1. Overview

Predictive outbound is an efficient batch outbound method in a call center system. It dynamically adjusts the dialing frequency through intelligent algorithms, implementing a "call the customer first, then transfer to an agent" process to maximize agent utilization and reduce wait times.

🎯 Core Value

• Improve Agent Efficiency: Agents only handle answered calls, effective talk time can exceed 50 minutes per hour.

• Intelligent Frequency Control: Dynamically adjusts dialing pace based on real-time data to balance answer rate and nuisance call rate.

• Fully Automated Execution: The system automatically completes the entire process of number dialing, status judgment, and agent transfer.

• Dual-mode Call Flow: Supports both direct-to-agent and AI-to-human transfer business scenarios.

⏱️ Estimated Operation Time

• Task Creation & Configuration: Approximately 5-8 minutes.

• Number Import & Verification: Approximately 5 minutes.

• Warm-up & Learning Phase: Approximately 5-15 minutes (completed automatically by the system).

• Task Monitoring & Adjustment: Continuous during task runtime.

📞 Two Call Flow Modes

The predictive outbound system provides two core call flow modes to meet different business scenario needs:

Mode One: Direct to Agent (Native Predictive)

• Customer is transferred directly to an agent upon answering, ensuring the best response time.

• Can overflow to IVR when no idle agents are available (requires pre-configured IVR flow).

• Applicable Scenarios: High-value customers, urgent business, scenarios requiring prioritized human handling.

• Advantages: Best customer experience, fastest response speed, highest agent utilization.

Mode Two: AI to Human (Intelligent Assistant Robot)

• Customer first enters the Intelligent Assistant Robot within the IVR upon connection.

• Intelligently judges whether to transfer to an agent based on interaction results.

• Applicable Scenarios: Business inquiries, information collection, lead qualification, standardized services.

• Advantages: Increases automation coverage, reduces labor costs, provides 7×24 service.

graph TD
    A[Predictive Outbound Task Starts] --> B[System Dials Customer]
    B --> C{Customer Answers?}
    C -->|No| D[Recorded as No Answer]
    C -->|Yes| E{Choose Call Flow Mode}
    
    E --> F[Direct to Agent Mode]
    E --> G[AI to Human Mode]
    
    F --> H[Direct Transfer to Idle Agent]
    H --> I{Idle Agent Available?}
    I -->|Yes| J[Both Parties Connected]
    I -->|No| K[Overflow to IVR]
    
    G --> L[Enter Intelligent Assistant Robot]
    L --> M{Meet Transfer to Human Condition?}
    M -->|Yes| N[Transfer to Human Agent]
    M -->|No| O[Robot Completes Service]
    
    K --> P[IVR Processing]
    N --> J
    J --> Q[Record Call Result]
    O --> Q
    P --> Q
    D --> Q
    
    style F fill:#54a0ff,color:#fff
    style G fill:#1dd1a1,color:#fff
    style J fill:#ff9ff3,color:#fff

2. Predictive Outbound Core Concepts

📚 Basic Definitions

Predictive Outbound is an advanced mode of batch outbound dialing in call centers. The system uses a prediction algorithm to calculate the optimal number of numbers to dial per cycle, calling customers first and transferring them to idle agents only after they answer.

Core Features:

• ✅ Effectively Filters Invalid Numbers: Automatically identifies statuses like disconnected, powered off, busy.

• ✅ Intelligent Frequency Adjustment: Real-time optimization of dialing strategy based on multi-dimensional data.

• ✅ Maximizes Agent Efficiency: Agents only handle successfully answered calls.

• ✅ Dual-mode Flow: Supports both direct-to-agent and AI-to-human business modes.

🎯 Core Purpose

To maximize the utilization of agents' working time by reducing agent waiting time for customers while improving overall outbound efficiency.

🔄 Basic Process

graph TD
    A[Create Predictive Outbound Task] --> B[Import Customer Number Data]
    B --> C[Start Task for Warm-up]
    C --> D[System Calculates Dialing Count per 10s Cycle]
    D --> E[Dial Customer Number]
    E --> F{Customer Answers?}
    F -->|Yes| G[Flow According to Configured Mode]
    F -->|No| H[Abandon This Call]
    
    G --> I{Flow Mode}
    I -->|Direct to Agent| J[Direct Transfer to Agent]
    I -->|AI to Human| K[Transfer to Intelligent Assistant First]
    
    J --> L[Agent-Customer Conversation]
    K --> M[Robot Interaction & Judgment]
    M --> N{Transfer to Human?}
    N -->|Yes| L
    N -->|No| O[Robot Completes Service]
    
