls with a positive
outcome, indicating the agent could
successfully resolvethe call during that
interaction.
Successful Rate (group Measures The percentage of calls with positive
by agent) outcome from the level-based total set of
outcomes grouped by agent. This calculation
is a call-level calculation.
Successful Rate Measures The percentage of calls with positive
(relative) outcome. This calculation is a call-level
calculation.
Successful Rate (totals) Measures The percentage of calls with positive
outcome from the level-based total set of
outcomes. This calculation is a call-level
calculation.
Unclassified Calls Measures Mineable calls that do not contain a driver.
Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.
Unsuccessful Calls Measures The number of classified calls with a
negative outcome, indicating the agent could
not resolve the call during that interaction.
Unsuccessful Rate Measures The percentage of calls with negative
(group by agent) outcome from the level-based total set of
outcomes grouped by agent. This calculation
is a call-level calculation.

20 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Policy Adherence Monitoring

Measure Measures Description

folder

Unsuccessful Rate Measures The percentage of calls with negative

(relative) outcome. This calculation is a call-level
calculation.
Unsuccessful Rate Measures The percentage of calls with negative
(totals) outcome from the level-based total set of
outcomes. This calculation is a call-level
calculation.

Policy Adherence Monitoring

The Policy Adherence Monitoring dashboard provides complete information on the compliance
and policy adherence levels by agents during calls. Different organizations have different
standards to comply with, ranging from legal reasons to business reasons. You can use SADAC
to define policy adherence profiles that your agents follow for different calls. Using the Policy
Adherence Monitoring dashboard, you can view reports that provide you with the adherence or
breach level details along with associated factors such as profile, call, and component. You can
apply the following filters to generate the dashboard:
● Dataset: Identification of the specific imported data.
● Call time (AM/PM): The time of the call in terms of AM or PM.
● Direction: The direction of the call: inbound, outbound, or internal.
● Driver: The main reason for the customer calling your contact center.
● Profile: The Profile that you define using SADAC.
● Call Dates: Two dates between which you want to generate the report.
The Policy Adherence Monitoring dashboard contains six tabs: Overview, Trend, Ranking,
Agent Compare, Agent Focus, and Levels by Profile. The reports show the proportions of all
expected call Components according to adherence to policy or breach of policy, where a call
might have multiple Profiles each containing multiple Components. A call can have multiple
Components across the profiles that might or might not exist, but the report counts the
Components and not the call-level summary such as fully adherent, fully in breach, or partially
adherent. Since Speech Analytics only counts components once, metrics can be calculated for
adherence and breach that are discrete and non-overlapping, as a proportion of 100%.
Organization is by Company* with drill-path through organizational structure to Agent.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 21

Chapter 3: Dashboards

Overview
The Overview tab provides the policy adherence metrics for your contact center. The report
uses a combination of pie charts, bar graphs, and pivot tables to present the metrics. The
Overview tab contains the following information:
● Calls in Period: The total number of calls that are present for analysis in the specified
date range.
● Calls With Adherence: The number of calls that have a profile associated to the call in the
specified date range.
● Calls Under Analysis: The number of calls that are included as part of the analysis in the
specified date range.
● Adherence Rate %: The percentage of calls that conform with the policy standards.
● Policy Breach Rate %: The percentage of calls that do not conform with the policy
standards.
● Adherence Ratio: The color-coded pie chart shows the proportions of all expected call
components according to presence, which means adherence to policy or absence, which
means breach of policy.
● Adherence Ratio By Driver: The graph shows the proportions of all expected call
components according to presence that is, adherence to policy or breach of policy, in
relation to the call driver. Using this graph, you can identify the type of calls, based on the
driver, which often result in policy breaches.
● Adherence Levels By Component: The table shows the adherence levels of calls based
on the following:
- Company
- Profile
- Component
- Calls (Adhere)
- Calls (Breach)
- % Calls (Adhere)
- % Calls (Breach)

