For a long time, call center quality management was carried out with manual spot checks. With rapidly increasing call volumes and emergence of new contact methods, manual quality inspection is no longer sufficient to meet modern contact centers needs. Agent quality inspection must be done efficiently and accurately, which can only be achieved by relying on advanced technology.                                                                                                         

  • Coverage

At present, the sampling rate of the industry is approximately 1-2%, which means that a large number of recordings go ignored. Hidden pointers and risks are not effectively identified.

  • Business value

Because coverage rate is relatively low, it is impossible to grasp customer needs and seize business opportunities accurately. The business value of quality inspectors is subsequently not well-established.

  • Quality inspection delay

Manual quality inspection is generally conducted after service. This method cannot respond to incidents and risks in real-time and in a timely manner.

  • Work efficiency

Manual quality inspection usually requires repeated playback. This results in inefficiency and limits the number of problems that can be found during working hours.

  • Quality inspection standards

 Manual quality inspections are limited by individual perceptions. Different people draw different conclusions from the same event; a person’s perception also changes over time. This caused difficulties in establishing a unified quality inspection standard.

  • Quality inspection costs

Manual quality inspections are highly repetitive and heavy. As call volume increases, companies will need to invest more and more manpower and material to meet inspection needs and inspection proportions.

 

 

Application Scenario

With the development of artificial intelligence technology (speech recognition, big data mining), mass recording data can now be converted (real time or offline) into unstructured data such as text. We can achieve 100% voice file coverage and greatly reduce the cost of quality inspection. Thanks to text analysis and data mining, it is now possible to achieve compliance inspection, risk warning, trend analysis, and business opportunity mining.

Technology Structure

 

System Structure

 

Functional Characteristics

Record Storage and Management

Converts large-scale voice calls to text indexes. Long-term support for high-speed retrieval, playback, and analysis of voice data.

Recording Search

The system provides extensive and flexible search conditions. It highlights search terms and color codes caller and callee.

Recording Playback

System visually displays call sound, call content, and timeline; caller and callee color coded. Click a word to accurately position playback.

Business Modeling

Standard manual quality inspection workflows can be converted into system-recognizable rules via role settings, position settings, range settings, and expression combinations. Freely customize and recreate different workflows.

AI-assisted Quality Inspection

Establish quality inspection rules (keywords, sentiment indicators, mutes, etc.) and automatically process and score service capabilities and attitudes. Solves the problem of low manual coverage.

Manual Inspection

The system can filter out low-rated calls for further manual quality inspection; quality inspection subsequently follows the model “AI batch inspection + operator targeted inspection”. This model facilitates comprehension and enhances customer service and contact center service quality.

Data Mining

The report engine and customizable report designer provide flexible and adaptable reports. Improve correlation and achieve data mining from report display to content query.