Last updated
Two thirds of mentoring relationships fail.
That's not a warning from a pessimist - it's the finding of Professor David Clutterbuck, co-founder of the European Mentoring & Coaching Council, after decades of research into what makes mentoring work. And the number that follows it is just as striking: add structured training for mentors and mentees, and the success rate jumps to 90%.
Think about what that means for how most programmes are designed. All the effort goes into getting a mentor and mentee connected. Almost none of it goes into what they do once they're there. That gap is where mentoring and mentorship programmes live or die.
Here's how to make those choices well.
The 5 steps to designing a mentoring or mentorship programme
Analysis - Understand what the programme is actually for before building anything. Talk to potential participants. Define success before you define the structure.
Programme Design - Make the key variables deliberate: enrolment, structure, matching approach, relationship duration, and how you'll collect feedback throughout.
Matching - The most underestimated step. Define your matching criteria before enrolment opens, and decide early whether you're offering self-matching or administrator matching.
Engagement - Where most programmes fail. Structured training for mentors and mentees, combined with clear milestones, takes the success rate from one in three to nine in ten.
Evaluation - Measure against the goals you set at the start. The point isn't a report - it's finding what to fix in the next cohort.
Step 1 - Analysis - understand the problem before you design the solution
Every mentoring programme that fails to gain traction has one thing in common: it was designed for a programme, not for people.
Before you build anything, talk to your potential participants. Surveys work. Small discussion groups work better. What you're listening for isn't just what people want - it's the gap between what they say they want and what the organisation needs from them.
The questions worth sitting with at this stage:
What is this programme actually for? Retention? Succession planning? DEI goals? Skill transfer ahead of a demographic shift? The clearer this is, the easier every subsequent decision becomes.
What type of mentoring or mentorship programme fits the purpose? Reverse mentoring works differently from career mentoring, which works differently from peer mentoring or group mentoring. The type shapes everything - matching criteria, relationship duration, what success looks like.
Formal or informal? Informal feels accessible but tends to drift. Formal feels structured but can feel mandatory and transactional. Most effective programmes live somewhere between the two.
What does failure look like here? If this programme ran for a year and you shut it down, what would have gone wrong? Starting with that question often surfaces the real risks before they become real problems.
Ayrshire Chamber of Commerce knew exactly what they were solving for before they built anything: their members had told them directly that mentoring was one of the most valuable things the Chamber provided. The problem wasn't whether to do it - it was how to grow it without the administration overwhelming the team. That clarity of purpose shaped every design decision that followed.
The output of this stage isn't a document - it's a set of constraints that stop you designing the wrong thing.
Step 2 - Programme Design - the variables that matter
Once you know what you're building toward, the design decisions become a lot less arbitrary. This is where you establish the guidelines that will govern how your programme runs - and every variable should be a deliberate choice, not a default. Here's what you're actually deciding:
Launch and promotion. A programme that people don't know about doesn't exist. How will you reach potential participants - and crucially, how will you reach both sides? Mentor recruitment is consistently harder than mentee recruitment, and it needs its own approach.
Enrolment. Open enrolment creates volume but can dilute quality. Invite-only maintains quality but limits reach. Administrator approval gives you control but adds friction. None of these is wrong - but they have consequences you should choose deliberately.
Mentoring roadmap and structure. An unstructured mentoring or mentorship relationship puts the burden of momentum entirely on two people who are already busy. A clear mentoring roadmap - with milestones for the first meeting, goal-setting, mid-point review, and close - removes that burden. Relationships that stall usually do so because neither party knows whose job it is to move things forward.
Matching criteria. Will you offer self-matching or administrator matching? The answer affects what data you need to collect at enrolment, and how much flexibility you want participants to have.
Connection limits. Should a mentor be able to take on multiple mentees simultaneously? Without a policy, you'll end up with both extremes, and neither serves participants well.
Relationship duration. Capped relationships create natural momentum and clear endings. Open-ended relationships can deepen further - but can also drift indefinitely without resolution.
Feedback mechanisms. Build an end-of-relationship survey in from the start, not as an afterthought. End-of-relationship feedback is what allows a programme to be refined and improved over time - it's how good programmes become great ones.
The Royal College of Midwives designed their programme specifically to address the shortage of senior leaders of colour in maternity - a clearly defined goal that shaped everything from their matching criteria to the type of mentors they recruited. That specificity is what allowed 76% of mentees who completed profiles to find a suitable matching mentor. Vague programme goals tend to produce vague results.
