Why am I still doing every part of it myself?

Why Do Two People Doing Everything Themselves Need Completely Different Things?

Because the visible result carries no information about what produced it.

One person is doing all of it because there has genuinely never been anybody. Another is doing all of it because the work has no describable form and therefore cannot be moved, even though two capable people are sitting ten feet away. A third has both of those solved and no money. From outside, all three are the same person: the one who does everything, who looks tired, who says they should really get some help at some point.

The general instruction aimed at all three is identical and it is aimed at willingness, which for at least two of them was never the constraint. Advice that bounces off somebody often enough teaches them that they must be the problem, since it works for other people and has never once worked for them.

Five roots follow. They produce the same week. They need different things, and the most common way somebody loses a year on this is by working hard on a root they do not have.

Read all five. Most people find one lands harder than the rest, and a fair number find that two are running at once and that one of them is quietly funding the other.

What If Nobody Around You Knows the Work Exists?

Then the absence of offers is an information problem, and reading it as indifference has probably been costing you something on top of the work itself.

Start with what other people can actually see. They see outputs: the thing that got shipped, the event that happened, the household that runs, the accounts that balanced. Outputs are visible by design, because an output is the part meant to be seen. Everything upstream of an output is invisible unless somebody narrates it, and you have never narrated any of it, because narrating it while doing it would take longer than doing it.

Star and Strauss made this argument carefully in 1999, under the heading of visible and invisible work. Their point is that visibility is a property of where the observer is standing and of how the surrounding system counts things, rather than a property of the effort itself. The same work is fully visible to the person doing it and structurally invisible to everybody else, and no bad faith is required anywhere for that to be true. Two honest limits: it is a conceptual and observational argument rather than a measured finding, and it names a category rather than establishing a magnitude. What it does well is describe the exact mechanism by which real effort becomes uncountable, which is that nobody is present at the moment the effort happens.

Now put that together with how offers get made. Nobody offers to take on something they have not noticed. There is no reservoir of people around you declining to help. There is a group of people who believe that whatever it is, it is handled, because it always is, and the evidence for that belief is overwhelming and comes from you.

This produces a loop that seals itself, and the sealing is the reason it can run for a decade:

• The work is not described, so nobody sees it.
• Nobody sees it, so nobody offers.
• Nobody offers, so there is no occasion that would require describing it.
• With no occasion to describe it, it stays undescribed.

Nothing inside that circuit generates the interruption. Every step is a consequence of the step before it, and none of the steps involves anybody behaving badly or thinking anything unkind.

The tell for this root is a particular flavor of irritation that arrives without a target. You are not angry at a specific person for a specific failure to help. You are carrying a low, general resentment at nobody in particular, which never resolves because there is nothing at the other end of it. The other tell is the answer you get when you finally describe one piece of it out loud. It is almost never I know. It is almost always some version of I had no idea you were doing that, delivered with real surprise, and that surprise is the most useful thing on this page if you can get somebody to say it to you.

What If the Arithmetic Genuinely Does Not Close?

Then this root is real, it is not a story you invented, and it deserves to be stated precisely rather than either dismissed or surrendered to.

There is a version of this advice that assumes the money is there and you are being cautious with it. Sometimes that is true. Often it is nowhere near true, and being told otherwise is why so many people carrying this stop reading anything about it. Somebody describing this out loud put the whole bind in one sentence: "that drains me. I personally, that, that drains me because I would like to spend more time trying to bring money in or doing. I've that bring money in, not the work, but I haven't got the money to pay somebody else to do it".

Read the shape of that rather than the words. The thing that would free the resource requires the resource. Paying somebody needs money the work has not made yet, and making more of it needs the time the paying would have freed. It is a closed loop, and unlike most binds people describe, this one is arithmetic rather than psychology. No amount of clarity about your own motivations opens it.

Two things are worth saying about it, and they point in opposite directions, which is why this root gets handled so badly.

The first is that the loop is often genuinely closed and pretending otherwise is not help. Where the numbers do not work, they do not work. One person put the whole calculation in a sentence: "I kind of afraid to hire a manager that will speak with the local people [...] Because, you know, a manager should get paid, and if he should get paid, I need to pay it from somewhere." That is correct arithmetic rather than an avoided feeling, and treating it as the second thing is how the conversation stops being useful.

The second is that the money is usually the only precondition actually missing, and the other three have been left untouched for years on the grounds that the money is missing. That is the expensive part. Describing the work costs nothing. Finding out who in your reach could take a piece of it costs nothing. Deciding in advance what an acceptable returned version looks like costs nothing. None of those requires a budget, all three take a few hours in total, and all three are almost always still undone on the day the money finally appears. Which is how somebody arrives at the moment they can afford help, hires, and discovers the situation gets worse for a season, because the only precondition they had solved was the one that got solved for them.

