Almost every organization that wants to support underrepresented talent reaches for the same tool. They launch a mentorship program, pair people up, hold a kickoff event, and feel good about it. Then, a year later, most of those pairs have quietly stopped meeting, and nobody is quite sure what went wrong. The intentions were real, the energy was there at the start, and still the thing fizzled. This pattern repeats so often that it deserves a hard look. Mentorship genuinely changes careers, so when programs fail this consistently, the problem is usually the design, not the idea.
The first reason is that good intentions get treated as a substitute for structure. Many programs match two people, wish them luck, and assume the relationship will grow on its own. But most people do not know how to be a mentor or a mentee without some scaffolding to follow. Without suggested topics, a meeting rhythm, or any sense of what success looks like, the conversations drift into vague small talk. After a few awkward coffees with no clear purpose, both people quietly let it lapse. A relationship that is everyone's nice idea and no one's actual responsibility tends to disappear.
The second reason is poor matching that looks fine on a spreadsheet. Programs often pair people by job title or department and call it a fit, ignoring the things that actually make mentorship work. A mentee growing in one direction gets matched with a mentor whose experience points somewhere else entirely. Sometimes the gap in seniority is so wide that the mentee feels too intimidated to ask the questions that matter most. Good matching considers goals, communication style, and what the mentee actually wants to learn, not just where two people sit on the org chart. When the pairing is wrong, no amount of enthusiasm will save it.
The third reason cuts deeper, especially for programs aimed at underrepresented talent. Mentorship gives advice, but advice alone does not open doors. A mentor can coach someone all year and still leave them stuck if that mentor never advocates for them in the rooms where decisions get made. What changes careers is sponsorship, when someone with influence puts their own credibility on the line to recommend you for the stretch project or the promotion. Many programs deliver friendly guidance and call it support, while the real barrier was never a lack of advice. It was a lack of access, and access is exactly what casual mentorship rarely provides.
The fourth reason is that nobody measures whether any of it is working. Programs celebrate the number of pairs created and never check whether those pairs still meet six months later. They do not ask whether mentees got promoted, took on bigger roles, or felt more connected to their work. Without that feedback, the same flawed design runs year after year, and the failures stay invisible. You cannot improve something you refuse to measure, and good intentions make it easy to skip the measuring. A program that tracked real outcomes would be forced to confront what is and is not actually helping.
There is a fifth problem that sits underneath the others, which is treating mentorship as a program to launch rather than a culture to build. A single annual event with a sign-up sheet sends the message that this is a box to check, not a real commitment. Mentors are often volunteers stretched thin by their actual jobs, given no time, training, or recognition for the role. When the organization does not visibly value the work, mentors deprioritize it, and the relationships starve from neglect rather than bad intentions. Real mentorship cultures protect time for it, train mentors on how to do it well, and reward the people who show up consistently. They also make room for informal connections to form, since the best mentorship often grows outside any official pairing. A program bolted onto an indifferent culture will always underperform, no matter how well it is designed on paper.
So the fixes are not mysterious, they are just rarely done. Give pairs a clear structure, a meeting rhythm, and topics to work through so the relationship has somewhere to go. Match people on goals and growth rather than titles, and check in to see whether the fit is working. Build sponsorship into the design, so that influence and access flow alongside advice. Then measure the outcomes that matter and adjust the program based on what you learn. Mentorship is one of the most powerful tools we have for building fairer paths, but only when it is treated as real work rather than a feel-good gesture. Done halfway, it becomes a photo opportunity that quietly wastes everyone's time. Done with real intention, it becomes one of the few things that genuinely moves people forward. The programs that take the design seriously are the ones that quietly change lives.
