Issue #13: Decay by Disuse
Authority that is held but never exercised does not stay held. It reverts to influence.
Brief: Issue 12 named the Venture Operator Function and set it as a test. A test focused on whether one accountable node holds the conversion authority, and whether that authority is backed, incentive-aligned and durable. What the test cannot do is tell you whether the function stays. Its scope is a single moment in time. And being only a moment in time, a function can pass every question and still be gone a year later, with no leadership change and no veto to account for it, because authority held but never used decays.
This issue names that decay, works through the three mechanisms behind it, and closes with a maintenance test to catch a function that has begun reverting to influence while the structure still records it as intact.
A function can pass the readiness assessment in March and be hollowed out by December. And it will not be because someone took away its authority. There was no leadership change, no veto by a committee, no single action you can point to. The chart still records the same name against the same four capabilities, and the authority behind that name no longer moves anything. In the nine months between the two snapshots, it was simply never used.
An authority that is never used stops being believed, losing credibility overtime.
The Venture Operator Function, “the function” defined in Issue 12, the capstone of the Corporate Venturing Under Constraint arc on Put Ideas To Work, as the accountable conversion layer between a validated signal and the structural commitment required to maximise its potential for success. Though it closed the arc, it closed on a six-question diagnostic and an instruction.
Run it on the opportunity area in the portfolio that carries the most weight, write down the name of the person who holds each capability, and if one name holds all four, the decision holds when it is made, and it would survive the sponsor leaving, the function exists as a structure.
That instruction though correct is incomplete. It answers a question in the present tense, and leaves untouched the one that decides whether the function was worth building at all, whether it will last. This issue reopens a closed arc to take on that question.
And address the limitation of all snapshot style diagnostics.
What the snapshot cannot read
The three conditions that separate a real function from a node with a title were backed, incentive-aligned, and durable.
Each of them tests a state:
Backed asks whether the authority holds when it is pushed.
The incentive-aligned condition is a question about motive, whether the rewards and penalties point the holder toward acting on the evidence or toward protecting an operating line.
Durable is the succession question, whether the authority would outlast whoever holds it now.
Run together, at a single moment, they tell you the function exists and is sound.
What none of them reads is whether the function is actually being used.
Both backed and durable are load tests, as each one assumes the authority is under pressure and asks whether it holds. Take re-litigation, a decision the node has made comes back, has to clear the strategic gate above and win divisional buy-in below, but can be quietly unmade at the next forum. Backed authority is what survives that.
Transition is different, a shock rather than a pressure. The sponsor leaves, and the question is whether the function walks out with them. Durable authority is what survives that. Notice what the two have in common: they fire only when the authority is doing something, and something outside it pushes back.
Incentive-aligned sits closer to the failure this issue is about, but it is still a snapshot. It will check whether the incentives are arranged to push the holder toward exercise. What it cannot check is whether a holder who is correctly incentivised on paper is actually exercising anything. The two come apart. A function can have its incentives set up exactly as the condition demands and sit idle for a year, because alignment that is designed but never tested against a live decision is a claim about the wiring, not a record of what the wiring did.
And here is the limitation, all three conditions read the function as a state, at a moment. None reads it as a practice, over time. A function that passes all three in March can be hollow by December, with nothing having been done to it at all. No relitigation, no transition, no change to the incentive design, just disuse.
The authority is backed, it would survive a transition, its incentives point in the right direction, and it decays anyway from within because the node never exercises it.
Overall performance is unchanged, yet the authority it had is gone, let us call it decay by disuse.
It is the maintenance failure of a function that passed every existence test, and it is invisible to a diagnostic that reads whether the function is there, never whether it is being spent.
Why does a held mandate decay?
Decay by disuse is not a lapse of will, and not a character flaw in whoever holds the seat. It is structural, with three mechanisms that produce it, and they run whether or not anyone is watching.
The first is that authority runs on belief, not on the organsiational chart.