22 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Policy Adherence Monitoring

Trend
The Trend tab contains a color-coded graph, which focuses on policy adherence and breach
percentage of all the calls under analysis for a selected set of weeks. The graphs provide a
visual view of the information in the Overview tab. Using the graph in the Trend tab, you can get
a snapshot of the policy adherence and breach level of the calls that the contact center agents
handle over a period of time. The trend analysis graphs are especially useful when you have a
large number of calls as part of the report, and you cannot analyze individual calls. For
example, by looking at the Adherence Trend graph, you can notice any significant change in the
adherence level of calls in one glance. Based on your analysis, you can take action to reverse a
negative trend or boost a positive trend.

Ranking
The Ranking tab contains the Adherence Levels table, which shows the adherence levels of
calls based on the following:
- Company
- Profile
- Component
- Calls (Adhere)
- Calls (Breach)
- % Calls (Adhere)
- % Calls (Breach)
Using the Ranking tab, you can analyze the policy adherence levels by ranking, which means
you can identify the type of calls that most or least often result in policy breaches. Speech
Analytics ranks the metrics with the largest share of calls that adhere to the policy first and the
largest share of calls that breach the policy last.

Agent Compare
The Agent Compare tab provides information on the adherence level of calls in relation to
individual agents who handle the calls. Using the report, you can gain insights on the
performance of individual agents. You can use the metrics to ensure that the agents who have a
low adherence rate are trained appropriately. The Agent Compare tab provides two analytical
tables, Agent Overall Balance of Adherence and Agent By Component Summary. The Agent
Overall Balance of Adherence table uses the following six columns to group information:
● Center: Groups calls based on the physical location of the contact center.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 23

Chapter 3: Dashboards

● Team: Groups calls based on the teams within one center.

● Agent: Groups calls based on an individual agent from one team.
● Policy Breach Rate (%): Shows the percentage of calls that breach policy guidelines for
an individual agent.
● Adherence Rate (%): Shows the percentage of calls that adhere to policy guidelines for
an individual agent.
The Agent By Component Summary table uses the following six columns to group information:
● Center: Groups calls based on the physical location of the contact center.
● Team: Groups calls based on the teams within one center.
● Agent: Groups calls based on an individual agent from one team.
● Component: Groups calls based on a call component, such as Greeting or Closing, for an
individual agent.
● Policy Breach Rate (%): Shows the percentage of calls that breach policy guidelines for
an individual agent in relation to a component.
● Adherence Rate (%): Shows the percentage of calls that adhere to policy guidelines for
an individual agent in relation to a component.

Agent Focus
The Agent Focus tab offers a different view to look at information that is present in the Trend tab
and the Agent Compare tab. The Agent Focus tab also contains the call recordings of the calls
under analysis. You can listen to the call recordings if you cut and paste the recording into a
player. With the recordings, you can understand the reasons and the severity of the policy
breach. The Agent Focus tab offers the following additional options to filter the data:
● Dataset
● Call time (AM/PM)
● Direction
● Driver
● Outcome Class
● Call Outcome
● Agent
Using the listed filters, you can view information for an agent in relation to the adherence levels
of the calls. With the Adherence Trend graph, you can identify any significant changes in the
adherence level for a given week.

24 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Policy Adherence Monitoring

Levels by Profile
The Levels by Profile tab offers a dedicated view to look at policy adherence and policy breach
metrics for a selected profile. You might want to monitor certain profiles, for example, profiles
that impact customer experience, more than other profiles. The information provided in the
Levels of Profile tab makes it simpler to monitor adherence and breach metrics. The Levels by
Profile tab offers the following options to filter the data:
● Dataset
● Call time (AM/PM)
● Direction
● Driver
● Profile
● Call Date range
The Adherence Levels by Profile graph shows the levels of calls according to either full
adherence to, full breach of, or partial adherence to business policy.