    L --> P[Record Call Result]
    O --> P
    H --> D
    
    style A fill:#ff6b6b,color:#fff
    style B fill:#feca57,color:#fff
    style C fill:#54a0ff,color:#fff
    style D fill:#1dd1a1,color:#fff
    style J fill:#ff9ff3,color:#fff
    style K fill:#95e1d3,color:#000

📊 Call Status Explanation

📖 Key Term Explanations

3. Predictive Outbound Full Process Details

Step 1: Create a Predictive Outbound Task

Page Location

Outbound Tasks → Predictive Outbound Tasks → Task Management

Operation Steps

1. Click the "New Task" button.

2. Enter the Task Name (within 50 characters).

3. Enter the Task Description (within 200 characters, optional).

4. Configure Basic Settings:

   • Nuisance Rate Coefficient: Default 3, affects dialing frequency.

   • Redial Strategy: Configure retry rules for unanswered numbers.

5. Set Agent-side Configuration:

   • Call Flow Mode: Choose Direct to Agent or AI to Human mode.

   • IVR: Configure the navigation flow after the customer answers.

   • Agent List: Specify agents participating in the task.

   • Agent Assignment Rule: Random / Sequential / Maximum Idle Time.

   • Agent Wrap-up Time: Default 30 seconds.

6. Configure Customer-side Settings:

   • Caller ID Number: Landline / Mobile / Number Pool.

   • Customer Timeout: Default 10 seconds.

   • Dialing Order: Priority between redial numbers and non-dialed numbers.

⏰ Scenario1: Task Start and End Configuration

The system provides flexible task start/stop control, supporting multiple start and end methods:

Start Method Configuration

Option One: Manual Start

• After creation, the task requires a manual click of the "Start" button to begin execution.

• Applicable Scenarios: Important tasks requiring manual confirmation before starting outbound.

• Advantage: Complete control over task start timing, avoids accidental operation.

Option Two: Scheduled Start

• Set a specific date and time; the system automatically starts the task.

• Applicable Scenarios: Planned outbound campaigns, automatic start outside working hours.

• Advantage: Automated execution, no manual intervention needed.

End Method Configuration

Option One: Manual End

• Manually click the "End" button to stop task execution.

• Applicable Scenarios: Scenarios requiring real-time control over task duration.

• Advantage: Flexible control over task end timing.

Option Two: Scheduled End

• Set a specific date and time; the system automatically ends the task.

• Applicable Scenarios: Fixed-duration outbound campaigns, avoiding overtime outbound calls.

• Advantage: Automatically stops, prevents forgetting to end the task.

Dialing Time Configuration

Default Configuration: Call all day (except blocked time periods).

• The system dials by default during all time periods except blocked times.

• No need to configure continuous dialing parameters additionally.

Dialing Time (Original Interval Dialing)

• Defines the specific time periods for task execution.

• The system only performs outbound calls during the configured dialing times.

• Complements blocked time periods to ensure complete time coverage.

Start Logic Explanation

The system adopts a multiple condition intersection start logic:

graph TD
    A[Task Creation Complete] --> B{Scheduled Start Configured?}
    B -->|Yes| C[Wait for Scheduled Start Time]
    B -->|No| D[Wait for Manual Start]
    
    C --> E{Within Dialing Time?}
    D --> F{Within Dialing Time?}
    
    E -->|Yes| G[Task Starts Automatically]
    E -->|No| H[Wait to Enter Dialing Time]
    F -->|Yes| I[Task Starts Immediately]
    F -->|No| J[Wait to Enter Dialing Time]
    
    H --> E
    J --> F
    
    style A fill:#ff6b6b,color:#fff
    style G fill:#1dd1a1,color:#fff
    style I fill:#1dd1a1,color:#fff

Key Rules:

• Scheduled start time must fall within the dialing time range to take effect.

• Manual start takes effect immediately (if within dialing time).

• Blocked time periods have the highest priority, overriding all start conditions.

🔢 Scenario2: Configure Dialing Order Strategy

The system provides flexible dialing order configuration, supporting three sorting methods for non-dialed numbers:

Option One: Sequential (By Priority)

• Dial numbers sequentially according to the batch import order and the specified priority order of numbers.

• Applicable Scenarios: Important business requiring strict control over dialing priority.

• Advantage: Ensures high-priority customers are served first, optimizes resource allocation.

Option Two: Sequential (By Time)

• Dial numbers sequentially according to the time order of import.

• Applicable Scenarios: Fair dialing scenarios requiring processing in the order of import.

