The Two Crossings

Founders don’t fail inside programs. They fail between them — and the gap belongs to no one.

I have watched a version of this play out in region after region.

A founder finishes an accelerator. She has a working product, three paying customers, a pitch she can deliver in her sleep, and financials that would survive a first meeting. Her cohort ends on a Thursday with a demo day. On Friday morning she wakes up and does not know who to call or what to do next.

Ninety miles away, a regional fund is sitting on committed capital and complaining — sincerely, not as an excuse — that it cannot find investable deals.

Nothing in that story is anybody’s fault. The accelerator did its job well. She did hers. The fund’s partners are not lazy; they are working the phones and their greater network. And yet the thing that was supposed to happen doesn’t happen, and eighteen months later she is running product for a hospital system and the fund has deployed into a company from Nashville.

For years I filed this under pipeline. Not enough deal flow, not enough readiness, not enough programming. Build more, and the two ends will find each other.

They don’t. And I have come to think the reason is structural, ancient, and almost never named.

Stages have owners. Gaps don’t.

Map any regional entrepreneurial ecosystem and you will find that the work sorts into stages. Exposure. Coursework. Applied venturing. Launch. Scale. Capital. The names vary; the sequence doesn’t much.

Now ask a different question. Not what happens at each stage but who is accountable for it.

You will always get an answer. The high school owns exposure. The community college owns coursework. The accelerator owns launch. The fund owns capital. Each has a budget line, a director, a board that reviews its performance, and a metric it reports upward. The stages of an ecosystem are, without exception, somebody’s job.

Now ask who owns the space between two stages.

I have asked this in a lot of rooms. The answer is a pause, and then somebody says “well, we all kind of do,” which is the sound an organization makes when the answer is nobody. There is no budget line for a seam. No program officer has ever been evaluated on a handoff. No annual report contains a section titled Transitions We Successfully Escorted.

So we build programs, because programs can be owned, and we do not build crossings, because crossings cannot. And then we are surprised that founders fall into the places nothing was built.

That is the whole finding, and it would be a modest one, except for what happens when you ask which gaps do the damage. Because it is not all of them. It is two.

Where the category changes

In most regions I have assessed, the programming inside stages ranges from adequate to genuinely good. The handoffs are almost uniformly unbuilt. But two do nearly all the killing.

The first sits between coursework and applied venturing. Call it the Bridge. A student takes an entrepreneurship course, does well, and then — nothing. The credits don’t articulate. The momentum doesn’t transfer. The institution that taught her has no mechanism to hand her to the one that would incubate her, because handing people off is not a service either institution is funded to provide.

The second sits between being ready for capital and receiving it. Call it the Accelerate-to-Capitalize handoff. This is where our founder and our fund fail to meet, ninety miles apart, both doing everything right.

Why these two and not the other four? Here is the pattern I could not see for a long time.

At every other transition, the founder is doing more of what she was already doing. She’s launched, now she’s scaling. She has one customer, now she has ten. The work intensifies. Her category does not change.

At those two gates, she stops being one kind of person and becomes another.

At the Bridge, she stops being a student and becomes a founder. At the capital handoff, she stops being a founder and becomes a counterparty. These are not increments. They are changes of kind — and institutions, whatever else they are, are machines for sorting people into kinds.

A college knows exactly what to do with a student. It has no idea what to do with a founder. A fund knows precisely how to transact with a counterparty. It cannot mentor a student, and it will not take the meeting.

Which leaves the founder, at exactly the two moments that matter most, belonging to nobody.

What the capital gate does to a person

The Bridge is the easier of the two to see, because its failure is bureaucratic and leaves a paper trail. Credits don’t articulate. Nobody signed the agreement. You can point at the missing document.

The second gate is stranger, and I want to be precise about it, because four things happen at once and they compound.

The people who helped her cannot follow her across. The accelerator director, the SBDC counselor, the mentor who took her calls at ten at night on a Friday — none of them has standing on the other side. They are not parties to the transaction. Several of them cannot advise on terms without incurring liability. Most simply cannot be in the room. The escort function evaporates at precisely the moment of maximum asymmetry. There is no ritual anywhere in human culture for and now we negotiate against each other, because everywhere else, the people you negotiate against are not the people who were mentoring you last week.

Her expertise inverts. Up to this point she has known more about her business than anyone alive, and every stage rewarded her for carrying that competence forward. Now she is across from someone who has done this a hundred times and will do it a hundred more, and she has done it never. She walks in assuming her fluency will transfer. The assumption is the thing that fails her.

The crossing looks like one small act and is made of a thousand invisible ones. What the ecosystem observes is a warm introduction: two sentences in an email. What actually made the introduction possible was years of relational work — trust accumulated, reputations quietly staked, a dozen small vouchings that made the introducer willing to spend his standing on her. The ecosystem can only see the email, so the person who sent it becomes the officiant, and officiants receive credit for ceremonies.

And the gate is worth something precisely because it is narrow. This is the part nobody says out loud. A warm introduction is currency, and currency derives its value from scarcity. If the investor takes every meeting, the introduction is worth nothing, and neither is the person who made it. Access is positional. It cannot be widely distributed without destroying the thing that makes it valuable.

That is not a conspiracy, and I want to be careful not to let it sound like one. It is arithmetic. And arithmetic operates perfectly well inside people who would be genuinely horrified to see it written down, which is why it is so durable, and why nobody is ever caught doing it.

I have been on the other side of that gate. I have run community angel funds. I have sat on the side of the table that decides which founder gets twenty minutes and which one gets a courteous email, and I have made introductions that a company’s entire future hung on, and declined to make others for reasons I could have defended out loud and never examined closely. I have watched a founder succeed after an introduction I made and felt a small, warm, entirely unearned sense that I had done something.

