HOW WE GOT HERE

Today’s problems come from yesterday’s solutions

Peter Senge’s line describes mining closely. As we see it, over a century the industry took the spare capacity out of its mines, one sensible decision at a time. This page traces how that happened and where it has left the industry in 2026.

Talk it through in 30 minutes

Run the Best-Day Diagnostic

THE INDUSTRY’S OWN LIST

Mining’s ten biggest risks for 2026

Each year EY asks mining and metals executives to rank the risks to their business. For 2026, operational complexity reached No. 1 for the first time, and rising costs and productivity climbed from No. 6 to No. 2.

SHAPED LARGELY BY HOW THE MINE IS RUN

Six of the ten

No. 1 Operational complexity, first time at No. 1
No. 2 Rising costs and productivity, up from No. 6
No. 3 Capital
No. 5 Licence to operate, unchanged
No. 6 Workforce, up from No. 13
No. 8 Digital and innovation, back on the list

MOSTLY OUTSIDE THE GATE

Four of the ten

No. 4 Resource and reserve depletion
No. 7 Geopolitics, down from No. 3
No. 9 Sustainability, down from No. 2
No. 10 Changing business models

Ranking by EY. The split into two groups is our own reading.

EY’s survey listed twelve bottlenecks. Each was chosen by between 19 and 31 per cent of respondents, who could pick up to three.

EY observed that miners gave almost equal weight to many different factors, from regulatory delays and labour to plant capacity and grade. We read that spread as a clue and no more. Executives across many mines picked from a set list, so an even spread says nothing certain about any one mine. In our experience, people at a mine with one stable bottleneck name it quickly, and where it moves from shift to shift they name several.

EY also warns that higher prices for some commodities have lifted revenue while hiding underlying productivity losses.

TWELVE YEARS EARLIER

EY described the same problem in 2014

In 2014 EY reported that mining productivity had been falling since about 2000. In its report Productivity in mining: now comes the hard part, Paul Mitchell, EY’s global mining and metals advisory leader, traced much of the decline to the boom, when volume came first and cost came second. EY described how hard it had become to run larger, more complex operations, where silos had grown and the parts of the business had lost touch with each other. One miner called it diseconomies of scale. EY’s answer was to manage the business end to end.

Two years later Barry Urban of TOCE Consulting and Hendrik Lourens wrote that EY had put its finger on the cause. They then asked why the slide was not turning, when executives knew the issues and were spending heavily on respected improvement programmes.

Up to 28 per cent: the fall in global mining productivity over the decade to 2015, after adjusting for ore grade and cost inflation (McKinsey, 2015).
45 per cent: the fall in the capital productivity of Australian mining since 2000, against 22 per cent for all other industries (EY, 2014).
About 25 per cent: how far global mining productivity remained below its mid-2000s level in McKinsey’s 2020 analysis, after gains since 2013 that came from holding down capex and opex.
No. 1: operational complexity on EY’s 2026 list of mining risks, the first time it has led (EY, October 2025; Mining.com).

In Australia, mining was among the industries with the largest productivity falls in 2024-25, on Ai Group’s reading of ABS figures. Twelve years after EY described the strain of running larger, more complex operations, operational complexity now tops its list.

A SHORT HISTORY

How mining took out its buffers

A buffer is whatever lets one part of a mine keep working when the part next to it has a bad hour. It might be a stockpile, a spare machine, critical spares close to the work, a planner with time to look ahead, or slack in the schedule. Operations science has a law for this: variability is always paid for, in stock, in spare capacity or in time.

Each step below made sense in its day. We would argue each one also treated buffers a little more as waste.

1911

Improve each part

Frederick Taylor’s scientific management broke work into measured tasks and improved each one. In stable, repetitive factory work it was a real advance.

Left behind: the instinct that if every part is improved, the whole will follow.

1920s

Manage by the numbers of each part

DuPont’s finance team and Alfred Sloan’s General Motors made return on capital and cost per unit the language of large companies, with efficiency measures for each division.

Left behind: the habit of judging the whole by the efficiency of its parts.

1925 onwards

Cost accounting and the budget

Budgets began as a way to plan cash. We would argue the habit of treating a dollar saved in any department as a dollar earned dates from a time when most cost rose and fell with output. In Relevance Lost (1987), Thomas Johnson and Robert Kaplan argued that most management accounting then in use had been in place since about 1925.

