In this article, we will cover how Match Scores are calculated, why they are important, and how you can make the most out of them.
What is a Match Score?
A Match Score is a percentage value generated by our matching algorithm. It measures the compatibility between a potential mentor and mentee based on the rules you set in your algorithm and how the mentee and mentee have filled out their questionnaire or what admin-entered data is attached to them. The score ranges from 0 to 100, with 100 indicating a perfect match.
How is a Match Score Calculated?
Match Scores are calculated by taking into account the profile information of mentors and mentees for any fields (such as Department, Skills, Goals, and Location) where an admin has set up an algorithm rule.
In your algorithm, you can choose the matching strength for any profile field. You can choose between "a must", "high", "medium", "low" or "custom" priority. Additionally, if you are using the match type "Let users indicate their preference in match" they will be able to select their own match strength for those particular fields. To learn more about how users can set their match preferences, you can check out this article: How to Update Your Match Preferences.
Any algorithm rules whose matching strength is "high", "medium", "low" or "custom" contribute to the match score. A "must" does not since it acts as a filter.
First, a raw score is calculated. The raw score is based on the following weighting:
- High priority: 5 points per answer overlap
- Medium priority: 3 points per answer overlap
- Low priority: 1 point per answer overlap
- Custom priority: N points per answer overlap
Imagine that a mentee has the following answers in their profile:
Imagine the algorithm in the program is set up such that matches for the Soft Skills field have a medium priority. In this example, the users will be able to set their match strength for Soft Skills:
Every time an answer overlaps between the mentee and mentor for Soft Skills, the raw match score gets an additional amount of points based on the match strength set by the user. In the screenshot below, the mentor and mentee overlap twice on Soft Skills and once on another field about their Seniority.
After the raw score has been calculated, an adjustment is made based on the maximum possible score attainable with your algorithm. Based on that, a percentage is calculated.
Where are match scores displayed?
Match Scores below 50% are not shown to mentors and mentees, only admins. Also, fields that are visible only to admins will not be visible as contributors to match scores for mentors and mentees, only admins. Match scores are displayed in the following places:
- Admin View: when manually matching two users
- Admin View: when hovering over match scores in staged or active matches
- User View: when hovering over the match score in the directory card. Top overall matches have a score greater than 75%.
- User View: when clicking on a mentor profile
How can I adjust match scores?
You can either:
- Add questions and/or algorithm rules with high/medium/low strength
- Change the strength of rules
- Reduce the number of options users can select for multiple-choice questions
- Recruit other mentors or mentees
Don't Rely Solely on Scores: While Match Scores are a great tool, remember they are not the only determinant of successful mentorship. Communication, commitment, and past feedback are also determinants.