• Advantage: Simple operation, easy to understand and manage.

Option Three: Random

• If number priority is not set, dial numbers randomly.

• Applicable Scenarios: Avoids answer rate fluctuations caused by batch number quality differences.

• Advantage: Spreads dialing risk, stabilizes overall answer rate performance.

📞 Scenario3: Configure Call Flow Mode

The system provides two call flow modes to meet different business scenario needs:

Option One: Direct to Agent (Native Predictive)

• Customer is transferred directly to an agent upon answering, ensuring the best response time.

• Can overflow to IVR when no idle agents are available (requires pre-configured IVR flow).

• Applicable Scenarios: High-value customers, urgent business, scenarios requiring prioritized human handling.

• Advantage: Best customer experience, fastest response speed.

Option Two: AI to Human (Intelligent Assistant Robot)

• Customer first enters the Intelligent Assistant Robot within the IVR upon connection.

• Intelligently judges whether to transfer to an agent based on interaction results.

• Applicable Scenarios: Business inquiries, information collection, lead qualification, standardized services.

• Advantage: Increases automation coverage, reduces labor costs, provides 7×24 service.

Step 2: Import Number Data

Method One: Page Import

1. Click the "Import" button on the Task Management page.

2. Download the import template and fill in number information.

3. Configure import parameters:

   • Deduplication Rule: Deduplicate within task / globally.

   • Start Type: Start immediately / Start manually.

   • Batch Name: For easy future management.

4. Upload the file and confirm import.

Method Two: API Import

• Call the platform API to import numbers in bulk.

• Supports finer-grained priority control.

• Suitable for large-scale outbound scenarios.

Step 3: Task Execution and Monitoring

Algorithm Core Logic

Predictive outbound dynamically adjusts the number of calls per cycle through intelligent algorithms:

Calculation Formula:

Number of calls per cycle = (Predicted Idle Agent Count / Customer Answer Rate) × Overdial Parameter

Key Metrics:

• Predicted Idle Agent Count: Idle agents + agents about to finish wrap-up.

• Customer Answer Rate: A combination of real-time and historical answer rates.

• Overdial Parameter: Dynamic adjustment factor (0-500).

4. Predictive Outbound Monitoring Details

📊 Task Monitoring Overview

Page Location: Task Monitoring → Predictive Outbound Task Monitoring

Data Scope: Displays the overall running status of all predictive outbound tasks.

Core Functions:

• Real-time Task Status: View the running status of all tasks (Running, Paused, Ended, etc.).

• Key Metrics Overview: Core metrics like nuisance rate, agent utilization, task completion rate.

• Task Progress Tracking: Number of numbers imported, number dialed, completion rate for each task.

• Abnormal Status Alert: Real-time display of task abnormal status and reasons.

Monitoring Metrics Explanation:

🔍 Task Monitoring Details

Page Location: Task Monitoring → Predictive Outbound Task Monitoring → Click on a specific task

Data Scope: Detailed running data of a single predictive outbound task.

Core Functions:

1. Real-time Data Monitoring

• Call Concurrency Monitoring: Real-time display of task call concurrency, total enterprise concurrency.

• Agent Status Distribution: Number of agents in idle, ringing, talking, wrap-up, etc., states.

• Dialing Frequency Adjustment: Real-time display of overdial parameter dynamic adjustment.

2. Time Period Trend Analysis

Trend charts include:

• Customer Answer Rate Trend: Recent 2-hour answer rate change curve.

• Nuisance Rate Trend: Fluctuation of real-time nuisance rate.

• Agent Inbound Time: Trend of average inbound time changes.

• Dialing Frequency Change: Overdial parameter adjustment record.

3. Multi-dimensional Metric Analysis

Customer-side Metrics Monitoring:

Agent-side Metrics Monitoring:

👥 Agent Real-time Monitoring

Page Location: Task Monitoring → Predictive Outbound Agent Monitoring

Data Scope: Real-time status of all agents participating in predictive outbound tasks.

Core Functions:

1. Agent Status Monitoring

• Real-time Status Display: Idle, Ringing, Talking, Wrap-up, Busy.

• Status Duration: Statistics on the duration of each status.

• Auto-answer Status: Auto-answer situation for WebRTC agents.

2. Personal Performance Metrics

• Today's Answered Count: Number of calls answered by the agent in the current task.

• Average Inbound Time: Personal inbound waiting time.

• Answer Timeliness: Personal performance on 2-second pick-up rate.

• Talk Time Distribution: Distribution of long, medium, and short calls.

📈 Call Record Analysis

Page Location: Call Records → Predictive Outbound Records

Data Scope: Detailed call records for all predictive outbound tasks.

