You gave them the right advice. They took notes. Nothing happened. Here is what is actually going wrong, and the delivery changes that fix it.
A dispatch on the delivery side of a mentorship practice — what to do in the weeks when the advice is right and nothing is moving.
TL;DR — the answer first: when a mentee does not implement, the reflex is to explain better. That is the wrong lever, because comprehension was rarely the bottleneck. The failure is almost always in the size of the first move, the absence of a visible record, and an engagement designed to deliver insight when the mentee needed an artifact. Three changes fix most of it: shrink the task until it survives a bad week, make progress visible without you chasing it, and spend a whole session on the obstacle after the second miss instead of restating the plan a third time.
The shape is consistent enough to set a watch by.
The first session is excellent. You do what you have done for twenty years — you listen for eight minutes, you recognize the pattern, and you name the thing they have been circling for a year and could not see. They go quiet. Then they say some version of nobody has ever put it that way. You both leave the call energized, and reasonably so: something real happened.
The second session is good. They have thought about it. They have questions, and the questions are better than the ones they arrived with.
The third session is where it goes. They apologize before you have said anything. It was a heavy few weeks. The client thing blew up. They are going to get to it. And you — because you are decent, and because you remember what running the actual business felt like — say of course, no problem, let's pick it back up.
By session five you are having the same conversation you had in session one, in slightly different words, and both of you can feel it. You start over-preparing. You add material. You explain the reasoning more carefully, because the only tool you were ever handed for this is explain it better.
It does not work, and it does not work for a reason worth understanding.
Here is the test that settles it. Ask them to explain the advice back to you in their own words, aimed at someone in their position. Most of the time they will do it accurately, and often they will do it well — with an example from their own business that you had not heard.
That is a person who has understood. Whatever is stopping them is downstream of comprehension entirely, and every additional hour you spend on the explaining side of the line is an hour spent on the one part of the problem that was already solved.
What is actually in the way is usually one of four things, and they are all mechanical rather than moral.
The first move is too big. You said "build a proper intake process." To you that is one thing, because you have built forty of them and the whole sequence is a single object in your head. To them it is an unbounded project with no obvious first hour, and unbounded projects lose every scheduling contest against a client emergency, permanently.
It has no home in the week. A commitment without a specific hour on a specific date is a wish. The mentee is not lying to you when they agree to it — they are agreeing to a version of next week that has no interruptions in it, and that week has never once arrived.
The cost of the change is real and unnamed. Raising prices means possibly losing a client they are financially dependent on. Narrowing the niche means saying no to work they can currently get. When the advice is correct and expensive, silence is not resistance to the idea. It is an unspoken calculation, and it will keep running underneath every session until someone says it out loud.
The engagement rewards insight instead of output. If every session is a conversation that ends in clarity, you have built a machine that produces clarity. It will produce it reliably, indefinitely, and it will never produce anything else. This is the same failure of design I traced from the other direction in The Scope You Never Wrote Down — what the structure quietly asks for is what the structure gets.
The version of this problem that shows up most in the mentor economy has a recognizable profile, and it is worth naming precisely, because the fix changes depending on who is stalled.
Picture a mentee with eighteen to twenty-five years inside one industry — operations in medical practices, or commercial insurance, or manufacturing quality. Deeply credible. Currently earning through one of two models: a salary they are trying to leave, or two or three retainer clients that consume every available hour. AI fluency somewhere between "uses ChatGPT for email" and "tried building something once and abandoned it." They came to you because they can see that their judgment is worth more than their hours, and they cannot see the path from here to there.
This person is not undisciplined. They ran teams. They shipped things under conditions far worse than a quiet Tuesday. What they are missing is not resolve — it is that every hour of their week is already spoken for by work that pays today, and you have asked them to fund a project that pays in six months out of a budget with nothing left in it. Until the first move is small enough to fit in the margins of that week, no amount of conviction will produce it.
Which is a different problem from the one facing a mentee who has time and is avoiding the work out of fear. Both look identical on a call. Only one is fixed by shrinking the task, and asking "what would have to be true for this to have happened by Friday?" separates them in about ninety seconds.
The Mentor Economy makes this point in Chapter Three, in a line about the two groups who woke up in the same economy on the same morning:
Both groups woke up in the same economy on the same morning. The distance between them is not talent — it is a decision, compounding quietly, one week at a time.
The phrase doing the work there is compounding quietly. Non-implementation does not announce itself. There is no week where the mentee fails visibly. There is only a series of reasonable postponements, each one defensible on its own, which is exactly what makes it so hard for the person inside it to see.
The same chapter puts a clock on it:
Six months on the wrong side is recoverable. A year is hard. Three years is brutal.