Measures
The following table describes the measures used for computing the Policy Adherence
Monitoring dashboards:

Table 3: Policy Adherence measures

Measure Measures Description

folder

All Calls Overview The number of all calls that occurred in the
Measures time period selected for a dashboard. This
measure is called Calls in Period in the
report.
Classified Calls Overview Mineable calls that have a primary driver.
Measures CBA supports only a single, primary driver.
Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.
Unclassified Calls Overview Mineable calls that do not contain a driver.
Measures Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 25

Chapter 3: Dashboards

Measure Measures Description

folder

% Calls (Fully Adhere to Measures by The percentage of calls that fully adhere to
Policy) Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
fully adherent call contains only components
with positive impact on business.
% Calls (Fully in Breach Measures by The percentage of calls that fully breach
of Policy) Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
fully breached call contains only components
with negative impact on business.
% Calls (Partially Measures by The percentage of calls that partially breach
Adhere to Policy) Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
call partially in breach contains a mixture of
components with positive and negative
impact on business.
Call Date Time Measures by The timestamp of the start of the call.
(CallAdherence) Call
Calls With Components Measures by The number of calls that contain at least one
Call component.
Fully Adherent Calls Measures by The number of calls that fully adhere to
Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
fully adherent call contains only components
with positive impact on business.
Fully Breach Calls Measures by The number of calls that fully adhere to
Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
fully adherent call contains only components
with positive impact on business.
Partially Adherent Calls Measures by The number of calls that partially breach
Call business policy. Since a call may reference
multiple profiles (business policy), each
containing multiple rules (components), a
call partially in breach contains a mixture of
components with positive and negative
impact on business.
Call Recording URL Measures by The URL that you can use to reach the call
Call recording file.

26 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Policy Adherence Monitoring

Measure Measures Description

folder

Call Recording Measures by The file name of the audio recording of the
Call call.
Call Date Time (Call Measures by The timestamp of the start of the call.
Adherence Profile) Call by Profile
Fully Adherent Call Measures by The number of call by profile records whose
Parts Call by Profile associated components are all positive.
Fully Breach Call Parts Measures by The number of call by profile records whose
Call by Profile associated components are all negative.
Partially Adherent Call Measures by The number of call by profile records whose
Parts Call by Profile associated components are both positive
and negative.
Total Call Parts Measures by The number of call by profile records,
Call by Profile basically a count of the intersection of calls
with profiles.
Adherence Rate (%) Measures by The percentage of calls with positive
Driver components, accessible to grouping by
driver.
Call Date Time (Driver Measures by The timestamp of the start of the call.
Adherence) Driver
Calls by Driver Measures by The number of classified calls.
Driver
Negative Calls Measures by The number of calls with a negative
Driver component.
Policy Breach Rate (%) Measures by The percentage of calls with negative
Driver components, accessible to grouping by
driver.
Positive Calls Measures by The number of calls with a positive
Driver component.
Positive Component Measures by Indicates whether or not a given component
Indicator Components is a positive component.
Negative Component Measures by Indicates whether or not a given component
Indicator Components is a negative component.
Calls with Components Measures by The number of calls that contain at least one
Components component.
Negative Components Measures by The number of negative components.
Components

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 27

Chapter 3: Dashboards

Measure Measures Description

folder

Negatives by Call Measures by The number of calls that contain at least one
Components negative component.
Positive Components Measures by The number of calls that contain at least one
Components positive component.
Positives by Call Measures by The number of calls that contain at least one
Components positive component.
Total Components Measures by The number of all components, independent
Components of business impact.
Adherence Rate (%) Measures by The percentage of components that adhere
Components to policy. This is not a call-level measure.
Policy Breach Rate (%) Measures by The percentage of components that breach
Components policy. This is not a call-level measure.

