Campaign Analytics Engine for Cloud Contact Center Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud-based contact center platforms provide limited data for campaign analytics, making it difficult to determine campaign effectiveness in customer conversion, which can lead to revenue loss for tenants aiming to expand their customer base.
Innovation Solution
A system comprising a Microservice for campaign management, a campaign configuration dashboard, a transcription serverless computing platform, and a campaign analytics engine that uses automatic speech recognition and Natural Language Processing to generate analytical data, including agent campaign delivery effectiveness scores and probability of conversion, to optimize campaign performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional campaign management systems are used, then system simplicity is maintained, but analytics data completeness is insufficient
Solution Approach 1:
The system is divided into independent microservices including transcription service, NLP service, analytics service, and campaign management service. Each service handles a specific function (speech-to-text conversion, sentiment analysis, analytics generation, campaign orchestration), allowing the system to process and capture comprehensive interaction data without creating a monolithic complex structure. This segmentation enables complete analytics data collection while maintaining manageable system architecture.
Solution Approach 2:
The patent introduces intermediary services between the contact center platform and the analytics pipeline. The transcription service acts as an intermediary to convert speech to text, the NLP service as another intermediary to extract sentiments and topics, and the analytics service as a final intermediary to generate insights. These intermediaries enable complete data capture and analysis without directly complicating the core campaign management system.
2Measurement precision
If basic campaign tracking is implemented, then implementation ease is maintained, but campaign effectiveness measurement is insufficient
Solution Approach 1:
The analytics service continuously monitors campaign interactions and provides feedback metrics including customer sentiment scores, topic analysis results, agent performance evaluations, and conversion probability predictions. This feedback loop enables precise measurement of campaign effectiveness by comparing actual interaction outcomes against campaign objectives, allowing for real-time optimization while managing analytics complexity through automated reporting.
Solution Approach 2:
The system measures campaign effectiveness by transforming raw interaction data into multiple analytical parameters including sentiment scores (positive/negative/neutral), topic categories, agent delivery effectiveness scores, and probability of conversion. These parameter transformations enable comprehensive effectiveness measurement by converting unstructured conversation data into quantifiable metrics that directly reflect campaign performance.
3Measurement precision
If manual campaign monitoring is used, then system simplicity is maintained, but agent performance evaluation is insufficient
Solution Approach 1:
The analytics service automatically performs agent performance evaluation by analyzing interaction transcripts without requiring manual review. The NLP service extracts sentiments and topics, the transcription service converts speech to text, and the analytics service generates delivery effectiveness scores and performance metrics automatically. This self-service automation enables precise agent performance measurement while eliminating time loss associated with manual evaluation processes.
4Loss of information
If comprehensive data collection is implemented, then analytics quality is improved, but processing complexity increases
Solution Approach 1:
The data processing pipeline is segmented into specialized services: transcription service handles speech-to-text conversion, NLP service handles sentiment and topic extraction, and analytics service handles insight generation. Each service processes a specific type of data transformation, enabling comprehensive interaction data capture while distributing processing complexity across multiple focused components rather than concentrating it in a single complex system.
Data Source
AI summary
A system for supporting effective campaign management in a cloud-based contact center platform. The system includes a campaign analytics engine. The campaign analytics engine includes retrieving campaign offerings of a running campaign; retrieving time-stamped text transcripts of interactions of agents of the running campaign; parsing the retrieved time-stamped text transcripts of interactions to yield parsed transcripts and related customer sentiment; comparing the yielded parsed transactions with the retrieved campaign offerings to generate analytical data. The generated analytical data includes at least one of: (i) agent campaign delivery effectiveness score; and (ii) probability of conversion. When an agent has an agent campaign delivery effectiveness score below a first threshold excluding the agent from the running campaign and sending the campaign delivery effectiveness score of the agent along with agent details and a preconfigured training list to be assigned to the agent by a QM application.