Core Functions:

1. Intelligent Query Filtering

• Dynamic Query Filtering: Select or enter query conditions in the filter bar, click query to refresh data in real-time.

• Multi-condition Combination: Supports combined queries across dimensions like task, batch, time period, call status, agent.

• Query Condition Management:

2. Call Status Analysis

In-depth analysis based on the four call status enum values:

🚨 Real-time Alerts and Intervention

Automatic Pause Condition Monitoring

The system monitors the following conditions in real-time, triggering automatic pauses:

Manual Intervention Functions

Real-time Adjustment Capabilities:

• Manual Overdial Parameter Adjustment: Temporarily adjust dialing frequency during task runtime.

• Call Pause/Resume: Immediately pause or resume task dialing.

• Agent Group Adjustment: Dynamically adjust the range of agents participating in the task.

• Batch Priority Adjustment: Modify the execution order of batches awaiting dialing.

📊 Data Export and Reports

Export Configuration Options

Operation Entry: After completing data filtering, click the "Export" button at the top of the list page.

Export Content Selection:

• Basic Call Records: Contains basic information for all calls.

• Detailed Call Logs: Contains detailed processes like status transitions, agent assignments.

• Agent Performance Data: Individual agent performance statistics.

• Task Statistical Reports: Aggregated data at the task level.

Report Type Explanation

Task-level Reports:

• Predictive Outbound Task Report: Analysis of overall task running effectiveness.

• Time Period Analysis Report: Trend analysis by hour/day granularity.

• Batch Effectiveness Report: Comparison of dialing effectiveness across number batches.

Agent-level Reports:

• Predictive Outbound Agent Report: Individual agent performance statistics.

• Answer Timeliness Report: Analysis of agent response speed.

• Call Quality Report: Analysis of talk time and effectiveness.

🛠️ Monitoring Configuration Suggestions

Key Monitoring Threshold Settings

5. In-depth Analysis of Predictive Outbound Algorithm

What is the Predictive Outbound Algorithm?

The predictive outbound algorithm is an intelligent call scheduling technology. By dynamically adjusting the number of calls per cycle, it achieves the goal of transferring answered customers to idle agents immediately, while maximizing agent utilization and controlling the nuisance rate. This algorithm is based on the following core idea:

By accurately predicting the number of idle agents and dynamically adjusting the dialing pace, it achieves the perfect match of "an agent is ready when the customer answers".

Three Major Pain Points of Traditional Outbound and Our Solutions

Quantified Business Value Brought by the Algorithm

graph TD
    A[Your Business Goal] --> B{Algorithm Intelligent Matching}
    
    B --> C[Increase Sales Conversion]
    B --> D[Reduce Operational Cost]
    B --> E[Optimize Customer Experience]
    
    C --> F[Agent Effective Talk Time<br/>From 30min→50min/hour]
    D --> G[Agents Needed for Same Volume<br/>Reduced by 40%]
    E --> H[Customer Wait Time<br/>From 15s→Under 3s]
    
    F --> I[Performance Increase<br/>+200%+]
    G --> J[Cost Savings<br/>-35%+]
    H --> K[Satisfaction Increase<br/>+25%+]
    
    style A fill:#ff6b6b,color:#fff
    style B fill:#1dd1a1,color:#fff
    style I fill:#54a0ff,color:#fff
    style J fill:#54a0ff,color:#fff
    style K fill:#54a0ff,color:#fff

Algorithm Core Advantage: Intelligent Closed-loop Workflow

Three Core Advantages Overview

graph TD
    A[Input Your Business Needs] --> B[Intelligent Algorithm Engine]
    
    B --> C[Advantage 1: Precise Resource Matching]
    B --> D[Advantage 2: Dynamic Risk Control]
    B --> E[Advantage 3: Continuous Self-learning Optimization]
    
    C --> F[Zero Agent Idle Time<br/>Zero Customer Wait Time]
    D --> G[Nuisance Rate Auto-managed<br/>Compliance Worry-free]
    E --> H[Gets Smarter with Use<br/>Effect Continuously Improves]
    
    F --> I[Maximized Operational Efficiency]
    G --> I
    H --> I
    
    I --> J[Your Business Goal Achieved]
    
    style B fill:#1dd1a1,color:#fff
    style C fill:#54a0ff,color:#fff
    style D fill:#54a0ff,color:#fff
    style E fill:#54a0ff,color:#fff
    style J fill:#ff6b6b,color:#fff

Advantage 1: Precise Resource Matching, Say Goodbye to Uneven Workloads

Traditional Problem: Agents are either waiting or overwhelmed.