Feedback Assessment
The Feedback Assessment dashboard provides information on what customers think about any
of the pre-defined feedback criteria on which you solicit customer feedback. You can use the
SADAC to define items on which you want customer feedback, such as product, agent
interaction, or pricing. Each feedback item is linked to a business category and has a positive,
neutral, or negative impact on your business. For example, for pricing, if the customer feedback
is high, the feedback will have a negative impact on your business. The Feedback Assessment
reports classify the impact as promote, passive, or detract.
The report shows the proportions of all feedback items according to their impact on the
business. A call can have multiple items of feedback, so the report counts the individual
feedback items and not the call-level summary. Since Speech Analytics counts feedback items
individually, metrics can be calculated for positive, negative, and neutral rates that are discrete
and non-overlapping, as a proportion of 100%. Organization is by Company* with drill-path
through organizational structure to Agent. You can apply the following filters to generate the
dashboard:
● Dataset: Identification of the specific imported data.
● Call time (AM/PM): The time of the call in terms of AM or PM.
● Direction: The direction of the call: inbound, outbound, or internal.
● Driver: The main reason for the customer calling your contact center.
● Feedback Class: The Feedback Class that you define using SADAC.

28 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Feedback Assessment

● Call Date: Two dates between which the system generated the dashboard.
The Feedback Assessment dashboard contains six tabs, Overview, Trend, Ranking, Agent
Compare, Agent Focus, and Levels by Profile.

Overview
The Overview tab provides information about what your customers think about you and your
organization. The report uses a combination of pie charts, bar graphs, and pivot tables to
present the metrics. The Overview tab contains the following information:
● Calls in Period: Shows the total number of calls that are present for analysis in the
specified date range.
● Calls With Feedback: Shows the number of calls that have a feedback associated to the
call in the specified date range.
● Calls Under Analysis: Shows the actual number of calls that are included as part of the
analysis in the specified date range.
● Promote Rate %: The percentage of all feedback that the customers rated favorably
across the various feedback items.
● Passive Rate %: The percentage of all feedback that the customers rated neutrally across
the various feedback items.
● Detract Rate %: The percentage of all feedback that the customers rated negatively
across the various feedback items.
● Feedback Ratio: The pie chart shows the Detract rate, Passive rate, and Promote rate in
percentage of calls according to the customer feedback.
● Feedback Ratio By Driver: The graphs shows the Detract rate, Passive rate, and
Promote rate in percentage according to the customer feedback in relation to call drivers.
● Feedback Levels By Class: The graphs shows the Detract rate, Passive rate, and
Promote rate of calls in percentage according to the customer feedback in relation to
feedback class.

Trend
The Trend tab contains a color-coded Feedback Trend graph, which focuses on feedback levels
for all the calls under analysis for a given week. The graphs provide a visual view of the
information in the Overview tab. Using the graph in the Trend tab, you can get a snapshot of the
feedback level of the calls that the contact center agents handle over a period of time. The trend
analysis graphs are especially useful when you have a large number of calls as part of the

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 29

Chapter 3: Dashboards

report, and you cannot analyze individual calls. For example, by looking at the Feedback Trend
graph, you can notice any significant change in the feedback class of calls in one glance. Based
on your analysis, you can take action to reverse a negative trend or boost a positive trend. The
Feedback Trend graph is also very useful when you are expecting a significant change in
feedback due to some other factors such as changed pricing or improved products.

Ranking
The Ranking tab contains the Feedback Levels (Ranked) table, which shows the type of
feedback received from your customer, based on the following:
- Company
- Feedback Class
- Feedback Item
- Number and percentage of negative calls
- Number and percentage of neutral calls
- Number and percentage of positive calls
Speech Analytics ranks the metrics with the feedback item with the largest share of positive
calls first and the feedback item with the largest share of negative calls last.

