How to Start a Mentoring Program: A Step-by-Step Guide

What is a mentoring program?

A mentoring program is a structured arrangement in which a more experienced person supports the growth of a less experienced one over a fixed period, with a sponsor, matching rules and a way to check results. The structure is what separates it from the informal advice people already trade in hallways. It gives every participant the same entry point, the same expectations and the same end date.

Mentoring is not the same as coaching, onboarding or line management. A coach works on a specific skill or goal, usually in a short engagement. A buddy helps a newcomer settle in during the first weeks, as described in our onboarding buddy program guide.

A manager owns goals and performance, which is the subject of one-on-one meetings. A mentor takes the longer view of a career and answers questions the manager is not placed to answer.

Why start one, and why most programs stall

The evidence for formal mentoring is better than it was, but it is still uneven. Harvard Business Review reported in September 2022 that more than 70% of Fortune 500 companies offer some form of mentoring, and also noted that hard evidence of firm-level gains has been scarce. The same article argues that mentoring pays off when it is mandatory, because the people who most need a mentor tend to decline an optional offer.

A second finding comes from Frank Dobbin and Alexandra Kalev. In their 2016 Harvard Business Review article, based on data from 829 midsize and large U.S. firms, mentoring programs raised the representation of several groups of women and minority men in management by 9% to 24% on average. They explain this with sponsorship: mentors who invest in a protege come to believe the protege deserves the opportunities.

MENTOR, the U.S. National Mentoring Partnership, adds a warning from its practice standards. It says mentoring done haphazardly can be more harmful than having no mentor at all. That is the main reason to design the program before announcing it.

Step 1: Define the goal and the audience

Write one sentence that says what the program is for, and name who it serves. Typical goals are career growth for early-career staff, faster promotion readiness for underrepresented groups, knowledge transfer before senior people retire, or retention in a team with high turnover. Pick one primary goal for the first cohort. A program that tries to do all four usually serves none of them.

Then decide the audience. Options include all employees, a single function, new managers or a specific talent group. A narrow audience is easier to match well and easier to evaluate.

You can widen it once the first cohort has run.

• The business need the program addresses, in one sentence.
• The group of participants in the first cohort and a rough headcount.
• An executive sponsor who will speak about the program and protect time for it.
• A program owner who handles matching, check-ins and reporting.
• Two or three measures you will look at when the cohort ends.

Put these five items on a single page and get the sponsor to sign off before building anything else.

Step 2: Choose the mentoring model

Chronus, a mentoring software vendor, and the Forbes Business Council both list the same set of design choices: one-to-one, group or team mentoring, how people enroll, how long the program runs and how often pairs meet. Treat these as practitioner advice rather than research, but the checklist is sound.

Start with one model. One-to-one pairs are the simplest to explain and the easiest to measure. Group mentoring, where one mentor works with three to five mentees, suits a small pool of senior people.

Reverse mentoring, where a junior employee mentors a senior one on a specific topic, works as a second program once the first is stable.

• One-to-one: best for career development and sponsorship.
• Group: best when mentors are scarce.
• Peer: best for new managers facing the same problems at the same time.
• Reverse: best for digital skills or insight into a different generation or role.

Step 3: Decide who joins and whether it is voluntary

The Harvard Business Review argument for mandatory programs is worth taking seriously. If sign-up is optional, the people who volunteer are often those who already have strong networks, and the people who would gain most stay out. A common compromise is to make mentoring a default for a defined group, such as everyone promoted to first-time manager, while letting people opt out with a reason.

For mentors, voluntary participation is usually wiser. A mentor who was told to join and has no interest will deliver little. Ask for volunteers, say clearly how many hours a month the role takes, and make the contribution visible in performance conversations.

Recognition of mentor involvement is among the steps Chronus recommends for attracting participants.

Step 4: Write matching rules before you match anyone

Agree the rules for matching first, then apply them to the people. If you match first, the rules bend to fit the pairs you already made. Collect a short profile from each participant: goals, experience, interests, location and preferences about the kind of mentor they want.

Both Chronus and the Forbes Business Council describe matching as the hardest step. Their advice is to account for differences in background, career goals and learning styles. One more rule is worth adding: mentors should normally not be in the mentee's reporting line, so that the mentee can speak openly.

• Mentor is at least one level above the mentee, but not the mentee's manager.
• Pairs come from different teams unless the goal is knowledge transfer inside one team.
• Each mentor takes no more than two mentees at a time.
• The mentee names their top goal and the mentor names the topics they can help with.
• Either side can request a re-match without giving a reason after the first two meetings.

Let participants see two or three suggested matches and pick one. People commit more to a choice they helped make.

Step 5: Train mentors and mentees

Training does not need to be long. A 60 to 90 minute session for each group is enough to cover the program goal, what mentoring is and is not, confidentiality, how to run a first meeting and how to end a relationship well. The Forbes Business Council piece recommends separate sessions for mentors and mentees, which makes sense because their questions differ.

Give mentors a short list of good questions and a reminder to listen more than they talk. Give mentees a template for preparing each meeting, including one topic they want to discuss and one thing they tried since the last time. Our guide on how to give feedback is a useful companion for both sides.

