Skip to content

Mentor-Mentee Matching Network

Run Mentor-Mentee Matching Network Fullscreen

You can include this MicroSim on your website using the following iframe:

1
<iframe src="https://dmccreary.github.io/organizational-analytics/sims/mentor-matching-network/main.html" height="502px" width="100%" scrolling="no"></iframe>

Description

This bipartite graph visualization shows the mentor-mentee matching process from the perspective of a graph database. The mentee (Priya Sharma, a Junior Data Analyst in Marketing Analytics) appears on the left, five candidate mentors appear on the right, and shared skill nodes form a bridge in the center. Pairing scores are computed from skill similarity, cross-departmental reach, and shared project history -- all factors that a graph-based matching algorithm would traverse.

How to use:

  • Hover over a mentor candidate to highlight the skills they share with Priya and see a detailed score breakdown tooltip
  • Click a skill diamond to highlight all people connected to that skill
  • Toggle Show Growth Skills to reveal skills that mentors have but Priya does not yet -- these represent learning opportunities
  • Adjust the Cross-Dept Weight slider to see how weighting cross-department exposure changes the pairing recommendations in real time
  • The gold ring highlights the best overall match based on current weights
  • Dashed amber lines connect the mentee to each candidate, with thickness proportional to the pairing score

About This Simulation

Mentor-mentee matching is a natural graph problem. In a labeled property graph, employees are nodes with skill, department, and tenure properties; HAS_SKILL edges link people to their competencies; and WORKS_WITH or COLLABORATES_ON edges capture network proximity. A matching algorithm traverses these relationships to find candidates who maximize skill overlap (for coaching relevance), cross-departmental reach (for career breadth), and shared project context (for trust and rapport).

The scoring model in this simulation combines three factors:

Factor Weight Source Purpose
Skill Similarity Jaccard-like overlap Ensures the mentor can coach relevant skills
Cross-Department Bonus Slider-adjustable Rewards mentors from different departments for broader exposure
Shared Project Bonus Number of past collaborations Reflects pre-existing trust and communication patterns

By adjusting the cross-department weight slider, students can observe how different organizational priorities -- deep specialization versus broad network exposure -- shift which mentor rises to the top.

Lesson Plan

Learning Objective: Students will assess the quality of mentor-mentee pairings by examining skill similarity, network proximity, and cross-departmental reach within the organizational graph.

Bloom's Level: Evaluate (L5)

Activities:

  1. With the slider at the default (40%), identify the best match and explain why that mentor scores highest by examining the shared skills and score breakdown
  2. Move the slider to 0% (pure similarity) and then to 100% (maximum cross-department weighting) -- describe how the best match changes and what each extreme implies for the mentee's development
  3. Toggle "Show Growth Skills" and identify which mentor offers the most growth skills -- discuss whether growth potential should factor into the matching algorithm
  4. Click individual skill nodes and observe which mentors share that skill -- propose a Cypher query that would find all employees who share a given skill with a target mentee

Assessment: Students design a scoring function for a different matching scenario (e.g., project team assembly or succession planning) and explain which graph relationships their function would traverse and why.

References

  1. Mentoring - Wikipedia - Overview of mentoring relationships and organizational mentoring programs
  2. Bipartite graph - Wikipedia - The graph structure used to model matching problems
  3. Jaccard index - Wikipedia - The similarity measure underlying skill overlap scoring