The matching algorithm compares responses to profile fields between mentees and mentors. Fields that will be used in the algorithm must follow the guidelines listed below.

Guidelines for matching algorithm profile fields:

*Text Entry questions:

Text entry questions may be used for matching but are only truly effective if the terminology used is consistent across profiles. Therefore, an SFTP employee information import into a user field that is set to be 'Editable by administrator only' is needed. This way, the likelihood of incorrectly spelled terms, inconsistent capitalization, or the use of different terms that may refer to the same thing - but not recognized by the algorithm - can be avoided.

Also, Match Labels cannot be used for text entry questions, which means they will not show up on the summary card in self-match programs as shown below:

There are two 'Match Config' (configuration) options for each field that will make up the matching algorithm:

In 'Normal Matching', there is direct match between what the mentor and mentee select as their response to that field, for example:

This option is used in the case where the Mentor and Mentee have identical dropdown lists to choose from within their respective profile fields.

In 'Set Matching', admins can create more complex alignments between profile field responses. The screenshot below shows the field choices and variation of possible matches depending on the profile field response to the 'Job Level'. A mentee that is a Front-line Associate will have mentors available for matching that are either a Professional Individual Contributor or a Front-line Leader.

Conditional matching allows admins to set multiple conditions across several profile fields for matching.

Example 1:

In this example, a condition is set based on whether the mentee would like to meet with a mentor from the same site, a different site, or don't have a preference. Depending on the mentee's choice, set matching options under 'Then' will coincide:

If the mentee selects 'I prefer to be matched with someone from a different site', then the expanded section under 'Then', would contain matching sets with only different location profile fields among mentees and mentors:

Thresholds and Weights refer to the level of influence a specific match configuration will have within the overall output of a match score. A simple way to think of the difference between the two is:

A Threshold is a rule. The mentor and mentee criteria must be aligned within the configuration in order to be considered a match. If the pair does not meet the criteria, their match score will drop to a 0% and they cannot connect.

A Weight is a preference. The mentor and mentee may align within the configuration in order to be considered a match, but it is not required. If the pair does not meet this criteria, their match score will drop The weight will define how steeply this score will drop.

Threshold - Consider Mentor as not a match if field score is:

If the threshold is set as less than or greater than the threshold value chosen, a match cannot be made, even if the pair is a good match based on the other criteria. If the threshold is not met, a mentee will not be able to request the mentor in a self-match program, and an administrator will not be able to match those participants in an admin-match program.

Examples of threshold use: Thresholds might be used to ensure that people with same or wide-ranging job levels don't match, or those that don't speak the same language don't match. Or that a match must contain a direct correlation between mentee and mentor such as the mentee's selected area of development and the mentor's area of expertise.

Weight - Factor by which the field score is multiplied and added to the total score

Examples of weighting: On a scale of 1-10 where 10 is the highest, rank the importance of each match configuration based on what you consider to be a quality match. This will equate to the weight score you see in this example below (e.g., if it's a 4 on a scale of 1-10, you would see a 0.4. If it's a 10, you would see a 1.0).

Goal: Increase career development by matching mentees and mentors within the same job function.

Weight: Job function alignment should have the highest matching weight.

Goal: Encourage mentees to connect with mentors that have a longer company tenure.

Weight: A job tenure match configuration might ideally match a mentee to mentor who has a longer company tenure, but a lower weight could be used if tenure is not necessarily considered a defining priority for a top match.

You can use negative weighting if you want to discourage matching on specific criteria. For example, if you want to promote but not require cross-departmental matching, you could use a negative weight for those that match on the same department. Simply select a negative value for the weight.