Mentoring Program ROI: How to Measure and Calculate It
Mentoring program ROI is the program’s net monetary benefit divided by its total cost: ROI = (monetized benefits − cost) ÷ cost × 100. The formula is simple. The hard part is everything around it: most mentoring benefits are not money, they show up months later, and they mix with everything else happening in the organization.
This guide teaches the method. It contains no borrowed benchmarks or industry averages, on purpose. A return figure is only credible when it is built from your own costs, your own outcomes and an honest estimate of how much the program caused.
Why is mentoring ROI hard to calculate?
Three reasons, and each one shapes the method below:
- The benefits are indirect. Mentoring changes confidence, networks and judgment. Money appears only when those changes move something the organization already pays for, such as an employee staying instead of leaving.
- The timing is long. A new hire who stays because of a good mentor shows up in next year’s attrition data, not this quarter’s.
- The attribution is messy. People who volunteer for mentoring are often already more engaged. Without a fair comparison, the program gets credit for differences that existed before it started.
So treat ROI as a careful estimate with stated assumptions, not as a measurement. That framing is also what makes it believable to a finance team.
What goes on the cost side?
Count everything, at full cost. An ROI built on a partial cost base is the fastest way to lose credibility. The largest line is almost always people’s time, not software.
| Cost line | Formula |
|---|---|
| Participant meeting time | [number of participants] × [meeting hours per person] × [loaded hourly rate] |
| Preparation and follow-up | [participants] × [prep hours per person] × [loaded hourly rate] |
| Mentor and mentee training | [people trained] × [training hours] × [loaded hourly rate] + [trainer or content cost] |
| Program coordination | [coordinator hours] × [coordinator’s loaded hourly rate] |
| Software | [subscription cost for the program period] |
| Events and materials | [kickoff, closing event, travel, catering, certificates] |
| Evaluation | [hours spent on surveys and analysis] × [loaded hourly rate] |
The loaded hourly rate is the real cost of one hour of someone’s time:
Loaded hourly rate = (annual salary + benefits + employer taxes + overhead share) ÷ annual working hours
Your finance or HR team usually has a standard figure. Use theirs rather than inventing one: it saves an argument later. If mentors and mentees sit at very different levels, calculate a separate rate for each group.
Which mentoring benefits can be turned into money?
Only convert outcomes your organization already puts a price on. For each one below, the inputs come from your own HR and finance data.
1. Retention: avoided replacement cost
If mentees leave less often than similar non-participants, each avoided departure saves a replacement cost.
Avoided departures = ([comparison group departure rate] − [participant departure rate]) × [number of participants]
Retention benefit = [avoided departures] × [cost of replacing one employee]
The replacement cost should be your organization’s own estimate: recruiting fees and advertising, interviewer hours × loaded rate, onboarding and training, plus the productivity gap while the new person ramps up.
2. Faster time to productivity
For onboarding and buddy programs, the question is how much sooner mentored new hires reach full output.
Benefit = [mentored new hires] × [days saved to full productivity] × [daily loaded cost] × [average productivity shortfall during ramp-up]
The “days saved” must come from a comparison with similar new hires who were not mentored, or from a previous cohort. Managers’ ratings of readiness at fixed points (day 30, 60, 90) are a practical source.
3. Internal promotion instead of an external hire
Leadership and high-potential programs aim to fill roles from inside. Internal moves usually cost less than external hires.
Benefit = [roles filled internally by participants above the comparison rate] × ([external hire cost] − [internal fill cost])
External hire cost includes search fees, longer vacancy and a longer ramp-up; internal fill cost includes backfilling the person’s previous role.
What about the benefits you cannot price?
Engagement, sense of belonging, knowledge sharing across silos, a more diverse leadership pipeline: report them, with evidence, but leave them unconverted. Forcing a monetary value onto them invites the challenge that weakens your whole calculation. Listing them separately as intangible benefits is honest and standard practice.
How do you isolate the program’s effect?
This step decides whether your ROI survives its first question. The approach widely known as the Phillips ROI Methodology, developed by Jack Phillips, adds a fifth level (ROI) to the four Kirkpatrick evaluation levels and treats isolation as a required step. Its practical tools fit mentoring well.
- Comparison group. Compare participants with similar non-participants in role, level, tenure and location. A waitlist or a phased rollout gives you one naturally. This is the strongest option.