There is a second currency worth checking before concluding the loop is shut. Money is one medium. Reciprocity is another, and most people carrying this have a favor account that has only ever run in one direction. If you have spent fifteen years doing things for people and have never once called anything in, the balance available to you is much larger than you think, and it has never been tested because testing it would require describing the work, which returns us to the first precondition again.

The tell for this root is that you can produce the numbers immediately. Ask somebody with a different root why they have not brought somebody in and you get a vague answer. Ask somebody with this root and you get a figure, a rate, and a monthly shortfall, because they have run it more than once and can do it from memory.

What If There Is Nothing to Hand Across, Even With Somebody Willing Standing There?

Then the constraint is the form of the work rather than anybody's willingness, and this is the root that most often hides inside all the others.

Picture the actual moment of transfer, because most advice skips it. Somebody capable is standing there with an hour. You want to give them a piece of it. What do you say? For a task that has ever been written down, this is easy. For a task that has only ever been done, you find yourself starting three sentences and abandoning all three, because the thing you are trying to describe has no beginning, no edges, and no separable steps. It is a set of conditions and thresholds you assembled over years and have never had to state.

Groups solve this without anybody planning it. There is a described model of how it works, usually called a transactive memory system: individuals in a group specialize, and the group carries a rough directory of who knows what, so nobody has to hold everything and any member can find what they lack by asking the right person. The model has an experimental literature behind it, developed by Hollingshead among others, showing that groups with a working directory recall and perform differently from groups without one. I am naming the model rather than a single finding, because the useful part for this page is structural. A person running everything alone has no directory. There is no entry pointing at anybody but them, so there is no route by which a piece of the work could be located, let alone taken.

Then there is the estimating problem sitting on top of it, which explains why the first attempt fails so reliably. Knowing something makes it extremely difficult to model not knowing it. Newton reported a study in 1990 in which people tapped out the rhythm of a familiar tune and estimated how many listeners would name it. Tappers expected around half. Roughly one listener in forty got it. Handle that number carefully: it comes from an unpublished dissertation study, it is retold in business writing far more often than it is examined, and the precise figures should not be leaned on. The direction is what matters, and the direction is that the person who knows the thing has almost no ability to feel how little of it is arriving.

Applied to your afternoon: when you estimate that explaining would take longer than doing, you are estimating from inside the knowledge. That estimate is wrong in a predictable direction, and it is wrong in the same direction every single time, which is how a repeatedly wrong estimate ends up governing fifteen years of decisions without ever being checked once.

The tell for this root is a specific and slightly embarrassing experience: you have tried to explain a piece of it, out loud, to somebody willing, and stopped part way through because you could not find the thread. Not because they were slow. Because you discovered, in the middle of the sentence, that you did not have the description either.

How Do You Tell Which of the Five Is Yours?

By checking what has actually happened, rather than by asking yourself how you feel about handing things over.

Two or three of these will point the same way, which is enough to start on. It is also worth knowing which ones are cheap. The first root and the second root are hours of work, not months. The third is arithmetic. The fourth is a document. The fifth is watching somebody. None of the five requires you to become a different person before anything can move, which is the assumption most advice about this quietly builds in.

Where Does This Hide When You Would Not Call It Doing Everything?

In a great many places that have nothing to do with a business, which is where it goes unaddressed longest.

Look down the third column. That is the actual finding, and it is the same in every row regardless of how different the settings look. The load is not what makes these situations similar. What makes them similar is that in each one, there is exactly one person who could say what happens next, and no record of it anywhere else.

The version people write about most is the one where a whole structure sits on a single person and the org chart has one box doing the work of nine. That version has its own page: what happens when an entire operation rests on one person, and what a week away does and does not fix. It is one setting among several and it is not the only one, and treating it as the whole subject is why a person running a household on the same mechanism never finds anything written for them.

There is a test that works in every row, and it takes a minute. Imagine you are unreachable for twelve days. Not dramatic, and not permanent. A bad flu and a phone left in a drawer. Now ask what actually stops, and then ask something harder: could anybody else even work out what has stopped? For most people carrying this, the first answer is uncomfortable and the second answer is the one that changes their afternoon.

Why Does Just Hire Someone Fail as Advice?

Because it answers the second precondition and quietly assumes the other three.

Somebody arriving does not produce a describable task. It does not produce the money, which is why the advice lands as slightly insulting when the money is the missing piece. And it does nothing at all about your ability to tolerate a version you would not have made, which is the precondition people discover last and usually in public.

