Underpitching

19 Sep, 2026

The staffing error that gets filed as a motivation problem

The mechanism, in ninety seconds

Pitching is adding yeast to wort. The pitching rate is cells per millilitre per unit of sugar, and it is decided in about four seconds by somebody holding a packet.

Underpitch and three things happen, in order.

First, the population spends itself reproducing before it can start the actual job, budding through more generations than it was designed for and producing thinner, less competent cells each round. Second, that stressed growth throws off higher alcohols - propanol, isobutanol, isoamyl - which taste like solvent and give you a headache before the second glass. Third, the batch stalls. The last sugars in the tank are the awkward ones, and an exhausted population drops out before it reaches them. The beer finishes sweet, cloudy, and faintly of butter, because nobody stayed sober enough to clean up.

Harsh output plus work that quits near the end. Two symptoms, one cause, and the cause was locked in before the first bubble.

Nobody has ever looked at a stuck fermenter and concluded that the yeast has lost faith in the mission. That courtesy is reserved for humans.

Lesson 1 - Output quality is mostly a function of a resourcing decision made before the work started. By the time you can taste the problem, you are tasting a decision, not an attitude.

How underpitching actually enters a company

It never arrives as a decision. Nobody presents a slide reading "we will resource this at sixty percent and absorb the flavour." It arrives as a sequence of individually defensible deferrals, each one signed off by a reasonable person having a reasonable week.

• The hiring freeze, described as a pause, which is a freeze with better manners
• The backfill that never lands because the team covered the gap so gracefully that the gap stopped being visible
• Attrition booked as savings rather than as capacity leaving the building
• The reorg that turns a team of nine into a team of six while the OKRs are copied and pasted forward untouched
• Span of control targets set by somebody who has never met the span
• Scope that grows continuously after the headcount number was fixed, which is the same as cutting the team without touching the team
• Contractors brought in as cells when they are in fact an adjunct, requiring supervision from precisely the people you were trying to relieve

The last one deserves its own paragraph, because it is the most common expensive mistake in the list. Adding three contractors to a stretched team of four does not create a team of seven. It creates a team of four with a new management responsibility and a knowledge transfer obligation, and it will look like a rescue for about five weeks.

Lesson 2 - Underpitching is rarely chosen. It accumulates. Audit the gap between the date the headcount was set and the date the scope was last changed, because that gap is your real pitching rate.

The stress flavours, translated

An understaffed team does not produce less work at the same quality. That would be legible, plannable, and easy to escalate. What it produces is work with a taste.

Every phrase in the right hand column is a real thing that real leaders say, in good faith, having correctly observed a symptom and then attributed it to the substrate rather than the census.

Lesson 3 - When output quality and morale degrade together, do not treat them as two problems. Codependent symptoms usually share a cause, and the cause is usually upstream of both.

The stall at ninety percent

Fermentation does not fail at the difficult sugars. It fails at the last ones. Corporate work has exactly the same shape, and the last ten percent has a name in every industry - the data migration, the security review, the compliance sign off, the accessibility pass, the error states, the training material, the runbook, the thing that turns a demo into a product somebody can actually operate on a Tuesday.

That work is unglamorous, unshowable, and requires the most vitality of anything in the project. It arrives at the precise moment there is least of it available. Which is why so many underpitched projects are described as ninety percent done for a period of time longer than it took to reach ninety percent.

The corporate cruelty is in the scoring. A project that ships at ninety percent is recorded as a delivery failure and attributed to the team. The staffing decision that guaranteed it is recorded nowhere at all.

Lesson 4 - Capacity planning that only covers the build phase is not capacity planning. Resource the finish explicitly, or accept that you are buying a demo at production prices.

Why underpitching keeps getting rewarded

Here is the part that explains the persistence of the whole practice, and it has nothing to do with brewing.

An underpitched batch looks excellent early. Longer lag, then a violent, impressive krausen. Somebody leans over the vessel and says look how hard it is going, and it is going hard, because that is what a small population at maximum effort looks like. Vigour is not health.

The organisational version is that the first quarter of an understaffed project is frequently the best looking quarter of that project. Velocity flatters the underpitch. Adrenaline and capacity are indistinguishable on a dashboard. The manager who absorbed the cut gets credited with efficiency in Q1 and the taste arrives in Q3, often after they have been promoted, moved, or reorganised into somebody else's problem.

This is not a culture failure. It is a measurement failure with a straightforward structure - the cost of an underpitch lands outside the window in which the underpitch is evaluated, and frequently on a different balance sheet.

Lesson 5 - Any organisation that reviews resourcing decisions on a shorter cycle than their consequences take to appear will produce underpitching indefinitely, no matter what its values statement says.

The misdiagnosis industry

Given a taste problem and a stall, the standard corporate response is to procure feelings.