The chart can record that a node holds stop authority. But it does not record whether that authority is being exercised on something the chart cannot show, namely, whether the organisation believes it will be used. A kill right everyone expects to see used changes behaviour long before it is invoked. Options get wound down early. Resources move before anyone has to force them because the system reads the call in advance. A kill right nobody expects to see used changes nothing at all, because the system has already priced in that it is not coming.
Every decision moment that passes without the authority being exercised is evidence to the organisation that it will not be exercised, and therefore opportunities will be lost. Over time, the belief in the function will erode. This is why the decay cannot be counted in months. A mandate can sit through a slow year and stay intact if nothing tests it, then be hollowed out in a single quarter if it declines three live decisions in a row. The unit of decay is the unexecuted / unexcercised decisions, and the organisation is counting them even when the holder is not.
The second mechanism is friction.
Non-use is always the cheaper path, and the same incentives give an enterprise its pace, as well as product non-use. As a result, the incentive-aligned conditions fail in motion even where they hold on paper, because the two costs arrive on different clocks. The previous issue established that an area-level stop is among the most politically costly decisions in the innovation system. This results from retiring a leadership-endorsed theme, which tells people who championed it that the evidence has turned. The bill arrives at the moment the authority is used. The bill for not using it is deferred, spread thin, and charged to no one in particular this quarter.
The holder facing each kill moment faces an asymmetry. Acting is expensive now where as not acting is often perceived as being “free.” The mandate is rarely lost through a decision to abandon it. It goes through a run of individually rational deferrals, each one defensible, each on leaving the authority a little less useful, a little less credible than the last. The holder never decides to give up the mandate. They only ever decide to wait.
The third mechanism is that organisations govern the function as a state to be reached rather than as one to be held.
The readiness assessment is a snapshot, and a snapshot invites a snapshot response. When the function is judged to exist, one name across four capabilities, backed and durable, the organisation books the result and stops watching. Unless it is worked in, the diagnostic has no re-run cadence, and there is nothing in the design forcing the authority to be re-earned once it has been granted.
Absent a renewal mechanism, the default is continuation. The function persists in the cart until something removes it, and disuse removes nothing visible; therefore, nothing ever triggers a second look. This is the condition that fuels decay. It lets the authority sit unexcercised indefinitely while the record of its existence stays clean.
Together, the three compounds. Belief and credibility are the substrate, so the authority is only ever as real as its expected use. Every deferral lowers what the system as a whole and the organisation expect; however, there is nothing to catch it. As a result, the function does not fail in any apparent way, with no flags on any formal indicators.
An Operator Note
Over six months observed a corporate spin-out that no one would either back or stop, from the next seat over. A year in, the theme above it still made sense, so it held its place as the only option anyone was pursuing under that theme, while the authority over it showed up as high-level performance monitoring and nothing else, no real contact with the potential customers or the division sponsor who would have had to take it on. Nobody killed it and nobody resourced it enough to find out whether there was anything there, so it was watched from altitude over another six months and the watching carried on after.
The occupied vacancy
Decay by disuse does not end in an empty seat. An empty seat would at least be noticed and maybe filled, the way any obvious vacancy pulls in a replacement. What disuse produces is more detrimental because it appears to be an occupied role.
Even if the holder does not exercise authority, the holder can still have influence, where they can still argue a case and advise. But what they can no longer do is decide and have the decision hold, because the belief and credibility that make the decision binding are reduced or absent. The seat is still theirs, the title and the standing invitation to the meeting continue, but influence, however senior, does not convert a signal into a structural commitment. It proposes conversations for other people to ratify, which is the tolerated-authority failure from Issue 12.
Reversion itself can be hard to catch because a vacant function is legible, but one hidden by influence and conversation indicates nothing is missing. The organisational chart accurately lists who holds the title and is silent on whether the title still carries weight.
A second cost is realised on the altitude up, on the resource, the authority had the right to move. The last issue is the opportunity area that persists after its hypothesis has been falsified, perhaps by the zombie pilot related to it. Decay by disuse is how that zombie pilot is made. An authority that will not exercise an area-level stop cannot retire the area, so the area keeps its claim on people and budget. The resource that should have been made available and most likely repurposed to the next thing stays trapped in a theme that is no longer alive.
