Contact center analytics looks a lot like Big Data
As enterprises seek to make more sense of complex customer interactions they come to recognize the limits of their siloed legacy data capture systems, and many are turning to new analytics systems that have similarities to IT-based Big Data tools. Some of the concepts behind Big Data have started to leak into the product marketing and messaging of contact center analytics vendors, and some of those contact center analytics vendors have in turn been acquired by traditional telephony vendors looking to solve the siloed data problem. Contact centers are ripe for this kind of effort: they sit at the nexus of multiple data streams but have been slow to make effective use of most of them. Decision-makers who want to position their contact centers as strategic and profit-centric (rather than merely operational and cost-centric) should pay close attention to the changes in how their own IT departments are managing other large corporate data sets.
Contact centers amass big data without considering it â€śBig Dataâ€ť
Contact centers collect massive amounts of data through the course of their daily operations. This data consists of both structured and unstructured information from sources as varied as telephony equipment, CRM systems, call- and screen-recording, and storage applications. Data from these sources is heavily siloed, which makes it very difficult for contact centers to see the whole picture of the customer relationship, especially as customer interactions become more complex. These interactions now frequently cross multiple contact channels, and are increasingly mobile. Companies typically lack a single view of the customer due to separate ownership of call center data, sales and marketing systems, and website tracking.
Most contact centers do very little to commingle and analyze their data on a large scale, even though contact center data fits most of the criteria for Big Data and this environment represents one of the best available use cases for Big Data analytics. Ovum defines Big Data as computational problems that are large and varied enough to demand new approaches to traditional SQL technology, problems that are typified by four Vs: Volume of data (huge), Velocity of input/output, Variety of sources, and high Value to the organization. Big Data is typically deployed in NoSQL or Advanced SQL databases.
Customer care organizations that record and store all of their calls can easily gather and index hundreds of thousands of hours of audio per month, racking up terabytes of data on an ongoing basis. Most traditional call recording vendors have built applications that store this data in proprietary databases that are non-SQL. Most are now also scraping and storing agent screen information and other pieces of the customer interaction, creating more complexity.
Contact centers have the data for Big Data, but most do not make use of the tools that could extract the most value from their stores.
Emerging contact center analytics tools stand in for Big Data tools
The contact center market has seen the emergence of specialty applications that aggregate and then analyze data from different sources. Contact centers have traditionally had two uses for analytics: to improve agent performance and to understand customer behavior. Established vendors in the contact center space have traditionally focused on questions of agent performance, which are much more limited in scope than those of customer behavior. These vendors have not taken a Big Data-style approach because the answers to agent performance questions are usually solved by sampling from large, segregated data sets.
However, contact centers have started to look at vendors that are exploring the customer experience side of the problem using databases designed according to the principles of Big Data. As businesses have asked harder and more complex questions about their customer interactions, contact centers have responded by turning to analytics tools that mimic business intelligence applications. They are subjecting a larger corpus of data to greater scrutiny through tools that are functionally a hybrid of IT-friendly business intelligence systems and traditional call center performance-optimization systems.
The spectrum of analytics tools available to contact centers includes offerings from the original telephony vendors that are based on storing and retrieving very basic information from call recordings and workforce management systems. These include legacy quality monitoring from NICE, Verint, VPI, and others.
There is also a new generation of tools that look and act like Big Data. These tools are produced by companies that are specialists in extracting meaningful information from giant sets of different data. Vendors such as Fizzback, Clarabridge, ClickFox, and Merced all have a grasp of the sophisticated manipulation of information, and at least some of their offerings are described as NoSQL or Big Data. It is no coincidence that they were all either acquired by legacy quality monitoring and call recording companies, or are working very tightly with them.
Centers should view their data stores as a key resource
Contact centers should immediately begin looking at their data as a meaningful resource, rather than as a simple by-product of their operations. They should invest in data management tools that, like Big Data systems, break down informational silos, and they should do it in collaboration with their enterprise IT organizations. Their IT colleagues are likely to already be looking at enterprise-level data, searching for the elusive big picture that the contact centerâ€™s data can help to provide.
Similarly, IT companies that are looking for effective use cases for Big Data technology should look squarely at the contact center. There they can find data sitting unexplored, just waiting to be integrated, analyzed, and monetized.
All Rights Reserved.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher, Ovum (an Informa business).
The facts of this report are believed to be correct at the time of publication but cannot be guaranteed. Please note that the findings, conclusions, and recommendations that Ovum delivers will be based on information gathered in good faith from both primary and secondary sources, whose accuracy we are not always in a position to guarantee. As such Ovum can accept no liability whatever for actions taken based on any information that may subsequently prove to be incorrect.