An engagement survey goes out. It correctly detects that people are stressed and that scores have fallen in the dimension labelled "I have the resources I need to do my job well." This finding is then routed to a working group, converted into a resilience programme, and delivered back to the exhausted population as a set of coping techniques. There is an offsite. Somebody says the word ownership in a tone that implies it is missing rather than dangerously oversubscribed. A wellbeing app is licensed. Somewhere in the process, a person who has been carrying two roles for eleven months is placed on a performance plan for insufficient attention to detail.

None of this is malicious and most of it is sincere. It is chosen because it is cheap, fast, and relocates the problem to the least expensive address in the building, which is the interior of the affected employee.

Lesson 6 - You cannot fix an underpitch by encouraging the yeast. If the remedy addresses sentiment while the constraint is arithmetic, you have bought a placebo and billed it to the patient.

Diagnostics - underpitch or genuine performance problem

Both exist. They are distinguishable, and it takes about twenty minutes.

1. Where does work fail. Performance problems fail at the hard parts. Underpitching fails at the last parts. Timing is the single most reliable tell.
2. What is happening to variance. If throughput is roughly holding while defect rates and rework climb, you are not looking at slack. You are looking at stress in the output.
3. Who is on the critical path. If the same one or two names appear across unrelated workstreams, the team is single threaded and the headcount number is fiction.
4. Who is leaving. Voluntary attrition concentrated in the strongest performers is the organisational equivalent of early flocculation. Weak teams lose weak people. Underpitched teams lose the ones with options.
5. The transplant test. Ask honestly whether a known high performer, dropped into this team tomorrow with these constraints, would succeed. If the answer is no, the problem is not in the cells.
6. Compare two dates. When was capacity last set, and when was scope last increased. If the second is later than the first, stop the investigation. You have your answer.

Lesson 7 - Diagnose by timing and distribution, not by impression. Underpitching and underperformance feel identical in a meeting and look nothing alike in the data.

Interventions, ranked by how fast they actually work

The uncomfortable implication is that by the time an underpitch is visible, most of the intervention menu has expired. Scope is what remains. Leaders who cannot bring themselves to cut scope are not choosing to keep the scope. They are choosing the flavour and the stall, and calling it a stretch goal.

Lesson 8 - Late in a project, capacity is fixed and scope is the only real variable. Refusing to move it does not create capacity, it just delays the confession.

Overpitching, for the CFO in the room

None of this is an argument for infinite headcount, and the brewing metaphor is honest enough to say so.

Overpitch and you get a ferment with almost no growth phase, finishing too fast, producing beer that is technically flawless and completely characterless. Thin. Clean. Nothing to say. Teams do this too. A project with more people than problem produces work that is correct and forgettable, adds coordination cost, diffuses ownership until no single person can be found who feels responsible, and denies everyone involved the stretch that develops them.

Twelve people can ferment a remarkably sober beer.

Lesson 9 - The goal is calibration, not generosity. The right pitch is a number, and both errors are arithmetic. Only one of them is routinely reframed as a virtue, and it happens to be the one where the discomfort is borne by whoever is inside the tank rather than whoever wrote the number down.

What to actually write down

The whole argument reduces to one operational change, which is to make the pitching rate explicit and durable.

• State the capacity to scope ratio in the plan, in whatever unit you actually trust, and record the date it was agreed
• Log every subsequent scope addition against that date, so that the arithmetic remains visible and cannot later be relabelled as attitude
• Name the descope trigger in advance, along with the person permitted to pull it, because a trigger with no owner is a wish
• Resource the last ten percent as its own phase with its own number, not as a rounding error at the end of the build
• Review resourcing decisions on the timeline of their consequences, not the timeline of the quarter, or accept that you have institutionalised the error

And when you report it, report it in the only language that is difficult to reframe. Not that the team is struggling, not that morale is fragile, not that people need support. Simply that the work is underpitched, here is the ratio, here is the date it was set, and here is what will be cut.

Yeast is honest because it has no incentive to be otherwise. It cannot be coached, cannot be inspired, will not tell you it is fine, and does not care that the number was convenient. People can do all four, and reliably will, and that willingness is the only reason underpitching survives as a management style.

The lessons, extracted

1. Output quality is set by a resourcing decision made before the work starts.
2. Underpitching accumulates rather than gets chosen. Audit the gap between the headcount date and the scope date.
3. Degrading quality and degrading morale are one symptom, not two.
4. Resource the finish. Ninety percent is not a partial success, it is a full failure with better optics.
5. Organisations that review resourcing faster than its consequences appear will underpitch forever.
6. You cannot fix an underpitch by encouraging the yeast.
7. Underpitching fails at the end, underperformance fails at the hard parts. Use timing to tell them apart.
8. Late in a project, scope is the only lever left that works.
9. The right pitch is a number, and overpitching is also an error.
10. Write the ratio down, or it will be renamed as a character flaw when the bill arrives.