Ender Turing's AI-powered quality evaluation tool is designed to help businesses automate and streamline the quality assurance process for contact center conversations (calls, chats, support tickets, etc).

By leveraging Generative AI, Ender Turing offers a scalable solution that evaluates conversations in real-time, providing actionable insights to improve agent performance and ensure customer satisfaction.

Generative AI for conversation quality evaluation involves leveraging LLMs (Large Language Models) to analyze and assess customer-agent interactions. The AI reviews conversations based on predefined criteria such as tone, resolution accuracy, compliance with guidelines, overall customer sentiment, etc. It automates the traditionally manual quality assurance process by offering consistent and unbiased evaluations at scale.

This guide will walk you through how to maximize Ender Turing's AutoQA for your business.

Using a generative AI model for evaluating conversations begins by anonymizing the customer-agent interactions to remove personal data, ensuring compliance with privacy regulations. The anonymized conversation is then sent to the AI with a predefined scorecard outlining evaluation criteria such as tone, resolution accuracy, compliance, and sentiment. The AI processes the conversation, generates a score for each criterion, and provides actionable feedback, highlighting areas for improvement. Managers with agents then review these results to help refine their performance, while trends can be monitored over time for continuous optimization.

To use AutoQA in Ender Turing, you need to create:

You can use the default Scorecard or set up your company's Scorecard/s. The process of creating a Scorecard is described here.

! Important - Provide a detailed description (explanation) for each point of the Scorecard

After creating a Scorecard, carefully add each evaluation point's description (explanation). This description will be passed on to AI as instructions for evaluating each point. The more accurate the description (explanation) you provide, the more accurate the automated scoring results will be.

Read our detailed guide on improving automated evaluations' description (explanation).

Here is an example:

Evaluation Point name: Greeting

Evaluation Point description: Is a polite and professional greeting present in the following conversation excerpt?

! Important - To use the created Scorecard for automation, switch on the 'Use as Automated' and 'Set as Protected' checkboxes on top of the Scorecard.

Ender Turing provides you with a default Scorecard, which can be used to evaluate conversation. Default Ender Turing Scorecard, which includes:

To start automated evaluations, you need to provide rules on which conversations to score and which to skip. You can also define multiple rules, where, for example, tickets will be evaluated using Scorecard-Tickets, but calls will be evaluated using different Scorecard-Calls; or process-based rules: Service conversations and Sales conversations by different Scorecards, etc.

Such rules are created using Enders. The process of creating Ender is described here.

Example:

Below, you can see an example of defined rules:

Which conversations to evaluate: Only Incoming calls that last duration of more than 30 seconds

Which scorecard to use: AutoQA Scorecard

Often, you don't want to have 5-10 Scorecards but wish to have only one unified Scorecard, and based on rules, Score or Skip some part of Scorecard evaluation.

You can select which points of Scorecard to score and which to skip by selecting them in particular Ender by clicking on the scorecard button:

By clicking on a Scorecard button, a new window will be opened where you can select which points to score and which to skip.

The example below shows that the Solution part of the Scorecard will not be evaluated, and N/A will be used as a value.

In rare cases, you might need to set the points that haven't been evaluated to Passed or even Failed without any generative evaluation. You can configure such parameters using the drop-down menu:

There are a few limitations on AutoQA usage: