Primary Use Case: Lead Scoring & Assignment

Background Overview

With the surge in lead volume, automated scoring and assignment mechanisms are needed to prioritize high-potential leads and assign them to appropriate sales teams. PowerX CRM provides scoring models, assignment rules, anomaly detection, and fairness analysis. This primary use case focuses on "Lead Scoring & Assignment" to help teams improve lead response efficiency.

Objectives & Value

• Scoring Modeling: Evaluate lead hotness by combining behavior, attributes, and interaction dimensions.
• Smart Assignment: Match best owners by region, industry, and product.
• Fairness Guarantee: Avoid individual overload or regional bias.
• Anomaly Detection: Identify abnormally high or low-scoring leads; continuously optimize model.
• Traceability: Scoring and assignment history logging to support review.

Participating Roles

• Marketing Operations: Maintains scoring rules and monitors results.
• Sales Manager: Sets assignment strategy and monitors team workload.
• Sales Rep: Receives leads and executes follow-up.
• Data Analyst/Agent: Trains models and evaluates performance.

Primary Scenario User Story

As a marketing operations staff, I want to automatically identify high-value leads and assign them fairly, so that I can improve conversion efficiency and ensure fairness.

Sub-scenarios Detailed

Sub-scenario A: Scoring Model Training & Application

• Roles & Triggers: Model calculates lead hotness score in real-time based on download, open rates, in-site behavior, and other metrics.
• Main Process:
  1. Regularly train scoring model with historical conversion data.
  2. Calculate new lead scores in real-time; classify as cold/warm/hot.
  3. Write scores to lead profiles; drive journey or assignment strategies.
  4. Monitor model performance; adjust weights or features.
• Success Criteria: High correlation between scoring and conversion; timely updates; strong interpretability.
• Exception & Risk Control: Model drift alerts; blacklist filtering; human intervention mechanism.
• Indicators: Scoring accuracy rate, model update frequency, feature contribution.

Sub-scenario B: Auto-assignment & Priority Lists

• Roles & Triggers: High-scoring leads auto-assigned to corresponding sales teams; generate priority follow-up lists.
• Main Process:
  1. Select owner based on assignment rules (region, industry, product).
  2. System generates priority list with follow-up suggestions and SLA.
  3. Sales views leads in todo list; records first contact results.
  4. Lead response fed back to scoring model.
• Success Criteria: Accurate assignment; timely lists; improved response rate.
• Exception & Risk Control: Auto-skip resigned/overloaded staff; rotation strategy ensures fairness; retry on assignment failure.
• Indicators: Assignment success rate, first response time, lead conversion rate.

Sub-scenario C: Anomaly Scoring Review

• Roles & Triggers: Leads with abnormal high or low scores trigger model evaluation; team can adjust weights and scoring rules.
• Main Process:
  1. System monitors score distribution; alerts on anomaly peaks.
  2. Marketing operations or analyst checks lead details; judges if model anomaly.
  3. Adjust model parameters or mark leads as black/white list when needed.
  4. Record adjustment reasons; form model governance log.
• Success Criteria: Accurate anomaly identification; timely adjustments; model stability.
• Exception & Risk Control: Misjudgment can be rolled back; permission control; audit logging.
• Indicators: Anomaly lead proportion, model adjustment count, scoring stability.

Sub-scenario D: Assignment History Review & Fairness Analysis

• Roles & Triggers: Scoring and assignment history reviewable; helps marketing and sales review scoring strategy effectiveness.
• Main Process:
  1. System records every assignment's lead, owner, time, and reason.
  2. Provide visualization reports analyzing personnel workload and conversion performance.
  3. Identify unequal distribution; give optimization suggestions (adjust rules, add alternatives).
  4. Review meeting outputs conclusions; optimize and version rules.
• Success Criteria: Complete data; improved fairness; reviewable conclusions.
• Exception & Risk Control: Data missing alerts; report permissions; conclusions require approval to take effect.
• Indicators: Assignment fairness index, review execution rate, rule change impact.

Scenario-level Test Cases

Test Preparation: Enable scoring model, auto-assignment, anomaly detection, and fairness reports. Prepare historical conversion data, feature weight configuration, regional sales teams, and model monitoring thresholds.

Use Case A-1: Scoring Model Real-time Calculation (Positive)

• Preconditions: Model trained; new lead includes downloading whitepaper and attending webinar behaviors.
• Steps:
  1. Import new lead and trigger scoring.
  2. Check lead profile score.
• Expected Results:
  • Score calculated with feature contribution displayed (download 40%, webinar 30%, etc.).
  • Lead classified as "hot" and enters high-priority queue.
  • Model monitoring panel updates real-time average score.

Use Case B-1: Auto-assignment & SLA (Positive)

• Preconditions: East China team rotation enabled; lead score 90.
• Steps:
  1. Trigger assignment engine.
  2. Check lead owner and todo.
• Expected Results:
  • Lead assigned to next available sales; creates 2-hour contact task.
  • Todo list displays lead source, score, and suggested scripts.
  • Sales feedback updates lead status and model training set.

Use Case C-1: Anomaly High Score Review (Negative)

• Preconditions: Lead incorrectly scored 98 due to bad data; anomaly threshold set to 95.
• Steps:
  1. Import lead with abnormal field.
  2. Observe anomaly monitoring list.
• Expected Results:
  • System marks lead as "anomaly high score"; auto-creates review task.
  • Reviewer can adjust score or mark as false positive.
  • Handling result recorded in model governance log.

Use Case D-1: Assignment History Report (Positive)

• Preconditions: Past week generated 200 assignment records.
• Steps:
  1. Open fairness analysis report.
  2. Check assignment ratio and conversion rate by region and owner.
• Expected Results:
  • Report displays each owner's lead count and close rate; highlights anomalies.
  • Can export with comments to form review materials.
  • Suggest rotation strategy adjustment with simulation plan.

Use Case D-2: Rule Change Rollback (Negative)

• Preconditions: Operations mistakenly changed weights causing major scoring fluctuation.
• Steps:
  1. Within 1 hour execute "rollback to previous version".
• Expected Results:
  • Model restores to previous version weights; related logs recorded.
  • Leads with abnormal scores during period enter review queue.

---

Related Resources

Business Domain: Marketing Automation

• Lead Nurturing - Journey design, content delivery, and human intervention
• Campaign Orchestration & Management - Visual campaign design, budget control, and real-time monitoring
• Content & Channel Outreach - Template management, multi-language support, and channel scheduling
• Testing & Conversion Optimization - A/B testing, funnel analysis, and best practice promotion

Other CRM Business Domains

• Customer Management - Customer lifecycle, segmentation, and lead management
• Membership & Loyalty - Member tiers, points, and engagement
• Sales Process - Opportunity management, quoting, and sales activities
• Customer Success - Case management, service delivery, and renewal operations
• Analytics & Revenue Intelligence - Sales forecasting, customer value analysis, and performance settlement
• Admin & Integration - Access control, workflow automation, and system integration

---