When it goes wrong
When a mentoring match goes wrong, the scheme names the failure, acts on it within the week, and closes the match deliberately if it cannot be repaired: the one thing it never does is let the match fade out.
A coordinator who has a match that is not working and has to decide this week whether to repair it, re-match it or close it. It is also for a volunteer who has begun to dread the visit and assumes that is their failure.
The person on the receiving end. Every failure on this page is survivable if it is caught and handled, and every one of them does real damage if it is left to run until somebody simply stops turning up.
The scheme learns about the failure from the person supported, months late, at the point where the only option left is the one that harms them most. A scheme is not judged on whether matches fail. It is judged on how it handles the ones that do.
That urgency is not tidiness. Matches that end early do harm, and youth mentoring relationships that broke down before three months were associated with outcomes worse than no match at all, in the analysis by Grossman and Rhodes published in 2002. Set against that, mentoring’s advantage is a small but real improvement in outcomes for the young person, around 0.2 of a standard deviation across the research, in the meta-analysis by DuBois and colleagues published in 2011. The disadvantage is sharper than the advantage, and that asymmetry is the whole subject. A mentoring scheme is not a low-risk intervention that either works or does nothing, and any page presenting mentoring as harmless enthusiasm misleads the coordinator reading it.
Four failures account for most of what goes wrong in one-to-one schemes, and they are listed here in the order a coordinator will actually meet them. Drift is the most common and the least reported. Dependency is next, and is often mistaken for success. Volunteer drop-out is third and is the one that converts a manageable problem into a harmful ending. Outright breakdown, including complaint and allegation, is the rarest and the one every policy is written about.
| Failure | Order a coordinator meets them | What it looks like | The intervention |
|---|---|---|---|
| Drift | Most common | Sessions still happen, but nothing is being worked towards and the logs repeat | Re-contract the match against its original goal, in a three-way conversation |
| Dependency | Second | Contact between sessions rises, the volunteer becomes the only person called | Restore the boundary, widen the network, name the end date early |
| Volunteer drop-out | Third | Cancellations, then silence, then a resignation by text | Hold a planned ending anyway, with the volunteer present if possible |
| Breakdown, complaint or allegation | Rarest | A refusal to continue, a formal complaint, or a safeguarding referral | Suspend contact, follow the safeguarding route, keep the two processes separate |
Drift, the failure that looks like everything being fine
Drift is a match that still meets and no longer does anything. The mentoring relationship has become a pleasant, purposeless hour: the same cafe, the same conversation, the same log entry three months running. It is the most common dysfunction in mentoring programmes and the least reported, because nobody involved is unhappy enough to complain.
The intervention is re-contracting, and it works best done in the open. Bring the volunteer and the person supported together with the coordinator, take out whatever was agreed at the start, and ask what has changed. Sometimes the original goal has been met and nobody noticed, in which case the match should move to a new goal or start planning its ending. Sometimes the goal was never the young person’s goal, only the referrer’s, which is a matching problem and needs saying out loud. Drift is easy to fix at three months and very hard to fix at nine.
Dependency, and why it gets mistaken for a good match
Dependency is the point where the volunteer stops being one relationship among several and becomes the only one. It shows in the contact between sessions: texts every evening, calls at night, the person supported describing the volunteer as their best friend or, worse, as family. It is easy to mistake for success: it looks like a strong bond, and the volunteer often finds it flattering before they find it frightening.
The intervention has three parts and none of them is a sudden withdrawal. Restore the agreed contact boundary and let the scheme, not the volunteer, be the one that says so. Widen the network deliberately, by bringing in a group activity, a social prescribing link worker where the person is registered with a practice that has one, or a second contact who is not the volunteer. Name the end date early and repeat it, because a dependent match is a match with a dangerous ending waiting at the end of it. This is also the failure where the volunteer needs support in their own right, since being needed that much is difficult to step back from.
Volunteer drop-out, which is where the real harm enters
Volunteer drop-out turns a scheme’s ordinary attrition into the one outcome the evidence warns about. Volunteers leave for legitimate reasons: a job, a move, illness, a new baby. The damage is not in the leaving, it is in leaving badly, with the last session being one that nobody knew was the last. For a referred child who has already been let down by adults, a mentor who disappears confirms the lesson the scheme was set up to disprove.
A volunteer resigns from a match that is under three months old
The volunteer, wherever they possibly can, gives a short notice period and attends one final session to say goodbye in person, even if it feels awkward and even if they are leaving because the match was hard. The coordinator does not simply re-match: the ending is handled first, with the reason given honestly and in a form the person supported can hold, and only then is a new match considered, with a gap and a fresh conversation. An abrupt ending inside three months is the specific circumstance associated with worse outcomes than never being matched at all, and it is worth considerable effort to avoid.
Breakdown, complaint and allegation
Breakdown is the rarest of the four and the one that needs the clearest procedure, because two different processes start at the same moment and must not be run as one. A complaint about the service is handled through the complaints procedure, with a named person, a timescale and a written outcome. A safeguarding concern or an allegation against a volunteer goes down the safeguarding route to the named lead, and in most cases beyond the scheme: to the local authority in England, Wales and Scotland, and to the Health and Social Care Trust in Northern Ireland, where councils have no social care function. That route runs whatever the complaints procedure says. Contact is suspended while that runs, the suspension is described to the volunteer as neutral rather than as a finding, and the person supported gets a different named contact immediately so that they are not left with nobody.
An allegation is made against a volunteer
The volunteer stops all contact at once, keeps any messages and logs rather than tidying them, and does not approach the person supported or their family to explain. The coordinator suspends the match, refers to the named safeguarding lead the same day, records the times and decisions, and does not investigate the allegation in-house. A scheme that tries to establish the truth itself before referring will destroy both the account and its own position.
Mentoring, mentorship and coaching, since the words are used loosely
Mentoring and mentorship describe the same thing at different scales: mentoring is the activity, mentorship is the relationship or the arrangement considered as a whole, and no UK scheme should design anything around the distinction. Coaching is a genuinely different activity, focused on performance against a defined goal over a short, contracted number of sessions, usually paid and usually professional. Peer arrangements, tutoring and workplace programmes each borrow parts of the mentoring model without borrowing its safeguards, which is why a corporate mentoring scheme can run on light coordination and a scheme working with referred children cannot.
The mentor or befriender
- Says early that a match is not working, rather than waiting to be asked.
- Gives notice instead of fading out.
- Attends the final session even when the ending is difficult.
- Passes a concern to the scheme rather than solving it privately.
The young person or person supported
- Can end the match at any time, without a reason and without losing the service.
- Is told why a match is ending, in words that do not leave them to fill in the blank.
- Is given a named contact at the scheme who is not the volunteer.
- Is not re-matched immediately as a way of tidying the problem away.
The scheme plans for failure before it happens: a written complaints route, a safeguarding route that is separate from it, a standing question in supervision about how the match feels, and a rule that no match ends without somebody senior knowing it has. It also records failures honestly, because a scheme reporting no breakdowns is not a safe scheme, it is a scheme that is not looking.