I did not notice I was doing it. Nobody does. That is what makes it a mechanism rather than a character flaw.

The shape was familiar, and it took me too long to say why

I did not get this from a book. I got it from a spreadsheet and two dozen years of the same conversation. But once I could see it, the shape nagged at me, because I had met it somewhere before.

Two crossings. A person leaves a known world, does the hard part in a place with no institution and no name, and comes back changed, carrying something the community needs.

That is Joseph Campbell’s monomyth: the hero crosses the first threshold, endures the ordeal, and crosses back on the road home bearing the elixir. Two crossings, and the danger is concentrated at them, not in the middle.

I am wary of leaning on Campbell. His comparative method was criticized by working folklorists almost from publication, and forty years of business-book adaptation have worn the framework smooth to the point of uselessness. Half the consultants in my sector will tell you that you are the guide and not the hero, and they will charge you for it.

So I went upstream, and found something better.

In 1909, a Belgian-French ethnographer named Arnold van Gennep published Les rites de passage, and observed that ceremonies marking a change of social status — birth, initiation, marriage, death — share one architecture across wildly unrelated societies. Three phases: separation from the old status, a liminal period belonging to neither, and incorporation into the new one. Campbell drew on him. Van Gennep got there first, and with data.

Sixty years later, Victor Turner picked up the middle phase and made it the interesting one. The liminal person, he wrote, is “betwixt and between” — outside the classification system entirely. Not the old status, not yet the new one. Structurally invisible. Societies find such people ambiguous and faintly dangerous, and every society that has thought carefully about the problem has built ritual to carry them through, because a person left in the liminal state does not simply wait there. They are lost.

Read that paragraph again with a founder in it.

What this changes

Here is the part that reorganized my practice.

A rite of passage is not performed by the initiate. The community performs it. There is an officiant. There are witnesses. There is someone whose recognized job is to stand at the boundary and escort the person across, and that person’s authority comes from the community, not from the initiate’s readiness.

Nobody in an initiation ceremony asks whether the initiate is ready enough. The community has decided to carry her.

Now consider what we build instead. We build investment-readiness programs. Pitch training. Diligence prep. Curriculum that makes the initiate more ready for a ceremony that has no officiant, no witnesses, and nobody at the boundary. Then, when she doesn’t cross, we conclude the pipeline needs more readiness, and we fund another cohort.

I have taught those programs. I have designed them, and I still do, because a founder who cannot read a term sheet will lose whatever room she gets into. But readiness was never the binding constraint. Escort is. And escort is infrastructure, not motivation — which means it can be named, funded, staffed, and held to account like any other piece of infrastructure, the moment somebody decides to own it.

This also explains something that puzzled me for a decade. In most regions, when a founder does cross, and you trace how, you almost always find one person. Not the person with the title. Usually somebody with an odd job description and no line item, who happens to know the dean and the fund’s chair and made a phone call. Every region has two or three of them. They are doing the officiant’s work, uncompensated, and nobody in the region can name them without being asked twice.

They are findable. In a network analysis, the measure is betweenness centrality — how often a person sits on the shortest path between two actors who otherwise have no connection at all. When you run that math on a real ecosystem and read out the top five names, the room goes quiet, because at least two of them are people nobody thought to invite to the strategy session.

That silence is worth more than any slide I have ever presented.

What I hold loosely

I want to be careful about how much weight this bears.

Campbell may simply be wrong about world mythology; several people who study it for a living think so. And there is a deflationary reading of my whole argument that I cannot rule out: any staged system has boundaries, and boundaries are where systems fail, and the myth adds nothing but a mnemonic. Perhaps I have found the fact that transitions are hard and dressed it in a robe.

Maybe. I have watched this pattern hold across enough regions to trust the observation about the two gates, and I have watched enough interventions succeed and fail to trust the claim about ownership. The mythology I hold more lightly than either.

But a model earns its keep by what it makes you build. Tell an economic development director that her pipeline needs more throughput, and she will fund another program, because that is what the sentence licenses. Tell her that her founders are stranded in a liminal state and that nobody in the county has been given the job of walking them across — and she will go find out who is doing it already, and pay them.

I have watched that second conversation change what a region builds. I have never watched the first one do it.

Where to start

If you run an ecosystem, or fund one, you can do this in an afternoon without hiring anybody.

Draw your stages. Ask who owns each one; you will get names. Then find the two places where a person changes category rather than intensity — usually the seam between education and venturing, and the seam between readiness and capital.

Ask who owns those.

Write down the name. If you can’t, you have found your binding constraint, and it is not a program.

If you can, do the two things that cost more than the diagnosis. Give that person standing — a title, a seat, a mandate that lets them convene people who currently outrank them. And pay them. In every region I have assessed, the person doing the escort work already exists, has been doing it for years, and has never once been compensated for it. They are not asking. They will not ask. They are the reason your last three deals happened, and if they take a job in another state, you will not understand for eighteen months why everything got harder.

Find them before that. They are almost certainly not the person who called this meeting.

We have a phrase in this practice: it takes a village to raise an investable entrepreneur. People hear it as a sentiment about support, and I have stopped correcting them, mostly.

But that is not what villages do, and it never was. A village does not raise a founder by teaching her.

It raises her by walking her across.

Jeremy Turner is the founder of EPIC Mission, where he designs regional entrepreneurial ecosystems and prepares founders for capital. He has studied entrepreneurship and taught it at three institutions, practiced it as a founder who built and exited a company, mentored and coached founders, invested in ventures and run diligence on hundreds, and built the ecosystems that connect those stages — which is why he can see the handoffs between them. His methodology is called Cradle to Capital.