The Pillars of Modern Entrepreneurship

Modern entrepreneurship is not short on advice. It is short on coherence.

Founders are handed frameworks, tactics and motivational narratives, and still cannot get from ambiguity to a venture that repeats. The advice is not wrong so much as misaddressed. Almost all of it assumes you already know what you are building and are looking to build it better.

That assumption is the error. A venture is not an execution problem with a personality attached. It is a learning system, and its rate of progress is set by how fast it turns what it does not know into something it does.

Which replaces the usual question with a harder one. Not how driven you are. Where your learning breaks, and what you have decided to do about it.

Venture Physics is what I call reading a venture that way: as a system with mechanics rather than a story with a hero. Three forces decide the outcome. Each forces a decision most founders would rather defer, and deferring is itself a decision, made by default.

Most founder advice is management advice

Management is the discipline of improving a business you already understand. It has a century of rigour behind it and it works: set objectives, forecast, allocate, review. It works because the thing being managed holds still long enough to be measured.

A venture holds nothing still. It is searching for a business while the ground under it is revised by customers, competitors and its own choices. Managing and searching are different activities with different failure modes, and the advice industry has spent thirty years selling the first to people doing the second.

So the advice arrives technically correct and practically useless. It offers ways to run a business better to people who do not yet know which business they are running. That is why working hard so reliably produces activity without producing progress. Effort applied to an unsettled question does not settle it. It makes the wrong version of it larger.

Searching has its own mechanics, and they are not mysterious. A learning system works in two motions. It diverges, widening the set of things that might be true, and it converges, committing to one of them and acting as though it were. Progress is the alternation between them. A venture that only diverges never builds anything. A venture that only converges builds the wrong thing faster.

Which is also where founders are told the wrong thing about themselves. Conviction, grit and vision are inputs to persistence. They are not inputs to knowledge. Your beliefs do not shape your outcomes. Your feedback loops do.

Say that plainly and the shape of the problem appears. If the loop is the mechanism, then anything that kills a venture has to break the loop somewhere, and a loop has only three places to break.

Nothing can enter, because what you need to know has not happened yet. That is uncertainty.

Something enters and cannot travel, because everything routes through one person. That is pressure.

Something travels and cannot land, because the market receives through a filter you did not design. That is compression.

Three forces, three breaks, three decisions.

Uncertainty, and the decision it forces

At seed stage the most important truths are not knowable by thinking. Frank Knight drew the line a century ago: risk has a distribution you can price, uncertainty does not. A founder lives almost entirely in the second condition, having been trained for the first.

Analysis cannot close that gap, and the reason it stays attractive is that it is available. It can be done alone, at a desk, on a schedule, and it produces artefacts that look like progress. Signal cannot be scheduled, and it usually arrives as a refusal.

So the decision uncertainty forces is not what to test. It is what you will accept as an answer, settled before you go looking. Most founders never make it, which is why their evidence keeps arriving conveniently. A conversation that could not have gone badly told you nothing.

This is where the received wisdom does most of its damage. Fail fast, pivot, learn relentlessly. Each is sound, and each becomes unfalsifiable in the hands of a founder who has not set that standard in advance. Failing fast without one is quitting early. Pivoting without one is divergence with better public relations. Learning relentlessly without one is a story you tell about a year in which nothing was decided.

Learning is not free either. Every unit of it costs a deal, a quarter or a relationship, which makes are we learning the wrong question. The strategic one is what this particular piece of knowledge is worth, and whether you are willing to pay that. Sarasvathy named the rule underneath it: entrepreneurs working under genuine uncertainty commit what they can afford to lose rather than what a projected return would justify, because the return cannot be projected and the loss can.

Most founders also diverge on the wrong axis. They hold options open on who they serve, because narrowing feels like losing, and they close options on how they might be wrong, because that is where the discomfort lives. The productive arrangement is the reverse. Converge hard on who, so that every signal is comparable to the last one. Diverge widely on how you might be wrong, so that something can still surprise you.

Two seed teams with equal talent and funding diverge here, in the ordinary sense. Team A works to be right before shipping: positioning debated, deck perfected, roadmap refined. Team B has already decided what would prove it wrong, and spends the same weeks finding out. Team B does not guess better. It holds a standing rule about what counts, and the rule settles questions Team A is still deferring.

Which is the first principle. Signal over narrative. A story is not validation and a waitlist is not traction, because neither cost anyone anything. Only a commitment a buyer could have refused carries information.