Step 6: Run the first meetings and the middle months

The first meeting sets the relationship, so give pairs an agenda for it. They should agree on goals, how often they will meet, how they will communicate and what they will keep confidential. Most programs run six to twelve months with a meeting every two to four weeks. These numbers are common practice, not a rule, so pick a rhythm your participants can keep.

Check in with the program owner at about 30 days and at the midpoint. A two-question pulse survey is enough: is the pair meeting, and is it useful? Pairs that have stopped meeting are the early warning sign of a failing program.

Treat the launch as the start of the work, not the end of it.

• Meeting 1: introductions, goals, ground rules, schedule.
• Meetings 2 to 4: the mentee's main goal and a first action plan.
• Midpoint: review progress against the goal and adjust.
• Final meeting: reflect on what changed, agree what happens next.
• After the program: the pair can continue informally, or each person returns to the pool.

Step 7: Measure and decide what happens next

Pick measures that match the goal you wrote in step 1. If the goal was retention, look at how many participants stayed compared with a similar group. If it was promotion readiness, look at role moves and manager feedback.

If it was learning, ask mentees what they can do now that they could not before.

Chronus suggests tracking outcomes at three levels: the individual, the pair and the program. In practice that means participant surveys, the share of pairs still meeting at the midpoint and the share who finish. Be careful with comparisons.

People who join a program differ from those who do not, so a simple difference in retention does not prove the program caused it.

• Completion rate: pairs that finish the full period.
• Meeting rate: pairs that met at least every month.
• Mentee outcome: self-reported progress on the stated goal.
• Mentor outcome: willingness to mentor again.
• Program outcome: the measure you chose in step 1, read cautiously.

Do you need software?

A pilot of 10 to 30 pairs can run on a spreadsheet, a shared calendar and a survey form. Software becomes useful when matching takes the owner more than a few hours, when you need to report across several cohorts or when you run more than one program model. Typical features are profile collection, matching suggestions, scheduling, goal tracking and surveys.

If you do evaluate tools, look at how you can export your data, whether matching rules are visible and adjustable, and whether pricing is published. Our mentoring software category compares options, and the one-on-one meeting software and learning experience platforms categories cover nearby tools. Vendors can pay for more visibility; paid placements are labeled and explained in our disclosure.

No commission is earned on clicks or purchases, and we do not work as an affiliate.

Common mistakes to avoid

The most common failure is launching without a sponsor or owner. Without a named person, nobody notices when pairs stop meeting. The second is matching on convenience, such as pairing people who are already friends, which reproduces existing networks.

The third is leaving the end undefined, so relationships either fade awkwardly or run forever.

A fourth mistake is treating mentoring as a substitute for sponsorship or for fixing a poor promotion process. A mentor can advise, but a mentee still needs access to real opportunities. If the organization does not open those doors, a mentoring program will raise expectations it cannot meet.

Frequently asked questions

How long does it take to start a mentoring program?

A small pilot can be ready in four to eight weeks: one week to write the goal and sponsor brief, a week or two to recruit, a week to match and a session of training. Larger programs take longer because of approvals and tooling. Starting with a single cohort lets you learn before you scale.

These timings are practical estimates, not research findings.

Should a mentoring program be mandatory?

The 2022 Harvard Business Review article argues that mentoring pays off when mandatory, because people who most need help tend to decline optional offers. A practical compromise is a default enrollment for a defined group, such as new managers, with an opt-out. Mentors should usually volunteer, since an unwilling mentor delivers little.

What is the difference between mentoring and coaching?

Coaching targets a specific skill or goal and is usually shorter and more structured, often with a trained coach. Mentoring is a longer relationship in which an experienced person shares perspective on a career or organization. A mentor does not need coaching credentials but does need time, relevant experience and a willingness to listen.

How many mentees can one mentor take?

One or two at a time is a sensible limit for a first program. More than that and the quality of each meeting drops, or mentors drop out. Group mentoring can serve more people per mentor if senior people are scarce.

This is common practice rather than a researched figure, so adjust it to what your mentors can sustain.

Should the mentor be the mentee's manager?

Usually not. The manager already owns goals, feedback and performance, and a mentee may hold back worries about the job from the person who rates them. A mentor outside the reporting line, ideally in a different team, creates space for open conversation, while the manager stays involved by supporting the time.

How do I know if the program is working?

Choose measures at the start and check them at the end. Look at whether pairs keep meeting, what mentees and mentors say they gained and the outcome tied to your original goal, such as retention or promotion readiness. Treat comparisons with non-participants carefully, since volunteers often differ from the rest of the workforce.

Do I need mentoring software?

Not for a pilot. A spreadsheet, calendar invitations and a survey form will do for a few dozen pairs. Software helps once matching, scheduling and reporting take too much of the owner's time, or when you run several programs at once.

Check data export and published pricing before you commit.

Sources and further reading

• Why Your Mentoring Program Should Be Mandatory (2022), Harvard Business Review
• Why Diversity Programs Fail (2016), Harvard Business Review
• Elements of Effective Practice for Mentoring, Fifth Edition, MENTOR: The National Mentoring Partnership
• Five Steps to Start a Mentoring Program (practitioner guide, vendor-authored), Chronus
• How To Start A Mentorship Program: A Complete Step-By-Step Guide (2022), Forbes Business Council