Mentoring as Infrastructure: Learning, Power, and Risk in Organisational Design
Mentoring is often framed as goodwill, yet it functions more like infrastructure. When designed well, it strengthens judgement and learning. When left to intention alone, it can narrow thinking, create dependence, and quietly amplify organisational risk.
By Dr Alwin Tan, MBBS, FRACS, EMBA (Melbourne Business School)
Senior Surgeon | Governance Leader | HealthTech Co-founder
Harvard Medical School — AI in Healthcare
Australian Institute of Company Directors — GAICD candidate
University of Oxford — Sustainable Enterprise
Institute for Systems Integrity
Executive summary
Mentoring is widely promoted as a low-cost, high-impact learning intervention. Evidence shows consistent — but modest — benefits. It also shows that when mentoring fails, the harm can be disproportionately large.
This paper reframes mentoring not as goodwill, but as learning infrastructure: a system that shapes judgment, power, and risk. When designed well, mentoring accelerates learning under uncertainty. When poorly governed, it quietly amplifies dependency, inequity, and loss of visibility.
This analysis builds on ISI’s earlier work on decision-making under system stress, oversight blindness, and the Failure Taxonomy, applying those system lenses to mentoring as a learning control mechanism.
1. What mentoring is — and is not
Mentoring is:
- a mechanism for transferring tacit knowledge
- a process that shapes judgment under ambiguity
- a social learning channel during role transition
Mentoring is not:
- sponsorship
- performance management
- therapy
- a substitute for structural equity
Failure to define this boundary is a primary source of risk. When mentoring is asked to perform functions it was not designed for, learning outcomes become unstable, and accountability diffuses.
2. What the evidence shows
Across sectors, mentoring is associated with:
- improved job satisfaction and commitment
- increased confidence and career clarity
- accelerated learning during professional transitions
However, the evidence also shows that:
- Effect sizes are moderate rather than transformational
- The benefits are unevenly distributed
- Negative mentoring experiences produce stronger adverse effects than positive mentoring produces gains
This asymmetry matters. From a governance perspective, mentoring should be evaluated not only on its average benefit, but on its tail risk.
3. Mentoring failure modes: a system view
Mentoring failures rarely present as misconduct. More often, they emerge through patterns that feel reasonable in the moment.
Common failure modes include:
Dependency creation
Learning is replaced by reliance on the mentor’s judgment. Decision-making capacity does not grow; it narrows.
Gatekeeping and capture
Access to opportunity becomes personalised rather than transparent, concentrating informal power.
Silencing through power gradients
Mentees self-censor to preserve safety, reputation, or favour — reducing learning visibility.
Role confusion
Mentors act simultaneously as assessor, advocate, and confidant, collapsing boundaries.
Program theatre
Pairing occurs without defined learning goals, mentor capability development, monitoring, or exit pathways.
These patterns closely mirror broader system failures described in ISI’s Failure Taxonomy — where harm emerges not from bad actors, but from misaligned incentives, unexamined authority, and gradual loss of oversight.
4. Mentoring versus sponsorship
Mentoring builds capability and sense-making.
Sponsorship reallocates power and opportunity.
Conflating the two creates predictable consequences:
- unmet expectations
- perceived unfairness
- erosion of trust in leadership development initiatives
Boards should expect explicit clarity on which function a program is designed to deliver — and which it is not.
5. Reverse mentoring: opportunity and risk
Reverse mentoring has gained attention as a way to surface blind spots in senior decision-making and accelerate learning around technology, culture, and frontline realities.
Its potential benefits are real. So are its risks.
Without explicit safeguards, reverse mentoring can:
- Exposes junior participants to unmanaged power asymmetry
- become performative rather than substantive
- generate insight without authority to act
As with traditional mentoring, reverse mentoring requires stronger design and governance, not greater reliance on goodwill.
6. Governance implications
Mentoring should be treated as a learning control system, not a benevolent add-on.
Key governance questions include:
- What learning outcome is this mentoring intended to produce?
- Where do power gradients sit, and how are they acknowledged?
- How can participants exit safely and without reputational cost?
- What signals would indicate that learning is narrowing rather than expanding?
These questions echo ISI’s earlier findings on oversight blindness: systems often retain formal compliance while quietly losing visibility over how decisions are actually shaped.
7. Designing safe mentoring infrastructure
Effective mentoring systems share several design characteristics:
- clear role definition and boundary setting
- mentor capability development, not assumption of competence
- opt-out mechanisms without penalty
- confidential escalation pathways
- periodic review of learning outcomes, not just participation rates
Many of these safeguards align with the diagnostic and intervention principles outlined in the ISI Toolkit, which emphasises early visibility, role clarity, and safe escalation as prerequisites for learning under pressure.
Trust is not a control.
Design is.
Conclusion
Mentoring remains a valuable learning mechanism. But when treated as intention rather than infrastructure, it can quietly undermine the very judgment it is meant to build.
Systems that rely on mentoring without governing it risk:
- stalled learning
- amplification of inequity
- and degradation of decision quality under stress
Mentoring does not fail because people lack goodwill.
It fails when systems rely on trust without design.
References
Allen, T.D., Eby, L.T., Poteet, M.L., Lentz, E. and Lima, L. (2004) ‘Career benefits associated with mentoring for protégés: A meta-analysis’, Journal of Applied Psychology, 89(1), pp. 127–136.
Argyris, C. and Schön, D.A. (1978). Organizational Learning: A Theory of Action Perspective. Reading, MA: Addison-Wesley.
Edmondson, A.C. (2018). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Hoboken, NJ: Wiley.
Eby, L.T., Allen, T.D., Evans, S.C., Ng, T., and DuBois, D.L. (2008). Does mentoring matter? A multidisciplinary meta-analysis, Journal of Vocational Behavior, 72(2), pp. 254–267.
Eby, L.T., McManus, S.E., Simon, S.A. and Russell, J.E.A. (2000) ‘The protégé’s perspective regarding negative mentoring experiences: The development of a taxonomy’, Journal of Vocational Behavior, 57(1), pp. 1–21.
Hewlett, S.A. (2013) ‘Forget a mentor, find a sponsor’, Harvard Business Review, January–February.
McDonald, S. and Westphal, J.D. (2013) ‘Access denied: Low mentoring of women and minority executives’, Social Forces, 91(3), pp. 853–877.
Scandura, T.A. (1998) ‘Dysfunctional mentoring relationships and outcomes’, Journal of Management, 24(3), pp. 449–467.