Our Solution:

graph LR
    A[Intelligent Sensing System] --> B[Real-time Data Collection]
    B --> C[Agent Status]
    B --> D[Call Rhythm]
    B --> E[Customer Answer Rate]
    
    C --> F[Intelligent Scheduling Center]
    D --> F
    E --> F
    
    F --> G[Precise Calculation]
    G --> H[Dynamic Dialing Frequency Adjustment]
    
    H --> I[Result: Agents Maintain Optimal Work Rhythm]
    
    style A fill:#ff6b6b,color:#fff
    style F fill:#1dd1a1,color:#fff
    style I fill:#54a0ff,color:#fff

Your Actual Gain:

• ✅ Agent Utilization: Increased from 30-50% to 75-95%.

• ✅ Effective Talk Time: Increased by 20-30 minutes per hour.

• ✅ Labor Cost: Same business volume requires 40% fewer agents.

Advantage 2: Dynamic Risk Control, Ensuring Compliant Operation

Intelligent Protection System:

graph TD
    A[Outbound Execution] --> B[Real-time Monitoring Layer]
    
    B --> C{Nuisance Rate Detection}
    C -->|Normal Range| D[Continue Execution]
    C -->|Approaching Threshold| E[Alert & Fine-tune]
    C -->|Exceeds Threshold| F[Automatic Circuit Breaker Protection]
    
    B --> G{Agent Status Detection}
    G -->|Normal| H[Optimize Assignment]
    G -->|Abnormal| I[Automatically Isolate]
    
    D --> J[Smooth Operation]
    E --> J
    F --> K[Protects Your Compliance Reputation]
    H --> J
    I --> K
    
    style B fill:#1dd1a1,color:#fff
    style F fill:#ff6b6b,color:#fff
    style K fill:#54a0ff,color:#fff

Your Security Assurance:

• ✅ Nuisance Rate Auto-control: Always maintained in the 1-3% safe range.

• ✅ Compliance Risk Alert: Proactively discovers potential issues.

• ✅ Automatic Circuit Breaker: Automatically protects your company reputation during anomalies.

Advantage 3: Continuous Self-learning Optimization, Gets Smarter with Use

Algorithm Evolution Capability:

graph LR
    A[First Use] --> B[Base Algorithm Starts]
    B --> C[Execute Task]
    C --> D[Collect Effectiveness Data]
    
    D --> E{Intelligent Learning Engine}
    E --> F[Learn Your Business Characteristics]
    E --> G[Learn Your Customer Habits]
    E --> H[Learn Your Agent Efficiency]
    
    F --> I[Optimize Parameter Configuration]
    G --> I
    H --> I
    
    I --> J[Next Execution More Precise]
    J --> C
    
    style E fill:#1dd1a1,color:#fff
    style I fill:#54a0ff,color:#fff

Your Long-term Value:

• ✅ No Manual Tuning Needed: System automatically adapts to your business changes.

• ✅ Continuous Effectiveness Improvement: The longer you use it, the more accurate the matching.

• ✅ Personalized Adaptation: Algorithm deeply understands your unique business model.

Intelligent Scenario Adaptation: How the Algorithm Serves Your Specific Business

Multiple Business Scenarios, One Intelligent Solution

graph TD
    A[Your Business Scenario] --> B{Choose Corresponding Mode}
    
    B --> C[High-value Customer Service]
    B --> D[Marketing Campaigns]
    B --> E[Daily Customer Follow-up]
    B --> F[Urgent Notification Calls]
    
    C --> G[Algorithm Mode: Quality First]
    D --> H[Algorithm Mode: Efficiency First]
    E --> I[Algorithm Mode: Balanced & Stable]
    F --> J[Algorithm Mode: Rapid Response]
    
    G --> K[Ensure VIP Experience<br/>Strict Nuisance Control]
    H --> L[Maximize Reach<br/>Reasonable Risk Control]
    I --> M[Stable & Reliable Execution<br/>Continuous Optimization]
    J --> N[Quick Target Reach<br/>Intelligent Time Selection]
    
    style B fill:#1dd1a1,color:#fff
    style K fill:#54a0ff,color:#fff
    style L fill:#54a0ff,color:#fff
    style M fill:#54a0ff,color:#fff
    style N fill:#54a0ff,color:#fff

