You have read how they did it. The order of the hires, the pivot in year two, the channel that worked. You are now building a plan shaped like their sequence.
That sequence was one of many routes to where they are, and it was selected by conditions you do not share. Study what the company is, structurally, not the order in which it got there.
Why these three models
The decision is what to extract from a company you admire. The features that fire are an open system exchanging with its environment, a history that may or may not be binding, and a target state you would need to describe precisely enough to aim at.
Three lenses that disagree with each other, which is the point. Equifinality produces an equilibrium answer saying the route is largely uninformative. Path dependence produces a cycle answer saying that in some regimes the route is the only thing that matters. Stock and flow produces a complex answer about how to describe a state in terms you could actually build toward. The first two are in direct tension and the third is how you tell which one you are in.
1. Many routes, one destination
The formal claim is older and stronger than the business literature that ignores it.
An open system, meaning one that exchanges matter, energy or information with its environment, can reach the same final state from different initial conditions and along different paths. That is equifinality, and it is a property of the class of system, not an observation about a few lucky companies. A closed system runs to an equilibrium determined by where it started. An open one does not, which is precisely what makes it open.1
Every company is an open system by this definition, which means the inference chain runs as follows. Their state was reachable by many routes. The route they took was selected by their initial conditions: the market at that moment, the people who happened to be available, the capital that happened to be offered. You do not have those initial conditions. Therefore the information content of their route, for you, is low, while the information content of their state is high.
And there is a selection problem stacked on top. The routes you read about are the ones that arrived. The same sequence executed by companies that did not arrive is not written down, so you cannot see the denominator and cannot tell whether the route caused the outcome or merely accompanied it.
Bertalanffy is also candid about what his own framework can and cannot do, in a passage worth carrying into any strategy discussion. System theory offers explanation in principle rather than prediction, and he illustrates it against his own side: professors of economics can explain economic phenomena well in principle, and are not able to predict fluctuations in the stock market with respect to certain shares or dates.1 Take the structural claim. Do not take a forecast from it.
Equifinality, the route lens
- Assumes: the system exchanges with its environment, so its end state is not fixed by its starting point.
- Fits because: you are about to copy a sequence from a company with different starting conditions.
- Breaks when: the system is effectively closed over the relevant horizon, or when lock-in has already occurred. That is model two.
- Evidence: grade B plus. It follows from open-system structure, and its practical reach in any particular business is a judgement.
- Counteracts: treating a founder narrative as a procedure.
- May reinforce: dismissing sequence entirely, which is the opposite error and is covered next.
2. Except when the route is the whole thing
The second lens is here to stop the first from being read as licence.
Some choices close other choices. A technology standard adopted early, a first large customer whose requirements shape the product, a market position that determines who will and will not partner with you: these constrain the reachable set from that point on. Where lock-in is real, the state you observe was not reachable by many routes. It was reachable by theirs, and possibly only by theirs.
So equifinality and path dependence are not competing theories. They describe different regimes, and the operative question is which one you are in. The test is whether the choice under discussion forecloses others. Hiring a second engineer does not. Choosing the payment rail your entire product will settle on does. Signing a distribution agreement with an exclusivity clause does.
This gives a usable division of their story. Their foreclosing decisions are worth studying. The rest of the narrative is texture. Those foreclosing points are where sequence genuinely mattered. Everything else in the narrative is texture, and texture is where founders spend their attention because it is where the anecdotes are.
Path dependence, the lock-in lens
- Assumes: some decisions foreclose options, so the reachable set narrows as history accumulates.
- Fits because: at least one decision in their story was almost certainly of this kind.
- Breaks when: a lock-in you assumed was binding turns out to be cheaply reversible, which technology shifts do regularly.
- Evidence: grade B. Well established as a mechanism, and identifying which specific decisions locked in is usually retrospective.
- Counteracts: assuming everything is reachable from where you stand.
- May reinforce: fatalism about a position that could still be changed.
3. How to describe a state you could actually build
The third lens is the practical one, because “copy the state” is useless until you can say what a state is.
A state is a set of stock levels. Not activities, not culture, not narrative: quantities that have accumulated. How many customers, in which segments, at what retention. How much cash, and how much of it is unrestricted. How many people, with which authorities. How much verified evidence about the business exists in a form a stranger can check.
Describing a company this way is unglamorous and immediately actionable, because a stock changes only through its flows. Once you have written their state as stocks, every gap between yours and theirs becomes a question about a flow you can name and possibly move. That is a plan. A sequence copied from a narrative is not.
Expect this to feel harder than it should. Highly educated adults routinely fail to infer the behaviour of simple stock and flow systems, and the failure is not attributable to graph literacy, contextual knowledge, motivation or cognitive capacity.2 Difficulty here is the default rather than a signal that you are doing it wrong.
Write down what they are. Then write down what you are. The difference is a list of flows.
Stock and flow, the description lens
- Assumes: a state can be written as accumulated quantities that change only through identifiable flows.
- Fits because: you need a target specific enough to aim at.
- Breaks when: the thing that matters genuinely is not a stock. Some advantages are relational and do not accumulate anywhere you can measure.
- Evidence: grade A. Structural, and the difficulty people have with it is well replicated.
