Techniques for estimating expected performance of a task assignment strategy in a task assignment system are disclosed. In one particular embodiment, the techniques may be realized as a method comprising receiving, by at least one computer processor communicatively coupled to a task assignment system, a plurality of historical agent task assignments; determining, by the at least one computer processor, a sample of the plurality based on a strategy for pairing agents with tasks; determining, by the at least one computer processor, an expected performance of the strategy based on the sample; outputting, by the at least one computer processor, the expected performance; and optimizing, by the at least one computer processor, the performance of the task assignment system based on the expected performance.

Systems and methods described herein may describe how enterprise system devices (e.g., servers) may be used to consolidate multiple disparate data sources (e.g., databases) into a single data source (i.e., “Datamart”). The Datamart may be a database or cluster of aggregated data from disparate data sources, which the Datamart may convert into a compatible format using various application programmable interfaces (APIs). In some cases, a software product may query the Datamart and then generate reports for understanding the data pulled from disparate sources. The reporting software application may enable enhanced analytics by humans or additional software applications, to build a more sophisticated understand around a member's call experience, reasons for call transfers, effectiveness of sales by a member service representative (MSR), and how to better train or equip MSRs to optimize their efforts.

Prompt snippets (typically 800 ms long) that are used to detect voice prompts within a call waveform may be divided into smaller sniplet portions (approx. 100 ms) long. The presence of a prompt in a call waveform may be detected by detecting the sniplets and determining if a sufficient number of the sniplets of a snippet were detected in sequence and within allowable time constraints. The use of sniplets improves accuracy of prompt detection in call waveforms in lower quality transmissions.

A computing system may include a plurality of first client devices associated with customers, a plurality of second client devices associated with support agents, and server configured to communicate with the first and second client devices. The server may establish support sessions between the first client devices and the second client devices via a communications network based upon customer support requests from the first client devices, collect customer feedback data on the support agents from the first client devices responsive to the support sessions, collect support agent feedback data on the customers from the second client devices responsive to the support sessions, rank the support agents based upon the customer feedback, rank the customers based upon the support agent feedback, and assign the support agents to customers as customer requests are received based upon the rankings of available support agents and the rankings of the customers making the customer requests.

A web-based contact center state engine provides data describing the state of the contact center system and actionable intelligence including key performance indicators. The contact center state engine may be utilized in conjunction with the network monitoring appliance which processes and manages exceptions to the call center data allowing for action, exceptions and escalation, thereby alerting an organization to an issue and providing recommended actions in addition to post event forensic data.

A real-time speech analytics system (“RSTA system”) detects a speech condition during a call involving a contact center agent and a remote party. Upon detecting the speech condition, an alert message is provided to an alert reporting module (“ARM”), which is configured to access various data to form a RTSA alert. In one embodiment, the RTSA alert is a transient alert indication overlaid on an agent icon on a grid where the agent icon represents the agent and is displayed to a contact center supervisor. Information on the type and severity of the alert may be conveyed by text and non-text images, such as icons, colors, or symbols. A number representing a cumulative number of alert messages received for each agent may be indicated in an alert bubble overlaid on the agent icon. A viewer is able to request detailed alert data upon selecting the alert bubble.

A method for determining evaluation information using audio data includes receiving a specification of an evaluation form, the evaluation form including a number of questions, each question having a number of possible answers. A user is received indicating that a first answer of the number of possible answers for a first question of the number of questions should be determined automatically from the audio data, the audio data including speech of a subject. A user input is received associating a first constraint of a plurality of constraints with the first answer. The audio data is processed to determine whether the audio data putatively satisfies the first constraint. The first answer is assigned as an answer to the first question if the audio data putatively satisfies the first constraint. The first answer is stored as the answer to the first question in the evaluation form.

A method, a system, and computer readable medium comprising instructions for real-time monitoring of agent adherence. The method comprises collecting events and data for an agent from at least one phone router, collecting time keeping data from a time clock system, collecting data and events from a scheduling system, normalizing the events, data, and generating at least one user interface comprising normalized data, presenting at least one view of the at least one user interface to at least one application, and refreshing the at least one view with updated events and data.

A system and method are presented for advanced campaign management in outbound dialing platforms. In a contact center environment, a dialing campaign may be configured such that the outbound dialing platform modifies its behavior automatically throughout the dialing campaign as conditions change. Dialing campaigns may be constructed from groups, which are organized into a campaign sequence. As the campaign sequence flows between groups, details of the campaign may be monitored and evaluated by the platform. Group transitions may be initiated based on activation triggers. In an embodiment, campaign sequences may also be automatically transitioned to new campaign sequences, paused, or reset based on activation triggers. A transition may also be initiated by manual override of the platform.

Systems and methods for assessing the skills of a customer support agent using one or more Artificial Intelligence/Machine Learning (AI/ML) models are disclosed. In at least one embodiment, one or more benchmarks against which the performance of the customer support agent is to be measured are established. The one or more benchmarks may be derived through direct and/or indirect analysis of historical customer service data by an AI/ML benchmark model. In at least one embodiment, data relating to performance of the customer support agent during a customer call is monitored. In at least one embodiment, the AI/ML benchmark model is used to determine one or more benchmark scores identifying whether the customer support agent is meeting the one or more benchmarks.