Together's mentorship platform delivers a 98% satisfaction rate by combining user-centred design with powerful, customizable features. Most of these match processes make use of automatching under the hood, so we recommend you read about that too.

You can choose from one of the following matching processes for your program.

 

Matching processes.png

 

User-led Matching with one-sided approval

Users choose their own match by selecting from a list of eligible matches. To improve decision-making, a user's top match, based on your algorithm rules, is featured. However, users can still browse through all eligible matches in the directory, sorted by match score.  

One-sided approval refers to the fact that users do not need the other party's approval to start the match. For example, in mentorship matching, mentees do not need the mentor's approval before the match starts. In colleague connect matching, colleagues do not need the other colleague's approval before the match starts. 

One-sided approval has the advantage of being simpler and faster to get everyone matched, and in mentorship matching, focuses on the needs of the mentee. 

It is acceptable not to have approval from both parties for the following reasons:

  1. You can give control of the matching algorithm to end users, allowing them to set their preferences up front about who they want to match with. 
  2. Users can pause their matching to stop getting new matches.

We typically find that mentors are willing to help any mentee, as long as they are currently available for new matches, so a two-sided approval is not required.

Note: User-led matching is available in both Cohort and Evergreen duration programs. 

Note: A user's top match may not have the highest match score of all their eligible matches. This is because we avoid advertising the same top match to everyone, and only recommend top matches as many times as they have room for matches. 

Note: This process is first-come, first-served, however, the featured top match is different for everyone, improving fairness. 

 

User-led Matching with two-sided approval

Users choose their own match by selecting from a list of eligible matches. Users browse through all eligible matches in the directory, sorted by match score.  

Two-sided approval refers to the fact that users need the other party's approval to start the match. For example, in mentorship matching, mentees will need the mentor's approval before the match starts. 

Since mentors might become picky and reject mentees in this process, this matching process introduces the concept of a shortlist. With two-sided approval, users must submit a shortlist of up to 5 top choices, so that if the first choice declines the match, they still have a high likelihood of being matched with one of the other 4 matches on their list.

To illustrate this, assume mentors decline a mentee 20% of the time. Without shortlists, one in five mentees will have to go back and choose another mentor, potentially getting discouraged. With a shortlist, the odds of being rejected are 20% x 20% x 20% x 20% x 20%, which is significantly less than 1% odds. 

Two-sided approval takes longer and is not optimized for the interest of mentees or getting the most matches. 

Note: User-led matching is available in both Cohort and Evergreen duration programs. 

Note: This process is first-come, first-served. 

 

Admin-led Matching

Admins create and approve matches with the help of the algorithm. You can either match users one by one or run automatching to generate the best set of matches all at once. A hybrid approach involves matching a few priority matches one-by-one, then running automatching on the remainder.

Because neither party in the match is involved in the decision-making process, we recommend using admin-led matching only in cases where you are familiar with the participants and can review the matches suggested, or in cases where you want to place strong faith in your algorithm rules.  

Note: Admin-led matching is available in Cohort programs only.

 

Roulette

The system creates matches with the help of the algorithm and announces them without any user approval. In Evergreen programs, roulette announces new matches on a regular cadence, making it suitable for short, informal matches lasting one session. Users can book additional sessions if they want to. 

In this format, admins set the day of the week that new matches are announced (such as Monday) and the frequency of new matches (such as biweekly). 

When the next matches are announced, users' previous matches are automatically ended, unless they have an upcoming scheduled session that hasn't been completed. These users do not get a new match until that session is finished. 

If you want a hands-off format that also avoids possible double matching (where users both choose a match and get selected by someone else to be their match), this is a great format. 

Note: Roulette matching is available in Colleague Connect, Evergreen programs only.

 

Have more questions? Submit a request here and let us know how we can help!

Share

Was this article helpful?

0 out of 0 found this helpful
Have more questions? Submit a request

Comments (0 comments)

Please sign in to leave a comment.