The sequence when only the second precondition is solved is predictable enough to write down in advance. A person arrives. There is nothing written, so the first two weeks is spent with them shadowing you, which slows you down and produces no artifact. They begin taking pieces, and every piece requires a conversation, because the conditions live in your memory. Their versions come back at seventy percent, correctly, because that is what a third attempt looks like. You repair them yourself, quietly, because the deadline is real. Three months in, your hours have not fallen, there is now a salary going out, and the honest internal summary is that it did not work.

That summary is wrong in an expensive way, because the experiment that failed was never the one people think. What failed was an attempt to move undescribed work into somebody else's hands. Nothing was learned about whether the person was capable, and nothing was learned about whether you can let go, although both will be blamed.

The related advice that fails for the same reason is the instruction to be less of a perfectionist about it. In a great many of these situations the standard is genuinely the thing that keeps the work working, and being told to lower it reads, correctly, as being told to be worse at the thing you are good at. What actually needs to happen is more boring and more specific than either instruction: the standard needs to be written down once so that it exists somewhere other than in your judgment. That is a document rather than a personality change, and it is why this whole subject responds better to an hour with a blank page than to a year of resolutions.

There is a third piece of standard advice worth retiring here. The suggestion to simply do less assumes that the pile is made of optional things. Some of it is, and the deleting is genuinely the fastest single reduction available. But a large part of what routes through you holds something up in the plain sense that something breaks if it stops, and telling somebody in that position to do less is an instruction with no available action attached to it.

What Actually Moves This?

The preconditions, in order, and the order is not the one people reach for.

Almost everybody starts at precondition two, because it is the one that feels like action. You look for a person. What happens next is the sequence described above, where a capable person arrives into a job that has no form and the whole thing gets read as a failed hire. Starting at two while one is missing is the single most common way this gets attempted and the single most reliable way it fails.

Start with the inventory, because it is cheap and nothing else works without it. One sitting, an hour, one line per item. Everything that routes through you, including the small ones that never take long and the ones that only come round twice a year, which are the likeliest to be invisible even to you. Against each line, one column: could anybody else currently name that this exists.

That hour does three things at once, and only one of them is the obvious one. It produces a list, which is the obvious one. It converts a single undifferentiated weight into a countable set, which changes how the load feels within about a day, because a countable thing can be reasoned about and an uncountable one can only be endured. And it reliably surfaces a group of items, usually somewhere between a fifth and a third, that no longer have any reason to exist and can simply be stopped. That last group is the fastest reduction available to anybody in this position, and it requires no other person, no money, and no conversation.

Then describe exactly one thing. Not the system. One recurring task, written down as though somebody else would be doing it next month. Half a page. What triggers it, what finished looks like, what the two or three judgment calls are and how you make them.

This is the step people skip because it feels like preparation rather than progress, and it is the actual work. It is also where the estimating error shows up in a way you can see. Time it. Compare the real figure with the number you would have guessed. The gap between those two numbers has been governing this for years and almost nobody has ever looked at it.

Then look for the receiver, with the description in your hand. This is a completely different search from the one that has never started. You are now able to say what the person would do, which means you can ask people whether they know somebody, which means the question is answerable. A search for somebody to help with everything has no answer and no first move. A search for somebody to do the specific written thing has both.

Then decide the tolerance in advance, in writing. Before anything comes back, write down what an acceptable version looks like, including what you will let stand that you would have done differently. Decide that on a calm morning rather than in the moment the version arrives, because the moment it arrives is exactly when you will be least able to judge it.

Then expect the rise and count the repetitions. Three to eight. Write the number down so that week two has something to be measured against other than your mood.

One thing worth saying about the emotional half, since a page about mechanisms can imply there is nothing else here. For a good number of people, working through those five steps is straightforward until the fourth, at which point something arrives that has nothing to do with logistics. If handing over a small piece produces a reaction out of all proportion to the piece, the preconditions are solved and the thing in the way is elsewhere, and that is worth knowing rather than pushing through.

Which Parts of This Are Research, and Which Parts Are Mine?

Worth separating, since the two carry very different weight and this subject attracts a lot of confident writing built on very little.

On work that goes uncounted, Star and Strauss published the argument about visible and invisible work in 1999, in the journal Computer Supported Cooperative Work. It is a conceptual and observational paper rather than a study with a measured result, it establishes a category rather than a magnitude, and its original setting is systems and organizations rather than households. Both authors are on one paper, so it counts here as a single line of evidence rather than as two. It is used as a lens and should not be read as a measurement of anything.