- Trend line. If you have a stable history of the metric (for example, two years of first-year attrition), compare the post-program value with where the trend was heading.
- Participant and manager estimates. When no comparison is possible, ask: “What share of this improvement do you attribute to mentoring?” and “How confident are you in that estimate?” Multiply the two to discount for uncertainty.
Attributed benefit = [total improvement value] × [share attributed to mentoring] × [confidence level]
A mentee who says 50% of a promotion’s readiness came from mentoring, with 80% confidence, contributes 50% × 80% = 40% of that benefit.
Two conservative rules, also central to the Phillips approach, keep the result defensible: when you have two estimates, use the lower one; and for a single program cycle, count benefits for one year only.
A worked example (hypothetical)
Every number below is a round placeholder, invented to show the arithmetic. None of it describes a real program or a typical result.
A company runs a 9-month program with 40 mentor–mentee pairs (80 people). Its finance team uses a loaded hourly rate of $50 for everyone, to keep the example simple.
Costs
| Item | Calculation | Cost |
|---|---|---|
| Meeting time | 80 people × 12 hours × $50 | $48,000 |
| Training | 80 people × 2 hours × $50 | $8,000 |
| Coordination | 200 hours × $50 | $10,000 |
| Software | placeholder | $3,000 |
| Events | placeholder | $1,000 |
| Total | $70,000 |
Benefits
- Retention. Over 12 months, a matched comparison group lost people at a rate equal to 3 more departures than the 40 mentees had. The company estimates one replacement at $30,000. Benefit: 3 × $30,000 = $90,000.
- Internal fills. Two mentees moved into roles that would otherwise have been filled externally, saving $10,000 each. They attribute 50% of it to mentoring, with 80% confidence. Benefit: 2 × $10,000 × 50% × 80% = $8,000.
- Total monetized benefit: $98,000.
Result
- Net benefit: $98,000 − $70,000 = $28,000
- ROI: $28,000 ÷ $70,000 × 100 = 40% (hypothetical)
- Benefit-cost ratio: $98,000 ÷ $70,000 = 1.4
Now test it. If the comparison supports only 1 avoided departure instead of 3, the benefit drops to $38,000 and the ROI turns negative. That sensitivity is the most useful part of the exercise: it shows the sponsor which assumption the result depends on. Here, the break-even point is $70,000 ÷ $30,000 ≈ 2.3 avoided departures, which is a question an approver can reason about.
When should you report ROI, and when ROE?
Return on expectations (ROE), a term associated with the Kirkpatrick model, asks whether the program delivered what its sponsors said they expected at the start. It covers outcomes that are real but not worth pricing.
| Situation | Better choice |
|---|---|
| A finance-led sponsor asks “what did we get for the money?” | ROI, with stated assumptions and sensitivity |
| The goals are inclusion, culture, belonging or leadership readiness | ROE, with indicators and participant evidence |
| First cycle, no comparison group, little HR data yet | ROE plus a break-even estimate |
| Mature program with clean retention and promotion data | Both: ROE every cycle, ROI once a year |
Agree on which one you will report before the program starts. The expectations conversation belongs in the mentoring program business case, and the indicators come from your mentoring program evaluation plan.
What are the most common ROI mistakes?
- Borrowing someone else’s number. A percentage from another company’s program is not evidence about yours. Leave it out.
- Counting software but not time. Participant hours are the biggest cost. Leaving them out inflates ROI and is easy to spot.
- Skipping isolation. Comparing mentees with “everyone else” credits the program with differences that were already there.
- Monetizing everything. Pricing engagement or belonging with a made-up rate damages the credibility of the solid numbers.
- No baseline. Without pre-program attrition, time-to-productivity or promotion data, there is nothing to compare against.
- Reporting a single figure. Show the assumptions, the conservative case and the break-even point. The how to report mentoring results guide covers presenting this to leadership.
How does software help with the numbers?
Most of the cost side and the participation evidence come from program records, which are hard to rebuild from email and spreadsheets. In Mentornity, meetings booked through the platform are recorded with their date, participants and session, feedback forms run with each session, and reminders go to anyone with missing feedback. Reports export as CSV, Excel or PDF with charts, so meeting counts and participation are ready when you calculate time costs. The monetary conversion and attribution stay with you and your finance team. See how this works for employee mentoring programs, or try it free with up to 10 users.
For more on designing, measuring and running programs, browse all mentoring guides.