Most often, the successor opportunity is the one that pays for this the most. Sometimes it will not be considered at all, because the slow is still occupied. Sometimes it cannot be funded to its next phase because the capital is already committed or is pegged to something the evidence no longer supports.
Reverting to influence only does more than fail to act. It will become a blocker of the organisation’s scarce resources devoted to the innovation agenda. Those resources often become trapped on a falsified hypothesis, preventing the next opportunity from being started.
The cost of the unexecuted authority does not stop at the the dead theme, it is the live one that was never explored.
Maintenance, not achievement
If decay by disuse is structural, the fix has to be structural too.
Telling the holder to be more decisive does nothing. It is just like saying “be more innovative." The problem is an incentive asymmetry atop a belief substrate that erodes through non-use, within a governance design that only checks for existence. The fix works on the design, not the person.
Five moves to consider:
Look to invert the default, where the function must be built so that continuation is what must be earned, not termination. Under the standard design, a theme continues until someone summons the cost to stop it. Invert it, and a theme continues only if it is actively re-justified, so silence reads as lapse rather than consent. This does not make the organisation more aggressive; it shifts the burden of proof from the person trying to stop a theme onto the theme itself, which is where it belongs, because resources are finite and every theme is spending them.
Second move to turn the snapshot into a cadence. The readiness assessment should not be run once and filed; it should be a recurring gate that the function passes through, at which the authority is re-earned by being re-exercised. An authority used on a schedule does not decay, because each exercise is a fresh deposit into the belief that it is real. This gate, however, will only work if renewal carries a genuine evidentiary burden. A checkpoint where the holder has to do nothing more than not lose the authority is a rubber stamp, and a rubber stamp is a disused wearing the costume of governance. Renewal requires evidence that the theme is still live, not merely the absence of an objection.
However, cadence raises an immediate design question about whether the review fires on a fixed schedule or only when a live decision moment occurs. The binary is a trap because a purely event-driven trigger assumes the live moment will show up, and it usually does not. The moment arrives, no one names it, and it passes in silence, which is the whole mechanism of decay by disuse.
A fixed cadence is not there to manufacture urgency; it should be used to force surfacing. The schedule review does not ask whether the mandate should be renewed in the abstract; the question that a rubber stamp waves through. It asks what decision moments arose in the period that were not acted on. The risk it guards against is not manufactured urgency but manufactured calm, a period that looks quiet only because no one surfaced the tests that occurred.
If the trigger is built as both a fixed floor cadence and a backstop, and event triggers that fire the moment a live call appears.Focused on what a cadence leaves open, what each renewal is measured against. Split the renewal by altitude, using the two tiers already defined in Issue 12. The function operates at two altitudes, and the two decay at different speeds against different evidence, so a single undifferentiated renewal tests nothing precise.
Opportunity Authority renews against the hypothesis: does the claim about where value exists still hold, on a slow clock, with thesis invalidation as the kill criterion?
Option Authority renews against progression: is this specific bet advancing toward its proof point, on a fast clock, with failure to progress as the kill criterion?
Conflate them, and you get both errors the portfolio already suffers from. A sound opportunity is killed because individual options are stalled, or a dead opportunity survives because one option still shows a pulse. Separate them, and each renewal has its own hypothesis, its own metric, its own cadence, and each becomes answerable rather than ceremonial.The fourth move governs what happens to the authority and the resource that a lapse releases. Issue 12’s rule that a stop must sit above the work it stops carries a corollary that the renewal design makes unavoidable: released authority has to lapse to a named node, not to the organisation at large.
The organisation at large is nobody. Authority that reverts to a collective evaporates rather than transfers, which recreates the vacancy the function was built to close, this time by design. A stalled option lapses its freed resource to the opportunity owner, who reallocates across the option set. A dead opportunity lapses to the portfolio-level node accountable for capital allocation across areas. Different recipients at each altitude, both of them named, neither of them the organisation.The last attaches a cost to the answer as every prior move designs a moment at which the holder must act or decline to act. However, none of them will be measurable if declining is free. It restates the incentive asymmetry, now at the level of the gate. If sitting on a decision carries no consequence, the rational holder sits, and the cadence becomes the rubber stamp it was built to prevent.