On how groups avoid one person carrying everything, the transactive memory model has been developed experimentally since the mid-1980s, including a well-known line of work by Hollingshead on how members retrieve what other members know. I am naming the model rather than resting the argument on one person's finding, partly because no single result carries it and partly because the useful part here is the structure it describes rather than any measured effect size.

On the difficulty of estimating what another person does not know, Newton reported the tapping study in a 1990 doctoral dissertation. It is unpublished, small, and retold in business writing far more often than it is read. The specific figures should be handled with care. The direction of the error has held up in other work on how badly people model the state of somebody else's knowledge, and the direction is the only part this page uses.

On why a heavy load with total authority does not read as strain, Karasek published the demand and control model in 1979 in Administrative Science Quarterly. The two main effects are supported by a large literature. The interaction claim at the center of the model, that authority buffers demand, has repeatedly failed to replicate cleanly and is quoted with far more confidence than it has earned. That limitation is stated here rather than left out, because the model is being used for the part it does establish.

Two things this page deliberately does not claim. It does not claim that the state you are in has a name, because the folk phrase names the load and nothing else, and no research term covers the situation described here. And it does not claim to know which of the five roots is yours, because nothing written down can establish that, and any page that tells you otherwise is selling the certainty rather than earning it.

The rest is mine and carries no research weight at all. I hold no clinical qualification of any kind and I am not a therapist. None of the researchers named here has any connection to this page or to the seminar. Where their published work and my reading of it disagree, theirs is the one to go and check.

How Long Does This Take, Honestly?

Faster than you would guess for four of the five roots, and the speeds are wildly different, which is why a single answer is useless.

The inventory is an hour and it pays out the same day. That is not a figure of speech about how it feels. The deleting alone usually removes real items from a real week inside the first week of the list existing.

The description of a single task is two hours and it is the highest return per hour available anywhere in this. Most people find it takes about a quarter of what they expected, and the discovery that it was a two-hour job rather than a two-week job does more to move this than any amount of resolve.

The receiver search runs on the description and takes as long as your particular circumstances take, which nobody can promise you. What can be said is that it becomes a real search rather than an intention on the day the description exists.

The first handover takes three to eight repetitions to stop costing more than it saves. In ordinary life that is four to ten weeks. Almost every abandoned attempt stops inside that window.

And the holding, which is the part that actually costs you, takes longest and moves in a way that surprises people. It does not shrink when the doing shrinks. It shrinks when the shape exists somewhere other than in your own memory, which is why the writing steps do more for the exhaustion than the transferring steps do. People routinely report the load feeling lighter before a single task has moved, and then distrust that because nothing has visibly changed.

Expect it to come back under pressure. A bad month, somebody leaving, a period where the arithmetic gets tight again, and the whole thing reassembles quickly. That is the conditions returning rather than everything you built being lost, and the second recovery is much faster than the first because the documents still exist.

What Is One Thing to Do This Week, Before You Give Anything to Anybody?

One hour, one page, and nobody needs to be told you are doing it.

Sit down with one sheet of paper and write everything that routes through you. Not organized, not prioritized, just written, including the small ones, the ones that come round twice a year, and the ones you would be embarrassed to admit take up space. Then go down the list once and mark each line with a single letter: D if it could simply stop, W if somebody else could do it once it was written down, M if it needs money that does not exist, and Y if you want to keep it and would fight to.

Most people expect the list to be about thirty items and find it is closer to seventy. Most people expect the D column to be empty and find it holds a tenth to a third of the page. And almost everybody discovers that the Y column, the things they actually want, is short, specific, and has been buried under everything else for years.

That sheet costs you an hour and gives you the one thing your situation has never had, which is a description of the work that exists somewhere other than in your own head. Nothing else on this page can start until it does.

If what landed hardest was not the logistics, and instead was the recognition that nobody in your life has ever heard the full description, or that the exhaustion has stopped responding to rest, or that you cannot picture a version of your life that is not this, our Sacred Valley seminar works at that layer rather than at the level of the workload. We start with what the arrangement is actually doing for you, because an arrangement held in place for fifteen years is very rarely doing nothing, and until that is visible every practical plan gets abandoned in week three for reasons that look like circumstances. Everybody is screened before arriving, which keeps the group small enough that somebody who has spent years being the only one who can does not quietly spend the week organizing everybody else. The aim afterward is that you are equipped to get all of this done on your own by yourself, which is a different sentence from the one you have been living, where every piece of the work is also yours. One limitation worth stating plainly: exact dates, the price and the final cap on group size are still being settled, so anybody comparing programs on those numbers this week cannot yet compare ours. Ask every program you are weighing the same questions, this one included, and decide from what comes back. The hour with the sheet of paper costs you nothing and is worth having either way.