That is the real stake in a stalled engagement, and it is why absorbing the stall politely is not the kind thing to do. Your mentee did not hire you to be agreeable about the sixth month. They hired you because you have already been through the part they are stuck in, and the single most valuable thing you can do is refuse to let a quiet quarter pass unnamed.
It also reframes what you are actually selling. You are not selling the answer — the answer is increasingly available to anyone who asks a competent model a good question. You are selling the compressed distance between knowing and having done, which is the argument underneath Sell Judgment, Not Time. If your engagement delivers only the knowing half, you are charging expert rates for the commodity part.
None of these require a harder conversation than you are already having. They require a different session structure.
Worth being precise here, because this is the era's most available wrong turn.
Pointed at understanding, AI actively deepens the stall. A mentee who is already over-informed and under-committed can now generate an unlimited supply of well-reasoned next steps, frameworks, and considerations. Every one of them feels like progress. None of them is. The tool has industrialized the exact activity that was already substituting for the work.
Pointed at execution, it removes the specific friction that stopped the task. The mentee who cannot start the intake document because a blank page is unbearable can have a rough draft in four minutes and spend their thirty on the part only they can do — correcting it against twenty years of knowing what actually goes wrong on a first call. That is the useful configuration, and it is the same distinction I drew in The 80/20 Delivery Layer.
One test, applied at the next session: did the tool produce an artifact that now exists in the business, or did it produce more clarity? Clarity is the failure mode wearing a convincing costume.
The uncomfortable part is that this is a change to your own behavior, not theirs. The stalled engagement is comfortable for the mentor. The sessions are pleasant, the mentee is grateful, the invoice clears, and nobody has to say anything awkward. You can run that indefinitely and tell yourself the value is in the thinking.
But you know within three sessions whether anything is being built, and so do they. Naming it in week four is a small conversation. Naming it in month six is a large one, and by then the person who paid for the difference is not you. If two full cycles pass after you have shrunk the task twice and named the obstacle plainly, you are no longer running a mentorship — and ending it cleanly is a more respectful act than continuing to take money for pleasant conversation.
The reassuring part is that almost none of this is a character problem. Most stalled mentees are competent, serious people who were handed a task too large for the week they actually have. Cut the task, make the record visible, ask the better question on the second miss — and the same person who produced nothing for a month starts producing something every week. Not because they finally found discipline, but because you finally stopped asking them to supply it.
The chapter behind this dispatch — why the distance between two operators in the same industry comes down to a decision compounding one week at a time, and what it costs at six months, a year, and three years — is Chapter Three of The Mentor Economy. Get your copy →
Stop explaining and start shrinking. Non-implementation is almost never a comprehension problem — if they can repeat the advice back to you accurately, they have understood it. What is missing is a first move small enough to survive a bad week. Cut the task until it fits inside thirty minutes on their worst day, attach it to a specific hour on a specific date, and make the next session open with that artifact rather than with a status conversation.
Not entirely, and not zero. You cannot make an adult act. But you can stop designing engagements that quietly depend on willpower you never verified. Most delivery is built as advice plus a calendar invite, which puts the entire burden of translation on the person with the least distance from the problem. Fixing the design is your job. Doing the work is theirs, and the line between those two is worth stating out loud in the first week.
Not on the first miss, and rarely on the second. One miss is information about the task — usually that it was too big. A repeated pattern after you have shrunk the task twice is information about the engagement, and at that point the honest move is to name it directly rather than absorb it. If three cycles pass with nothing built and no stated reason, the engagement is no longer producing anything worth either party paying for.
By making the work visible instead of making yourself the enforcer. Managers chase; mentors design a structure where the absence of progress is obvious to the mentee before the session starts. A shared artifact the mentee updates — not a report they send you — does this without any chasing. The pressure comes from the record, not from your disappointment, which is what keeps the relationship peer-to-peer.
Both, depending on where you point it. Pointed at understanding, it makes things worse — it produces an endless supply of plausible next steps and gives the mentee the sensation of progress without any of it. Pointed at execution, it removes the specific friction that stalled the task: the blank page, the first draft, the format nobody wants to build. The test is whether the tool produced an artifact that exists in the business, or only produced more clarity.
Two cycles. Miss one, shrink the task and try again. Miss two, stop the agenda entirely and spend a full session on the obstacle rather than the plan, because at that point the plan is not the thing that is broken. Waiting longer than two cycles is usually the mentor avoiding an uncomfortable conversation, and the cost of that avoidance compounds on the mentee, not on you.
Author of The Mentor Economy and co-founder of MentorMe. He writes about turning hard-won expertise into AI-leveraged one-person businesses.
The Mentor Economy is the full system — free, you just cover $9.95 shipping.
Get Your Free Copy