The cost has to be written for the quality of the call, not its existence, because the crude version introduces the opposite failure. A holder paid for exercise as a count is paid to kill, and will retire live themes to post one, which is decayed by disuse and more destructive, because it ends what was working. The measure is action, the evidence supported, in either direction: a justified stop and a justified continuation both pass, and an unjustified kill fails exactly as a condemned deferral does.
Spend-and-kill rights inherited with the role, and compensation tied to acting on the evidence rather than to the survival of the portfolio, are what convert the renewal gate from a ceremony into a decision the holder cannot avoid. Without the cost, the other four moves are a process no one is penalised for ignoring. With the wrong cost, a process that manufactures churn.
Taken together, the five moves convert the function from something an organisation achieves once into something it maintains continuously. A built function that is never maintained is not a permanent capability but a depreciating one, and the depreciation is silent.
The Mandate Maintenance Test
The readiness assessment in Issue 12 tests whether the function exists. The Mandate Maintenance Test assumes the function has passed that assessment and asks whether it is still relevant or has begun to revert to influence, thereby decaying. It is built to be run by someone above the holder, not by the holder alone, because a node in decay is the least able to surface the moments it lets pass.
Run it on the same opportunity area, and answer based on what has actually happened, not from what the design intends.
Q1: Last exercise.
When did this node last exercise stop authority against resistance, retreating from an area or killing an option that other people wanted to keep, rather than cancelling something already dying on its own? Name the instance and the date.
Failure mode: Reversion, if no instance comes to mind within living memory, the authority exists on paper and has not been spent. It has most likely already degraded to influence, whatever the chart says.
Q2: Unused moments.
In the last twelve months, how many moments arose where the evidence justified a stop and the stop was not made? Name them.
Failure mode: Deferral accretion. Each unmade call is a deposit against the belief that the authority is real. A short list is maintenance, while a long one, or an inability to reconstruct the list at all, is decay in progress.
Q3: Renewal cadence.
Is there a scheduled point at which the mandate must be actively re-justified to continue, or does it continue until someone objects? Name the cadence, or name its absence.
Failure mode: Continuation by default. With no renewal point, silence reads as consent, which is the exact condition disuse needs to run unchecked.
Q4: Evidentiary burden.
At that renewal point, does the holder have to present evidence that the hypothesis is still alive, or is renewal a formality no one has ever failed? Name what renewal requires.
Failure mode: Rubber stamp. A gate that tests nothing is out of use in the context of governance. The presence of a checkpoint proves nothing if the checkpoint cannot be failed.
Q5: Cost of sitting.
Does declining to exercise a live call cost the holder anything, or is sitting on the decision free? And does that cost fall on unjustified action as heavily as on inaction, or only on inaction? Name what a deferral costs, what an unsupported kill costs, and to whom.
Failure mode: Costless deferral, or one-sided cost. If sitting carries no consequence, the rational holder sits. If only inaction is penalised, the holder overkills to be seen acting. The gate binds correctly only when it prices the quality of the call in both directions.
Q6: Lapse destination.
When authority or resource is released, does it pass to a single named node accountable for redeploying it, or does it revert to the organisation at large? Name the recipient.
Failure mode: Disguised vacancy. Released authority that lapses to no one evaporates rather than transfers, recreating the gap the function was built to close.
Interpreting the results, the existence test and the maintenance test are separate reads, and a function can pass one and fail the other. A node that cannot name a recent exercise on Q1, has no cadence on Q3, and lapses to nobody on Q6, has reverted to influence regardless of how cleanly it passed the readiness assessment. The chart will not tell you this. The organisational chart records the title, and the title is the last thing to go.
Two objections
Two objections are worth stating fairly because both raise real concerns and are resolved once the design is understood rather than assumed.
The first is that constant renewal is just bureaucratic churn, and that making a theme re-justify itself every period erodes the very durability that made the function worth building. But this conflates renewal with re-litigation, even though the two are opposites.
