Saturday, November 20, 2010

Tuesday, October 19, 2010

SATMAP

SATMAP, an acronym for SATisfaction MAPping, is an artificial intelligence application for matching callers to agents in a contact center, based on a real-time analysis of diverse personality attributes. It is a neural-network based call-mapping engine that connects customers to the contact center agents by taking into account key psychographic and demographic factors. SATMAP is the first ever intelligent call routing system available for contact centers that significantly and quantitatively improves customer satisfaction, reduces average call handle time and increases sales. It is a registered trademark owned by The Resource Group (TRG).

History

SATMAP was founded in 2004 as a wholly owned subsidiary of The Resource Group, a global business process outsourcing and technology company. After providing audited performance improvements to the global operations of TRG, SATMAP has increasingly provided technology solutions to the broader contact center industry since 2007. With SATMAP, the calls no longer need to be routed to the best performing or the longest waiting agent. Instead they can now be routed to the agent that has the best personality match for each caller. Skill Based Routing (SBR), a traditional call routing methodology, has been implemented extensively in the contact center industry. Over the years, skill based routing has proved to be a widely implemented solution for contact centers. Some of the global telecommunication companies following the SBR call routing methodology include Avaya, Nortel, and Cisco etc. SATMAP strives to add intelligence to the existing call routing systems by selecting the best agent from the agent pool based on the demographic and psychographic attributes of both the caller and contact center agents. SATMAP makes the ultimate decision as to which individual agent is assigned to a particular caller, based on complex but highly intuitive personality fits.

SATMAP Solutions

SATMAP has numerous call routing solutions which are based on different optimization metrics, as discussed below:

SATMAP Enterprise (SME) customers are given up to eight different optimization metrics, including the traditional metrics of revenue optimization, cost reduction, and customer satisfaction. With five additional slots, SME customers can optimize diverse metrics including insurance claims pay put, customer collections, medical diagnosis accuracy, and other industry-specific metrics.

Performa enables a proactive approach to contact center management by facilitating dynamic control of operations based on rapidly changing calling conditions. Performa is an entry-level system that is upgradable to SATMAP Enterprise.

TS is a neural-networks- based call routing system tuned specifically to optimize technical support functions at contact centers. TS customers are given options to customize their optimization metrics including those that are critical to determining the success of technical support objectives: truck rolls, equipment replacements and call escalations.

Revmax is used purely as a revenue-enhancement application for contact center environments. It analyzes detailed caller and agent attributes in real-time to determine which caller-agent combination will result in the highest likelihood of a revenue outcome.

L2 is configured to dynamically manage contact center call queues in order to optimize the result of every customer-agent interaction. L2 tracks every call's unique characteristics and detailed caller attributes to create a holistic view of incoming calls. A neural-network algorithm is used to match these incoming calls with relevant agents, based on performance and personality fit ensuring that agents are connected to the calls they can most likely resolve successfully.

SATMAP Features

Connecting caller with the right agent

In contact center environments, purchase behavior is driven by the degree to which agents build empathy and rapport with a caller. This, in turn, is a direct function of the degree to which the personality of agents and callers fit together. By connecting the caller with the right agent, SATMAP caters three different attributes related to client’s business preferences. These goals which are concentrated upon by the SATMAP team are:

  • Customer Satisfaction
    In a competitive marketplace where businesses compete for customers, customer satisfaction is seen as a key differentiator and has increasingly become a key element of business strategy. SATMAP reaches the customer satisfaction level by connecting a caller to an agent who has the potential of handling the call in an effective manner that leaves the caller satisfied. It also maintains data on satisfaction score of the agent by getting a feedback from the customers. So in this case the factor that is taken into consideration for the decision making is the Historical Satisfaction Score of the Agent.
  • Revenue
    A company’s performance is measured by the extent to which its revenues compare with its expenses, thereby making revenue boost a critical business objective. Revenue can be optimized with more n` more calls resulting in sales i.e. a customer ends up buying a product as a result of successful communication with the agent. Keeping this concept in mind SATMAP considers the following factors when creating agent and caller mapping: historical agent data, background of the agent, customer’s buying propensity, agent mood.
  • Cost
    Cost can be reduced by connecting an incoming call to an agent that takes less time in handling the call; ultimately saving agent’s time as well as the cost incurred on the call. Apart from the ability to handle calls in lesser time, an agent who shares a similar background with the customer in terms of language, ethnicity or same age group stands a better chance of conveying the right message in short time. SATMAP mainly focuses on the historical handle time of the agent and agent’s background for cost reduction.

Support for customized algorithm implementation

SATMAP can customize the algorithms that help in decision making based on client’s requirements and preferences. The changes in the algorithm also vary due to the SATMAP solution used to optimize the caller-agent pairings. Thus the algorithms used by SATMAP for the purpose of decision making can easily adapt to the particular choices laid out by the client.

Self–learning Capability

SATMAP is capable of updating and refining itself based on the machine learning algorithm selected for a certain client. This is performed through an integral component which keeps on executing updating cycles after regular intervals. During the course of an update cycle, the performance of an implemented algorithm is analyzed and based on the results, the algorithm can be replaced if required. This self-learning capability serves as a means of better optimization of the overall working of the system.

Support for multifarious switches and related software products

SATMAP works natively with most commercial skill based routing and cloud-routing systems and integrates seamlessly with most major third party technology platforms. SATMAP functions on many different platforms which include switches and other related products.