Agent Compare
The Agent Compare tab provides information on the feedback type of calls in relation to
individual agents who handle the calls. Using the reported information, you can gain insights on
the performance of individual agents. You can use to metrics to ensure that the agents who
enjoy a high positive feedback rate are retained in their roles where as agents who frequently
encounter a neutral or negative feedback rate are trained and that you analyze the reasons for
the negative feedback. The Agent Compare tab provides two analytical tables, Agent Overall
Impact of Feedback and Agent By Feedback Item Summary. The Agent Overall Impact of
Feedback table uses the following six columns to group information:
● Center: Calls based on the physical location of the contact center.
● Team: Calls based on the teams within one center.
● Agent: Calls based on an individual agent from one team.
● Detract Rate (%): The percentage of total feedback items that received a negative
feedback from the customer, for an individual agent.
● Promote Rate (%): The percentage of total feedback items that received a positive
feedback from the customer, for an individual agent.

30 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Feedback Assessment

● Passive Rate (%): The percentage of total feedback items that received a neutral
feedback from the customer, for an individual agent.
The Agent By Feedback Item Summary table uses the following six columns to group
information:
● Center: Calls based on the physical location of the contact center.
● Team: Calls based on the teams within one center.
● Agent: Calls based on an individual agent from one team.
● Feedback Item: Calls based on a call component, such as Greeting or Closing, for an
individual agent.
● Impact on business - Negative: The percentage and number of calls that had a negative
impact on the business in relation to a feedback item.
● Impact on business - Neutral: The percentage and number of calls that had a neutral
impact on the business in relation to a feedback item.
● Impact on business - Positive: The percentage and number of calls that had a positive
impact on the business in relation to a feedback item.

Agent Focus
The Agent Focus tab offers a different view of information from the Trend tab and the Agent
Compare tab. The Agent Focus tab offers additional options to filter data by:
● Dataset
● Call time (AM/PM)
● Direction
● Driver
● Feedback Class
● Feedback Item
● Call Dates
● Agent
Using the listed filters, you can easily view information for an agent in relation to the feedback of
the calls. With the trend graph, you can identify any significant changes in the feedback trend
for a given week.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 31

Chapter 3: Dashboards

Levels by Item
The Levels by Item tab offers a dedicated view to look at feedback levels, which includes
promote rate, passive rate, and detract rate, for a selected feedback item. You might want to
monitor certain feedback classes, for example, classes that impact competition or pricing, more
than other feedback classes. The information provided in the Levels of Item tab makes it simpler
to monitor feedback related metrics and their impact on the business. The Levels by Item tab
offers the following options to filter the data:
● Dataset
● Call time (AM/PM)
● Direction
● Driver
● Feedback Class
● Call Date range
The Feedback Levels by Item graph shows the levels of calls according to the impact on the
business, which might be positive, neutral, or negative.

Measures
The following table describes the measures used for computing the Feedback Assessment
dashboard:

Table 4: Feedback Assessment measures

Measure Measures Description

folder

All Calls Overview The number of all calls that occurred in the
measures time period selected for a dashboard. In the
reports, this measure is named, Calls in
Period.
Classified Calls Overview Mineable calls that have a primary driver.
measures CBA supports only a single, primary driver.
Note that Non-Mineable calls + Classified
calls + Unclassified calls = All Calls.
Call Recording Measures The file name of the audio recording of the
call.

32 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Feedback Assessment

Measure Measures Description

folder

Recording URL Measures The URL that can be used to access the call
recording file.
Calls with Feedback Measures The number of calls that have at least one
(relative) associated feedback item.
Calls With Feedback Measures The number of calls with at least one
(feedback totals) associated feedback item totaled/
level-based across all feedback classes.
The system uses this measure as the
denominator in percentage calculations
required to sum to 100% across all feedback
classes.
Calls With Feedback Measures The number of calls with at least one
(Feedback Class, Team) associated feedback item totaled/
level-based across all feedback classes.
The system uses this measure as the
denominator in percentage calculations
required to sum to 100% across all feedback
classes and grouped by team.
Calls With Feedback Measures The number of calls with at least one
(cwf - agent, feedback associated feedback item totaled/
totals) level-based across all feedback classes.
The system uses this measure as the
denominator in percentage calculations
required to sum to 100% across all feedback
classes and grouped by agent.
Detract Rate (%) Measures The percentage of negative feedback items
from the total set of feedback items. Since a
call can have multiple feedback items, this
measure is based on individual feedback
items and not the call-level summary.
Detract Rate (%) (cwf - Measures The percentage of negative feedback items
agent, feedback totals) from the level-based total set of feedback
items grouped by agent. Since a call can
have multiple feedback items, this measure
is based on individual feedback items and
not the call-level summary.
Negative Feedback Measures The number of negative feedback items.
Neutral Feedback Measures The number of neutral feedback items.
Passive Rate (%) Measures The percentage of neutral feedback items
from the total set of feedback items. Since a
call can have multiple feedback items, this
measure is based on individual feedback
items and not the call-level summary.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 33