Re-litigation is other parts of the system reopening a decision the node has already made, the tolerated-authority failure, authority revisited until it dissolves. Where as renewal runs the other way. The node re-earns its own continuation, on its own cadence, with evidence, which is simply the authority being used. And, subjecting a decision to external veto erodes the function; making the holder act sustains it.
A renewal cadence, run properly, is not overhead bolted onto the authority. It is the authority in use, and use is what keeps it from reverting. The churn objection is really a description of a badly designed renewal, one that has drifted into a standing invitation for others to reopen settled calls. That is a failure to hold the line between renewal and re-litigation. It is not an argument against renewal.
The second worth noting, is that making lapse the default will kill good long-horizon bets before they mature, punishing exactly the patient capital corporate venturing is supposed to protect.
This is a legitimate concern and the design answers it directly. The renewal test runs against the hypothesis and the progression, not against short-horizon results.
A long-horizon bet re-justifies on whether its thesis still holds and whether it is advancing toward its proof points, both of which a genuinely patient bet can demonstrate. Inverting the default does not shorten the horizon, it forces the horizon to be defended rather than assumed, which is different.
The evidentiary burden is what keeps it honest: a bet survives on demonstrated advance toward its proof point, not on the quality of the story told about it, because a well-told thesis and a live thesis are not the same object and the gate is built to tell them apart. A long bet that cannot show it is still live was never patient capital. It was a zombie with a longer story, and the renewal gate is precisely what separates the two.
Built is not kept
Issue 12 argued that a portfolio for which no one is accountable for conversion is a list with a budget and a sponsor, and that the function is what remains when the budget and the sponsor are not enough.
There is a second half to this argument addressed in this issue.
The function is not enough either if it is only built and never maintained. Building the conversion layer is the achievement of the readiness assessment measures. Keeping it is a separate discipline that the assessment cannot see, because authority is not a possession that stays where it is put. It is a claim the organisation honours only for as long as it is exercised, and it drains quietly, without a triggering event, in exactly the organisations most confident they have already solved for it. The most dangerous function is not the one that fails the existence test. It is the one that passed eighteen months ago and has not been asked the question since.
A portfolio without kill discipline is a list.
A portfolio without resource velocity is a list with a budget.
A portfolio no one is accountable for converting is a list with a budget and a sponsor.
And a portfolio whose conversion authority is held but never exercised is the most expensive version of all, because it carries the full cost of the function, the seat, the salary, the governance time, and delivers none of the conversion, while every formal indicator reports that the problem is solved.
Authority that is not exercised reverts to influence. Silence reads as a lapse, not consent. A function you do not spend is a function you no longer have.
Ideas do not scale. Execution does. And, execution is not a structure you install once. It is a discipline you spend, or it spends down without you.
This is Issue #13. It extends the Venture Operator Function, introduced in Issue 12 as the capstone of the Corporate Venturing Under Constraint arc, from the question of whether the function exists to the question of whether it lasts. The Mandate Maintenance Test above is designed to be run alongside the Issue 12 readiness assessment, on the opportunity area in your portfolio that has gone longest without an exercised stop.
This essay is based on the author’s professional experience and interpretation of publicly available information. It is provided for general informational purposes only and does not constitute advice. Any views expressed are the author’s own and do not refer to any specific organisation, program, or individual.
References & Further Reading
The Venture Operator. “Issue #12: The Venture Operator Function.” Put Ideas To Work, 9 June 2026. Available at: https://www.putideastowork.com/p/issue-12-the-venture-operator-function
The Venture Operator. “Issue #8: Zombie Pilots.” Put Ideas To Work, 9 April 2026. Available at: https://www.putideastowork.com/p/issue8-zombie-pilots
The Venture Operator. “Issue #10: Resource Asymmetry.” Put Ideas To Work, 5 May 2026. Available at: https://www.putideastowork.com/p/issue-10-resource-asymmetry
The Venture Operator. “Issue #11: When Exposure Outpaces Governance.” Put Ideas To Work, 20 May 2026. Available at: https://www.putideastowork.com/p/issue-11-when-exposure-outpaces-governance