Knowledge at this stage is not deduced. It is bought, and the price is being told no by people you hoped would say yes.

Pressure, and the decision it forces

Pressure does not only change how founders feel. It changes how they behave, and the change is documented rather than anecdotal. Staw, Sandelands and Dutton named it threat rigidity: under threat, organisations restrict information processing and concentrate control. At the moment a system most needs to learn, it narrows.

In a venture that narrowing has a signature. Decisions climb back up to the founder. Approvals lengthen. The founder becomes the routing layer through which everything must pass, and calls it responsibility.

Read against the two motions, pressure is a convergence disorder. It does not stop a company deciding. It makes every decision converge on one person, which is the fastest way to make a system that can no longer diverge at all.

The failure this produces never appears in the account afterwards. The company did not stall for want of talent. It stalled because knowledge could not move through it faster than one person could process it, and that person was certain they were helping.

So the decision pressure forces is about authority rather than effort. Which decisions leave your desk permanently, and at what cost are you willing to let them be made worse than you would make them? A founder who cannot answer the second half has not delegated anything. They have lent it out.

Which is the second principle. Systems over heroics. If the venture depends on founder energy it will fail at the moment that energy is most expensive, and the failure will look like bad luck.

The test is not whether the company moves. It is whether it moves when you are not in the room, and whether you can live with how it moves.

Compression, and the decision it forces

The interface between a venture and its market has narrowed. Customers meet decisions through summaries, default recommendations and shortlists, each of which is a filter applied before any judgement of yours reaches a person.

Buyers have always narrowed. Hauser and Wernerfelt explained why: evaluating options costs something, so a rational buyer considers a small set and ignores the rest. What has changed is who assembles the set. It used to be the buyer, using their own knowledge and their own shortcuts. Increasingly it is a system, using neither. Search rankings and category shortlists did this first. Generative assistants now do it with reasoning, reading your material and deciding on a buyer’s behalf what deserves to be mentioned at all.

Which is worth stating precisely, because it means the first assessment of your venture is increasingly made by something that will never meet you, cannot be persuaded, and has no interest in your intentions.

The consequence is not gradual. Where choice is compressed, second place disappears. Second best in a market of twenty is a position with a business attached. Second best on a list of three that nobody scrolls past is not a position at all.

The decision compression forces is the one founders most want to avoid: what are you willing to be compared on? You do not get to be judged on everything you are good at. Something will be used to sort you, and if you have not chosen it, the filter chooses for you, usually the dimension on which you are most ordinary.

Which is the third principle. Legibility over complexity. Clarity here is not presentation. It is the decision to be narrower in scope than you are capable of being, so that what remains can be understood and checked quickly by someone who owes you no attention.

This is not a branding problem. It is a structural one.

The rule

THE RULE

Three questions, one for each force.

1. When did you last pay for information you did not want?

   Learning has a price, and it is usually a lost deal, a wasted quarter or an uncomfortable conversation. If nothing in the last three months cost you something to find out, you have been diverging inside your own building. Evidence that arrives comfortably is not evidence, it is confirmation, and you were already holding that.

2. What did you decide this month that someone else should have decided?

   Name it, then name what it would have cost to let their version stand. If that cost was survivable and you stepped in anyway, you did not step in for the company. The company has now learned it cannot proceed without you, and it will act on that lesson for years.

3. When you lose, do you lose for the same reason each time?

   A venture that loses differently every time has not been sorted yet, it has been ignored. Losing consistently on one dimension means the market has already decided what you are compared on. The only question left is whether you chose that dimension or inherited it.

If you cannot reduce uncertainty with signal, hold decisions away from yourself under pressure, and stay legible where choice is compressed, you do not have a strategy.

You have motion.

The three are not equally visible, which is why they are misdiagnosed. Uncertainty is invisible by definition. Pressure feels like commitment. Compression happens somewhere you are not. The force you can see is rarely the one binding you, and the decision it demands is rarely the one you want to make.

None of these decisions has become easier. What has changed is how little time now sits between making one and paying for it.

So the work is not to chase trends, stack frameworks or perform confidence. It is to build a venture that can learn faster than its assumptions, decide faster than its fear, and adapt without losing what it is.

Replace noise with constraints. Replace ambiguity with mechanics.

Where these three forces meet arithmetic is a separate question, and I have taken it up in the Entrepreneurial Efficiency Equation.

Entrepreneurship is not a test of character. It is a sequence of decisions made without the information you wanted.