Chapter 3: Dashboards

Measure Measures Description

folder

Passive Rate (%) (cwf - Measures The percentage of neutral feedback items
agent, feedback totals) from the level-based total set of feedback
items grouped by agent. Since a call can
have multiple feedback items, this measure
is based on individual feedback items and
not the call-level summary.
Passive Rate (%) (cwf - Measures The percentage of neutral feedback items
feedback totals) from the level-based total set of feedback
items. Since a call can have multiple
feedback items, this measure is based on
individual feedback items and not the
call-level summary.
Positive Feedback Measures The number of positive feedback items.
Promote Rate (%) Measures The percentage of positive feedback items
from the total set of feedback items. Since a
call can have multiple feedback items, this
measure is based on individual feedback
items and not the call-level summary.
Promote Rate (%) (cwf - Measures The percentage of positive feedback items
agent, feedback totals) from the level-based total set of feedback
items grouped by agent.Since a call can
have multiple feedback items, this measure
is based on individual feedback items and
not the call-level summary.
Promote Rate (%) (cwf - Measures The percentage of positive feedback items
feedback totals) from the level-based total set of feedback
items. Since a call can have multiple
feedback items, this measure is based on
individual feedback items and not the
call-level summary.
Total Feedback Measures The number of positive, negative, and
neutral feedback items.

34 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

Appendix A: Support information

Avaya Application Support provides coverage for Avaya-developed custom applications,

software, and code with which Avaya solutions meet specific business requirements of
customers. Customers continue to rely on Avaya to provide a single point of accountability for IP
telephony, Contact Center, and Unified Communications applications of customers. Avaya
Application Support extends the value of continuous, comprehensive, and cost effective
maintenance support to ensure maximum system performance, reliability, and availability for the
communications applications of customers.
Customers with a Full Coverage Avaya Maintenance Agreement can extend their Avaya
Software Support (maintenance) services to include coverage of Avaya Contact Flow Analytics.
For standard Avaya hardware and software platform products, an Avaya Global Services
Maintenance Agreement provides proactive management and value to support and maintain
the IP Telephony, Contact Center, and Messaging applications of the customer. Avaya provides
a single point of accountability for remote and on-site technical support, priority response, rapid
trouble isolation and resolution, and parts replacement to maximize communication system
uptime while protecting equipment investments. By purchasing Avaya Application Support,
customers with existing maintenance agreements not only have comprehensive product
support but also obtain break-fix coverage for outages that might occur with respect to Avaya
Contact Flow Analytics.

Information required for support

When you contact an Avaya Professional Services representative, you must provide the
following information:
● Your full name, the name of organization, and the telephone number where an Avaya
representative can contact you.
● The Avaya Sold-To number, also known as the Functional Location (FL) number.
● A description of the problem.
● The type of service contract your organization has with Avaya.
● A description of the Avaya Contact Flow Analytics configuration.
● A description of any APS contracts for Avaya Contact Flow Analytics.
● Remote access to the Avaya Contact Flow Analytics systems. For more information, see
Remote Access for Technical Support.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 35