Actual Effectiveness Comparison: Traditional vs. Intelligent

Usability Design: Intelligent Algorithm with Zero Technical Barrier

Three Steps to Enable the Intelligent Algorithm

graph TD
    A[Step 1: Choose Task Type] --> B[Select Pre-set Mode Based on Business]
    B --> C[Step 2: Set Basic Parameters]
    C --> D[Enter Your Business Expected Goals]
    D --> E[Step 3: Start Task]
    E --> F[Algorithm Automatically Takes Over & Optimizes]
    
    F --> G[You Just Focus on Business Results]
    
    style A fill:#ff6b6b,color:#fff
    style C fill:#1dd1a1,color:#fff
    style E fill:#54a0ff,color:#fff
    style G fill:#ff9ff3,color:#000

Intelligent Parameter Recommendation System

No Need to Be an Algorithm Expert:

• ✅ Scenario-based Presets: Recommends optimal configurations based on your business type.

• ✅ One-click Application: Takes effect upon selection, no complex adjustments needed.

• ✅ Safety Boundaries: All parameters adjust automatically within safe ranges.

Visual Effectiveness Monitoring

Real-time Understanding of Algorithm Work Status:

graph LR
    A[Monitoring Dashboard] --> B[Real-time Efficiency Dashboard]
    A --> C[Risk Alert Panel]
    A --> D[Effectiveness Trend Charts]
    
    B --> E[Agent Utilization<br/>75%-95%]
    C --> F[Nuisance Rate<br/>1%-3%]
    D --> G[Continuous Optimization Curve]
    
    style A fill:#1dd1a1,color:#fff
    style E fill:#54a0ff,color:#fff
    style F fill:#54a0ff,color:#fff
    style G fill:#54a0ff,color:#fff

Effectiveness Guarantee and Service System

Algorithm Effectiveness Expectation Goals

Algorithm Value Summary

The intelligent predictive outbound algorithm is not a complex technical concept, but a practical tool for your business growth:

What Does It Mean for You?

6. Configuration Notes

🔧 Key Configuration Item Checklist

🎯 Start/Stop Configuration Selection Guide

📊 Call Status Analysis Guide

🚨 Common Issue Handling

Task Cannot Start:

• Check if the scheduled start time is within the dialing time.

• Confirm if the current time is within a blocked time period.

• Verify if the task status is normal.

Task Stops Abnormally:

• Check scheduled end time configuration.

• Confirm if automatic pause conditions were met.

• View task status description information.

Call Status Abnormal:

• Check business logic corresponding to enum values.

• Confirm if status statistics are accurate.

• Verify if nuisance rate calculation is correct.

Caller ID Not Effective:

• Check number's predictive outbound permissions.

• Confirm API parameter priority.

• Verify if number status is normal.

7. Full Process Verification Checklist

8. Typical Business Scenario Configuration Guide

This chapter details the configuration methods for predictive outbound tasks under different business needs based on actual scenarios, helping you get started quickly and configure various outbound tasks correctly.

Scenario 1: One-time Immediate Outbound Task

🎯 Business Expectation

Create a task, import data immediately, and start manually. The task automatically executes outbound calls during specified working hours and ends naturally after data is dialed, requiring no manual intervention. Suitable for urgent outbound tasks requiring immediate execution.

📋 Business Expectation Example Explanation

Application Scenario: A company needs to send time-limited promotional notifications and must start outbound calls immediately during working hours on the same day to notify all target customers before the campaign starts.

Specific Parameters:

• Task working hours: 10:00-12:00, 14:00-18:00.

• Current time: 11:00 (within the working time period).

• Goal: Start outbound calls now.

• Action: Complete task creation, after importing data, manually start the task immediately.

⚠️ Core Points Explanation

1. Start Time vs. Execution Time Separation:

   • A task can start successfully at any time, but actual outbound calls are only made during configured working hours.

   • In the example, starting at 11:00, the system immediately begins outbound calls (because 11:00 is within 10:00~12:00 working hours).

2. One-time Task Characteristic:

   • Tasks are designed for single use, archived after data is dialed.

   • Facilitates data statistics and effectiveness analysis.

   • Recommended to use independent tasks for each campaign for traceability and management.

3. Incomplete Dialing Handling for the Day:

   • If imported data is not fully dialed within the day's working hours, the task remains active.

   • It will automatically continue dialing remaining numbers during the next day's working hours (starting at 10:00).

   • Manual ending is required to stop.