- Counteracts: describing a target company in adjectives.
- May reinforce: measuring what is countable and missing what is not.
The levers, cheapest first
- Rewrite their story as a balance sheet of stocks. One page, present tense, no chronology. This alone changes what you take from the case.
- Mark the foreclosing decisions. Go through their narrative and flag only the choices that closed other choices. Usually there are two or three, and they are rarely the ones the story dwells on.
- Ignore the rest of the sequence. Deliberately. The order of their first ten hires is a fact about their labour market in that year.
- List the gaps as flows, not as goals. “Reach their retention” is a goal. “Reduce the second-month cancellation rate” is a flow, and only flows have levers.
- Ask what their initial conditions were. If the answer includes a founder who had already sold a company in that sector, that is not a step you can sequence into your plan.
What to do this week
Do now, sized at one afternoon, effect immediate. Take the company whose path you were about to copy and write its present state as six to ten stock levels. Reversible, free, and dominant across every scenario about whether their route would work for you.
Hedge, where the premium is the whole loss, live before the next planning cycle. Pick the single largest gap between their stocks and yours, and name the one flow that closes it. If the gap turns out not to matter you have spent an afternoon and learned which stock you were wrong about.
Defer and trigger, size fixed now. Do not rewrite your strategy against this. Pre-commit the trigger: the next time a plan in your company is justified by how another company did it, the justification has to be restated as a stock gap and a flow before it gets funded. Decide now who applies that test, because a standard with no owner is a preference.
Watch the arrivals. The state description lands this afternoon. Any flow you change lands after the stock responds, which for retention or cash is months, and the temptation to conclude it failed will arrive well before the evidence does.
What usually happens next
Run the break test first. Has a rule changed, has an actor entered or left, has a measurement become a target? If the company you are studying operated before a regulatory change that now governs your market, their state was reachable under rules that no longer exist, and neither their path nor their state transfers cleanly.
If nothing broke, the pattern is consistent. Path copying produces a plan that is defensible in a board meeting, because every step has a precedent, and that is exactly why it survives scrutiny it should not. State copying produces a plan that looks less impressive and names actual flows.
Subtract the counterfactual before crediting a copied move. If the channel that worked for them also worked for you, ask whether it worked because they proved it or because it was going to work for anyone in that market that year. The second is far more common and is invisible from inside the success.
What this ensemble cannot see
All three lenses assume you can observe their state. You mostly cannot.
What is public is a subset chosen by them, and it is selected to look like a coherent story. Retention by cohort, unrestricted cash, and the real authority structure are exactly the stocks that decide outcomes and exactly the ones not disclosed. So the honest version of this method is that you will build a partial state description from incomplete public information, and the missing stocks are systematically the important ones.
There is also a genuine tension left unresolved here. Equifinality says the route carries little information; path dependence says that in lock-in regimes it carries almost all of it. This article gives you a test for which regime you are in, and that test is a judgement call made by the person who wants an answer. Two thoughtful people will place the same decision differently.
And one property none of these models contains: reading a founder story does something to your conviction that a stock table does not. The narrative is doing motivational work as well as informational work, and stripping it to quantities removes both. Whether that trade is worth it depends on what you are currently short of.
The one action that survives the ignorance: before your next planning session, write the target company’s state as stocks on one page, and mark each line as observed, inferred or unknown. If more than half are unknown, you are not copying their state. You are copying a story, and now you know it.
Who has to move
This only bites if the person building the plan reads it, and the path narrative is usually what got everyone excited about the plan in the first place. The cheapest first test is the one-page stock description with confidence marks. If most lines come back unknown, that is a stronger argument than any objection to the strategy itself, and it takes an afternoon.
Sources and notes
- Ludwig von Bertalanffy, General System Theory: Foundations, Development, Applications, George Braziller. Open systems, steady states and equifinality are developed in chapters 5 and 6, with equifinality treated at length as the property distinguishing open systems from closed ones. The concession on explanatory reach, that in complex and theoretically underdeveloped fields we must be satisfied with what Hayek termed explanation in principle, together with the illustration that professors of economics can explain economic phenomena in principle but cannot predict fluctuations in the stock market with respect to certain shares or dates, is in chapter 2. Bertalanffy adds that explanation in principle is better than none at all, which is the spirit in which this article uses it.
- Matthew A. Cronin, Cleotilde Gonzalez and John D. Sterman, Why don’t well-educated adults understand accumulation? A challenge to researchers, educators, and citizens, Organizational Behavior and Human Decision Processes 108(1), 2009, pages 116 to 130. Author copy: https://www.mit.edu/~jsterman/CroninGonzalezSterman061210.pdf. The abstract states that highly educated people are often unable to infer the behaviour of simple stock-flow systems, and that persistent poor performance is not attributable to an inability to interpret graphs, contextual knowledge, motivation, or cognitive capacity.
A note on what this argues against. Most founder education is path narrative, and this piece says the path is the low-information part. That is a strong claim and it should be held with its limit attached: equifinality is a statement that a state is reachable by many routes, not that all routes reach it. Sequence still matters where a choice forecloses others. The argument is against copying sequence by default, not against sequence.
Joshua Agonya Pi’Rwot, Founder.