  • Aspect Call Center ACD
  • Aspect Rockwell Spectrum
  • Asterisk
  • Avaya
  • Cisco UCCE
  • Genesys
  • Interactive Intelligence
  • Nortel
  • Stratadial
  • Sytel Softdial Contact Center
  • TeleDirect Liberation 6000 Dialer
  • trg Dialer

Seamless integration of SATMAP in a contact center

SATMAP integration is a low-impact process and does not require changes to recruitment, training, compensation or other existing business processes of the contact center. SATMAP only utilizes the existing resources that the client has invested in the contact center.

SATMAP failsafe feature

In case SATMAP is unable to cater the call route request, the call continues the normal course which results in the call being sent to a skill queue and assigned to an agent by the PBX. In this way SATMAP executes a failsafe approach for handling calls and ensures that the caller receives the same treatment when waiting for an agent, regardless whether SATMAP is running or not.

SATMAP Architecture

SATMAP interfaces with multiple modern and legacy PBX switches. The communication between switch and SATMAP is managed through a switch interface, responsible for providing a translation between the switch and internal SATMAP events. The data for decision making is continuously refined by SATMAP through the use of several algorithms which assist in filtering calls and agents for SATMAP’s decision making. Some additional information in the form of customer and agent demographic as well as psychographic attributes aids this decision making process. The output i.e. the optimal caller-agent pair is sent back to the switch interface which in turn sends it to PBX for routing. In order to perform its integral functionalities, SATMAP interacts with a number of external systems that support the overall operations of the application. These external systems include:

Agent Information Management System (AIMS)

The data pertaining to the agents is managed by the Agent Information Management System. AIMS is a key component of SATMAP deployment as it maintains the agent survey application which the agents fill out at the onset of their employment in a SATMAP-enabled contact center. This survey becomes the source of the agent’s demographic and psychographic information which is then used to generate optimal caller-agent mappings. Demographic data are the characteristics of a population, like sex, race, age, income, location, disabilities, mobility, educational attainment, home ownership, employment status. Psychographic data, on the other hand, encompass attributes pertaining to lifestyle, attitudes, beliefs, values and personality. The contact center agent enters the data into AIMS by completing a survey consisting of approximately 150 questions. To ensure the accuracy of the data, every 30 days the agent is asked to complete a refresh survey. Both the initial survey and the refresh survey can be customized for specific applications.

SATMAP Portal

The SATMAP Portal is a web based application which provides comprehensive turnkey reporting and allows the management to monitor the real-time performance of a campaign. Reports are available on:

  • Agent and customer demographics
  • Agent performance
  • Call center and campaign performance

Call History Server

The Call History Server is a logical server used to collect and collate call history from one or more sources. To collect the historical call data, the Call History Server uses different data access methods.

Acxiom

Acxiom, a global interactive marketing services company, provides information on the caller demographics and psychographics.


Science Behind SATMAP

A typical contact center consists of a number of human agents, each assigned to a telecommunication device, such as a phone or a computer for conducting email or Internet chat sessions, that is in turn connected to a central switch. Using these devices, the agents are generally used to provide sales, customer service, or technical support to the customers or prospective customers of a contact center or a contact center's clients. Industry research consistently shows that the central driver of customer satisfaction is the degree to which agent and client established a rapport over the course of their interaction. In turn what drives rapport and engagement is the degree to which agent and customer personalities coincide. Through SATMAP intelligent call routing system; this concept is incorporated to deliver best services to the valued customers. SATMAP optimizes the routing of callers to agents in a contact center. In general, contact routings are optimized by routing contacts such that callers are matched with and connected to particular agents in a manner that increases the chances of an optimal interaction that is deemed beneficial to a contact center. This optimal interaction is improved by grading agents and matching a graded agent with a caller. SATMAP increases the chance of an optimal interaction by matching a caller to an agent using a computer model derived from data describing demographic, psychographic, past purchase behavior, or other business-relevant information about a caller, together with data describing demographic, psychographic, or historical performance about an agent. Agent and caller demographic data can comprise any of: gender, race, age, education, accent, income, nationality, ethnicity, area code, zip code, marital status, job status, and credit score. Caller demographic and psychographic data can be retrieved from available databases by using the caller's contact information as an index. Available databases include, but are not limited to, those that are publicly available, or those that are commercially available, or those created by a contact center or a contact center client. Once agent data and caller data have been collected, this data is passed to a computational system. The computational system then, in turn, uses this data in a pattern matching algorithm to create a computer model that matches each agent with each caller and estimates the probable outcome of each matching along a number of optimal interactions, such as the generation of a sale, the duration of contact, or the likelihood of generating an interaction that a customer finds satisfying. The pattern matching algorithm can comprise any correlation algorithm, such as a neural network algorithm or a genetic algorithm. To generally train or otherwise refine the algorithm, actual contact results (as measured for an optimal interaction) are compared against the actual agent and caller data for each contact that occurred. The pattern matching algorithm can then learn, or improve its learning of, how matching certain callers with certain agents will change the chance of an optimal interaction. In this manner, the pattern matching algorithm can then be used to predict the chance of an optimal interaction in the context of matching a caller with a particular set of caller data, with an agent of a particular set of agent data.


Friday, October 8, 2010