Step 3 - Matching - the hardest part to get right
Good matching is the most labour-intensive part of running a mentoring programme, and the part most often underestimated. If you're running a large programme and trying to match participants manually via spreadsheets, you'll spend more time on logistics than on anything that actually helps participants.
The NHS Clinical Entrepreneur Programme - the world's largest entrepreneurial training programme in healthcare - replaced a manual spreadsheet-based matching system with a digital solution when they moved to PLD. The scale of their network made a cost-per-user model unworkable; the complexity of their needs made a manual process unsustainable. Getting this step right unlocked everything else.
The first decision is self-matching versus administrator matching. Self-matching gives participants agency and tends to produce stronger buy-in, but it requires surfacing the right information about mentors so mentees can make an informed choice. Administrator matching gives you more control over quality and balance, but it shifts the responsibility - and the blame if it goes wrong - onto you.
Whichever approach you choose, the matching criteria need to be defined before enrolment begins, not after. Prioritise the criteria that are most predictive of a useful relationship for your specific programme goals.
One thing that's often overlooked: matching on difference can be as valuable as matching on similarity. In the alta mentoring programme for women in aviation and aerospace, 99% of users said their relationship on the platform was a good match - a result that reflects careful thought about what variables actually predict a useful relationship, not just shared background.
Step 4 - Engagement - where most programmes fall short
This is the Clutterbuck gap. You've got two people matched. Now what?
Training is non-negotiable. The reason that one-in-three to nine-in-ten statistic is so dramatic is that most people have never been in a mentoring relationship before - or they've been in one that drifted. Mentors need to understand the difference between mentoring and advising, between guiding and directing. Mentees need to know how to prepare, set goals, and drive the relationship forward rather than waiting to be helped.
This doesn't need to be extensive. Clear guidelines on the role of mentor and mentee, how to conduct a first meeting, how to set goals together, and how to give and receive honest feedback - that's enough to make the difference.
Milestones keep relationships on track. The first meeting is the most important moment in any mentoring or mentorship relationship - and the one most likely to be awkward and underprepared. Build a structured approach to it. After that, a clear mentoring roadmap with milestones at regular intervals - goal-setting, a mid-point review, and a defined close - gives the relationship a shape that carries both parties through the natural dips in momentum that every long-term commitment faces.
IOSH - the world's largest Chartered Membership Body for Safety & Health Professionals, with 49,000 members across 130 countries - found that structured skill-based matching and a clear framework for relationships transformed what their members could do together. In their own words, it revolutionised the way members interact with each other and form meaningful mentoring relationships.
CIAT, the Chartered Institute of Architectural Technologists, have been running their PLD mentoring programme for eight years. Their platform - supporting 9,500 members and affiliates - now largely manages itself. That's not an accident of technology. It's the result of a programme designed with sufficient structure that participants can navigate it independently.
Step 5 - Evaluation - measure what matters
Setting goals at the start is only useful if you measure against them at the end. The metrics worth tracking regardless of your programme's purpose:
- Relationship completion rate - did participants reach the end of the programme, or did relationships stall?
- Goal achievement rate - did mentees reach the goals they set at the start?
- Participant satisfaction - from both sides. Mentor experience is often neglected.
- Qualitative feedback from end-of-relationship surveys.
ICAS - the Institute of Chartered Accountants of Scotland, with 23,000 members globally - monitors their programme's performance on a daily and weekly basis. That level of visibility lets them see in near real-time how marketing activity is driving engagement, and make adjustments while the programme is running rather than waiting until a cohort closes.
The Salvation Army uses their platform to track exactly where each coach and coachee is in the process at any given time - when first meetings have happened, when final sessions have been completed, and whether the programme is delivering what it promised. For a coaching programme running across a charity of nearly 6,500 employees, that oversight would simply not be possible without evaluation built in from the start.
The goal isn't to produce a report - it's to find the two or three things that will make the next cohort better than this one. Every programme has friction points. Evaluation tells you where they are.
Where to go from here
If you're designing a mentoring or mentorship programme from scratch, the five steps above give you a framework. But framework and execution are different things - and the details of execution, particularly around matching, training content, and journey design, are where most programmes either gain traction or quietly lose it.
If you’d like to see how PLD approaches programme design, book a demo, take the free 3 minute quiz or get in touch if you want to see the mentoring platform in action, and to explore whether a dedicated mentoring platform is right for your organisation.