⚙️ Configuration Parameters

📋 Operation Steps

Step 1: Create Task Basic Configuration

graph TD
    A[New Predictive Outbound Task] --> B[Basic Settings]
    B --> C[Task Name: Immediate Outbound Task]
    B --> D[Nuisance Rate Coefficient: 3]
    B --> E[Redial Strategy: Configure as needed]
    
    A --> F[Agent-side Configuration]
    F --> G[Call Flow Mode: Direct to Agent]
    F --> H[IVR: Configure overflow IVR]
    F --> I[Agent List: Select participating agents]
    
    A --> J[Customer-side Configuration]
    J --> K[Caller ID: Configure caller ID]
    J --> L[Dialing Order: Sequential by priority]
    
    style A fill:#ff6b6b,color:#fff
    style B fill:#54a0ff,color:#fff
    style F fill:#54a0ff,color:#fff
    style J fill:#54a0ff,color:#fff

Step 2: Configure Working Hours

1. Go to 'Basic Settings → Dialing Time'.

2. Add time period: 10:00-12:00.

3. Add time period: 14:00-18:00.

4. Ensure no overlap between periods and coverage of required working hours.

Step 3: Import Data and Start

1. Click the "Import" button to upload number data.

2. Configure import parameters:

   • Deduplication Rule: Deduplicate within task.

   • Start Type: Do not auto-start.

   • Batch Name: Immediate Outbound Batch.

3. After upload completes, click "Start Task".

🔍 Monitoring Focus

• Real-time Task Progress: Monitor changes in task completion rate.

• Execution in Working Hours: Confirm normal outbound calls during specified periods.

• Nuisance Rate Control: Ensure within acceptable range.

Scenario 2: One-time Scheduled Outbound Task

🎯 Business Expectation

Create a task, import data immediately, but set a scheduled start for a future specific date. Suitable for planned outbound campaigns. Businesses can prepare in advance; the system executes automatically at the scheduled time.

📋 Business Expectation Example Explanation

Application Scenario: A financial institution plans a new product promotion outbound next Monday. They need to configure the task on Friday, setting it to start automatically next Monday, avoiding disturbing customers on weekends or non-working hours.

Specific Parameters:

• Task working hours: 10:00-12:00, 14:00-18:00.

• Current time: 2025/9/24 16:00 (within working hours).

• Goal: Planned task start time: 2025/9/25.

• Action: Complete task creation, after importing data, set task start time to 2025/9/25.

⚠️ Core Points Explanation

1. Creation vs. Execution Time Separation:

   • Creation time is at current working hours (16:00), but execution is set for the future.

   • The system does not execute outbound calls immediately; starts as per schedule.

   • Suitable for outbound campaigns planned and prepared in advance.

2. Schedule Accuracy:

   • The scheduled start time must fall within the dialing time range.

   • Recommended to set slightly later than dialing start time (e.g., 10:05).

   • Avoids start anomalies due to timing precision issues.

3. Task Status Monitoring:

   • After creation, task status is "Waiting".

   • Automatically changes to "Running" when scheduled start time arrives.

   • Can view preparation status via the task monitoring page.

⚙️ Configuration Parameters

📋 Operation Steps

Step 1: Scheduled Task Configuration

graph TD
    A[New Predictive Outbound Task] --> B[Basic Settings]
    B --> C[Start Method: Scheduled Start]
    C --> D[Start Time: 2025/9/25 10:00]
    B --> E[End Method: Scheduled End]
    E --> F[End Time: 2025/9/25 18:00]
    B --> G[Dialing Time: 10:00-12:00, 14:00-18:00]
    
    style A fill:#ff6b6b,color:#fff
    style C fill:#54a0ff,color:#fff
    style E fill:#54a0ff,color:#fff

Step 2: Data Import Configuration

1. Click the "Import" button to upload number data.

2. Configure import parameters:

   • Deduplication Rule: Deduplicate within task.

   • Start Type: Do not auto-start.

   • Batch Name: Scheduled Outbound Batch.

3. Confirm import; task enters waiting state.

🔍 Monitoring Focus

• Task Waiting Status: Confirm scheduled configuration is correct.

• Automatic Start Execution: Observe if scheduled start triggers normally.

• Planned Execution Effectiveness: Evaluate reasonableness of planned time.

Scenario 3: Reusable Daily Scheduled Data Import

🎯 Business Expectation

Create a predictive outbound task for long-term use. Continuously import data during specified time cycles; the task automatically executes outbound calls during daily working hours. Suitable for ongoing business like daily customer service, regular follow-ups.

📊 Example Configuration Comparison

⚠️ Core Points Explanation

1. Task Continuity:

   • Tasks designed for long-term use, support continuous import of new data.

   • System automatically manages data priority and dialing order.

   • Suitable for daily operations and customer service scenarios.

2. Automatic Execution Logic:

   • Current date is within task cycle.

   • Current time is within daily working hours.