Appendix A: Support information

Support limitations and restrictions

The following limitations and restrictions apply to Avaya Contact Flow Analytics software
support:
● The Avaya Contact Flow Analytics software warranty only applies to the software media,
not to non-Avaya supplied solution components. At this time, the only exception to this rule
is a software defect.
● For ongoing support of Avaya Contact Flow Analytics, you need an advanced Avaya
Application Support service contract. If you are uncertain about the details or expiration
date of an existing service contract, contact your Avaya sales representative.
● You can only add the hardware or software that meets Avaya requirements to the Avaya
Contact Flow Analytics configuration. If your Avaya support representative suspects that a
non-standard hardware or software is causing Avaya Contact Flow Analytics-related
problems, the representative will ask you to remove that component. Application Support
coverage for Avaya Contact Flow Analytics includes:
● Case management, resolution, and follow-up.
● Status on all open issues associated with the Avaya Contact Flow Analytics Application.
● Application Correction as appropriate for the Advanced Application.
● Repair of critical errors in the code.
● Escalation to the Avaya Contact Flow Analytics Application Development team.
● Participating in recreating, testing, and resolving issues with the Avaya Contact Flow
Analytics application.
Support aspects
● Remote Break-Fix Support: Includes the entire spectrum from receipt of the customer
call and diagnosis of the fault toThe graph presents information over a period of time r
rectification and return of the Avaya Contact Flow Analytics Application to its functionality
as originally delivered to the customer.
● Avaya Contact Flow Analytics Application Bug Fix: Consists of actions or changes to
resolve anomalies that are causing or contributing to the fault reported by the customer.
● Scope of Problem Fixes: Limited to correcting problems that are reproducible in a
supported version of the software, running unaltered in the original or subsequently
approved configurations.
● General Usability Questions: Questions related to features or functions of the application
that are answered during the business day. Questions do not include consulting on
Application Programming Interfaces (APIs), Software Development Kits (SDKs), or any
other advanced services.

36 Speech Analytics Reporting Comments? infodev@avaya.com March 2013

 Remote access for technical support

● Support for the Avaya hardware platform standard software or operating systems on which
the Advanced Application resides. Support for the hardware platform is covered by an
Avaya Maintenance Agreement.
Application Support coverage for Avaya Contact Flow Analytics does not include:
● Testing of customer-provided equipment.
● Resolving interoperability issues with other vendors.
● New or additional feature functionality, capabilities or capacity beyond the scope of the
features sold with the Avaya Contact Flow Analytics including Major Release Upgrades to
optional features or functionality that the customer did not previously license or which
Avaya licenses as separate products. For example, any program or functional
enhancements to Avaya Contact Flow Analytics.
● Modifiying the Avaya Contact Flow Analytics Application when an upgrade to an
associated standard platform element results in functionality changes to the Avaya Contact
Flow Analytics application.

Remote access for technical support

Avaya support personnel require remote access to provide technical support. To provide remote
access, set up Avaya Web Conferencing

Support contacts

Contacting Avaya Technical Support

Application Support is available to customers through the same Maintenance Help Desk
support number applicable to the core Avaya Software Solution. By calling one number, the
customer has full access to both the core solution support and the Avaya Contact Flow
Analytics application covered by the Application Support agreement.
To report problems or ask questions about your product, call the following telephone numbers:
For support within the United States:
1-800-242-2121
For international support:
See the 1-800 Support Directory listings on the Avaya website.

Speech Analytics Reporting Comments? infodev@avaya.com March 2013 37

Appendix A: Support information

Contacting Avaya Professional Services

For more information about maintenance and software support offers, contact Avaya
Professional Services (APS).
Within the United States, call 1-866-282-9266.
U.S. and international customers can also visit the Contact Center APS website.

Escalating a technical support issue

To escalate urgent service issues, contact Avaya Global Services Escalation Management. For
more information, see the Escalation Management listings on the Avaya website.

38 Speech Analytics Reporting Comments? infodev@avaya.com March 2013