Mentorship programs that survive month three
Most don't. The launch goes well, the matches get made, and by the second quarter nobody can tell you who's still meeting. This is the design and the infrastructure that prevents that.
Every organization wants one. Almost nobody runs one well.
It gets assigned to someone in HR or L&D who has never built one, on top of their existing job. They find a platform, send an interest survey, match people by title and department, and hold a kickoff.
Then the calendar takes over. Pairs meet twice. A few keep going because they'd have found each other anyway. Nobody wants to say it failed, so it quietly stops being mentioned — and two years later someone proposes launching a mentorship program.
The failure isn't enthusiasm. It's that the program had a launch and no operating model.
End-to-end design, built to be handed over.
The deliverable isn't a strategy document. It's a program your team administers, with the decisions already made and written down.
The matching model
Who gets matched with whom, on what basis, and who decides. Most programs match on title and function, which produces polite pairs with nothing to work on. The model gets designed against what your people actually need — skill, exposure, sponsorship, or transition.
Session structure
What a pair does in their first meeting, and the next five. Left open, mentorship becomes coffee. A light structure with real prompts and checkpoints is the difference between pairs that continue and pairs that fade.
Platform configuration
Whatever you're using, or a recommendation if you aren't — configured to your org structure, matching rules, cohort cadence, and reporting. Configuration is where most programs are quietly lost, because the defaults belong to somebody else's organization.
Measurement and runbook
What gets tracked, who reviews it, and what triggers intervention when a pair stops meeting. Plus the administrator runbook — the document that lets the program survive the person who launched it changing jobs.
A program your team owns, and a view leadership can act on.
The program
- A documented matching model and the rules behind it
- Session structure and prompts for the full cycle
- A configured platform with your cohorts loaded
- Launch communications and manager guidance
The organization
- An administrator runbook your team owns
- Participation and continuation reporting
- A leadership view of who's engaged and who isn't
- A defined intervention point before a pair goes quiet
The runbook is the part that matters. Programs don't usually die from bad design — they die when the person who built it moves on and nobody else knows how it works.
Organizations building formal mentorship without formal mentorship experience.
Typically an HR or L&D team that has been asked for this by leadership or an employee resource group, has budget, and doesn't have someone who has run one before. Also organizations relaunching a program that stalled.
Not built for organizations wanting a platform recommendation alone, or a one-time workshop on how to be a mentor.
Four phases, remote by default.
Discovery
What leadership expects, what participants need, what already exists, and what constraints the platform and calendar impose.
Design
Matching model, session structure, cohort cadence, and measurement plan — reviewed with you before anything gets built.
Build and configure
Platform configuration, curriculum, launch communications, manager guidance, administrator runbook.
Launch and handover
Kickoff delivered with your team, first-cycle support, then a handover review at the point most programs start to drift.
Scope and timeline
Scoped to headcount, whether a platform already exists, and how much of the administration your team wants to own. A twenty-minute call is usually enough to size it.
Built by an operator, not an advisor.
Two decades building talent infrastructure — mentorship, early-career pipelines, apprenticeship and leadership programs — for organizations where programs had to run across regions, time zones, and reporting lines. Most recently as Global Head of Learning & Development for Technology and Engineering at Chubb. The relevant experience isn't program theory; it's having administered these at a scale where informal coordination stops working, and knowing which design decisions cause a program to stall six months later.
Chubb
Former Global Head of Learning & Development, Tech & Engineering
UNC Kenan-Flagler
Talent Management Institute, executive development under Marc Effron & Zac Upchurch
Penn State Smeal
MBA
Anthropic
Certified AI practitioner
A twenty-minute call is enough to know whether this fits.
Bring what leadership has asked for and what you have in place today. If it isn't a fit, I'll tell you on the call.