   • Task has data pending dialing.

   • Outbound calls executed automatically only when all three conditions are met.

3. Data Management Strategy:

   • Supports mixed dialing of multiple batches of data.

   • Can set data priority via API import.

   • Supports distinguishing data sources with different batch names.

⚙️ Unified Configuration Scheme

Core Configuration Parameters

📋 Standard Operation Process

Step 1: Long-term Task Basic Configuration

graph TD
    A[New Predictive Outbound Task] --> B[Basic Settings]
    B --> C[Start Method: Scheduled Start]
    C --> D[Start Time: Cycle Start Date]
    B --> E[End Method: Scheduled End]
    E --> F[End Time: Cycle End Date]
    B --> G[Dialing Time: Configure working hours]
    
    A --> H[Advanced Settings]
    H --> I[Task Attribute: Long-term Task]
    H --> J[Dialing Frequency: 100]
    
    style A fill:#ff6b6b,color:#fff
    style B fill:#54a0ff,color:#fff
    style H fill:#54a0ff,color:#fff

Step 2: Data Management Strategy

1. Initial Data Import: Import initial data immediately after task creation.

2. Continuous Data Appending: Import new data irregularly during the task cycle.

3. Batch Management: Use different batch names to distinguish data sources and times.

4. Priority Setting: Set data priority via API import.

🔍 Monitoring Focus

• Cycle Validity: Confirm task is within the set cycle.

• Data Continuity: Monitor data import and consumption.

• Execution Stability: Long-term task runtime stability.

Scenario 4: Time-sensitive Data Outbound Task

🎯 Business Expectation

Outbound data has clear timeliness; continuing dialing beyond a specified time point loses business meaning. Strict end time control is required. Suitable for businesses with time requirements like limited-time campaigns, appointment reminders.

⚠️ Core Points Explanation

1. Timeliness Control:

   • Set strict scheduled end time.

   • End time slightly earlier than actual data expiration time.

   • Supports data-level timeliness control.

2. Cross-scenario Applicability:

   • Can be combined with Scenarios 1, 2, 3.

   • Supports task-level + data-level dual control.

   • Flexibly adapts to different business needs.

⚙️ Configuration Scheme

Scheme One: Strict Timeliness Control (Recommended)

📋 Strict Timeliness Configuration Process

Step 1: Timeliness Task Configuration

graph TD
    A[New Predictive Outbound Task] --> B[Basic Settings]
    B --> C[Start Method: Scheduled Start]
    C --> D[Start Time: Start as early as possible]
    B --> E[End Method: Scheduled End]
    E --> F[End Time: Data expiration time]
    B --> G[Dialing Time: Valid working hours]
    
    A --> H[Advanced Settings]
    H --> I[Task Attribute: Time-sensitive Task]
    H --> J[Dialing Frequency: Appropriately increase]
    
    style A fill:#ff6b6b,color:#fff
    style F fill:#ff6b6b,color:#fff

Step 2: Data Timeliness Management

1. Data Preprocessing: Ensure imported data is within validity period.

2. Batch Identification: Use batch name to identify data timeliness.

3. Priority Setting: Increase priority for data nearing expiration.

4. Monitoring Alert: Set completion progress alerts.

🔍 Monitoring Focus

• Timeliness Progress: Monitor completion before expiration.

• Alert Mechanism: Set alerts for lagging progress.

• Exception Handling: Automatic pause for timed-out data.

Scenario Configuration Summary

🎯 Scenario Selection Guide

⚙️ Configuration Checklist

Basic Configuration Check

• Task name and description are clear and explicit.

• Call flow mode matches business needs.

• Agent configuration is complete and online.

• Caller ID configuration is correct.

Time Configuration Check

• Start method matches business scenario.

• End method setting is reasonable.

• Dialing time covers business needs.

• No conflicts in time parameters.

Data Configuration Check

• Import data format is correct.

• Deduplication rule setting is appropriate.

• Batch management is clear.

• Priority configuration is reasonable.

🔧 Troubleshooting Guide

Task Cannot Start

1. Check if current time is within dialing time.

2. Confirm scheduled start time is set correctly.

3. Verify if task status is normal.

4. Check if there is data pending dialing.

Task Stops Abnormally

1. Check if scheduled end time has been reached.

2. Check if automatic pause conditions were triggered.

3. Confirm if nuisance rate exceeded limit.

4. Verify agent availability.

Through the above scenario-based configuration guides, you can quickly configure suitable predictive outbound tasks based on actual business needs, ensuring outbound effectiveness is maximized.

9. Quick Reference Summary