Left behind: managers judged department by department on their variance to budget.

1988 to 1990s

Activity costs and lean

Activity-based costing (Cooper and Kaplan, 1988) traced cost to every activity so the fat could be found and cut, and benchmarks rated each operation against standards and its peers. Lean, drawn from Toyota and popularised in 1990, taught a generation that inventory and waiting are waste. Hendrik saw lean work well in a manufacturing plant he ran earlier in his career, because the process could be held tightly. In a mine much of the variation comes from the ground and the weather, and no standard holds it still.

Left behind: stockpiles and spare machines filed as waste.

1990s to 2000s

The ERP

SAP, widely used across mining, started in 1972 as a financial accounting system. ERP systems put budgets and approval rules into software. Spending or stock outside the budget is flagged at once and needs a senior override.

Left behind: protective capacity that shows up as a variance to be corrected.

Mid-2000s to 2012

The boom

With prices high, volume came first. Operations grew larger and organisations were scaled up to match. McKinsey’s index shows productivity falling from 2004 to 2009, and EY later traced much of the decline to this period.

Left behind: bigger, more siloed operations that were harder to coordinate.

2012 to 2016

The cost-out

When prices fell, budgets were cut to match. In our experience the roles cut hardest were the ones that connect work rather than perform it: planners, superintendents, coordinators and discipline leads. In our experience many sites also ran down stockpiles to free working capital and sized fleets to the plan. Productivity began to recover from 2013, and McKinsey found the gains came from holding down capex and opex.

Left behind: thinner organisations and thinner buffers. In our experience few of those roles came back when prices recovered.

2016 to 2026

Operating systems and digital tools

Management operating systems brought rhythm and accountability, often with a cascade of targets for every department. Digital tools gave each area a far richer view of its own performance.

Left behind: well-governed parts, and a whole that is still hard to predict. In 2026 operational complexity tops EY’s list.

As we see it, the mine without buffers was built one sensible decision at a time, by capable people doing what they had been taught was good management.

ON SITE

Where the buffers used to sit

In our experience these are the buffers a mine most often gives up, and what it pays instead.

Run-of-mine and intermediate stockpiles

Why it went: stock looks idle and ties up working capital.

What the mine pays instead: the crusher or plant waits whenever the pit has a bad shift.

A spare winch, loader or drill on hand

Why it went: capital rules price the machine and miss the hours it protects. See the winch story.

What the mine pays instead: days of lost blasting or loading while one breakdown is repaired.

Critical spares close to the work

Why it went: stock above standard is flagged in the ERP.

What the mine pays instead: crews wait while parts come from a central store.

People who connect the work

Why it went: they do not appear on the mine plan, so they go first in a downturn.

What the mine pays instead: supervisors spend the shift reacting, and the plan fails by midweek.

Slack in the schedule

Why it went: plans are built to average rates.

What the mine pays instead: weekend catch-up and overtime.

What a chain without buffers delivers

Picture three machines in a row, a loader, a truck and a screen, each rated at 100 tonnes an hour and each available for 8 hours of a 10-hour shift. Most people expect 800 tonnes. In the worst case, where each machine’s two hours of downtime fall at different times and nothing sits between them, the chain runs for four hours and makes 400 tonnes. Every machine met its availability target.

AT THE MINE

The same list, one shaft at a time

In their 2016 paper, Barry Urban and Hendrik Lourens told the story of Eugene, a mine manager running a shaft. Every Monday for five years he and his twelve mine overseers went through the previous week’s figures, and the picture hardly changed.

Each line on the report had its own owner and its own remedy. Eugene blamed the overseers, and the region blamed Eugene, for figures nobody in the room could move alone.

Two of EY’s ten, rising costs and productivity and the workforce, are on his list too. We read EY’s list as Eugene’s Monday meeting held across a whole industry.

Six phone calls

The same paper records what happened on an underground platinum mine when the financial controller of one shaft agreed to hold stores underground to keep work flowing. The ERP flagged the extra stock at once. Within an hour she had six phone calls from different parts of the organisation, each asking what she thought she was doing.

The paper reads this as the system pushing back: it was built to remove exactly the kind of buffer she had just created.

Monday production meeting

The figures, week after week

Production tonnes

flat

Operating costs

rising

Equipment breakdowns

rising

Safety record

worsening

Absenteeism

rising

Overtime

rising

Weekend catch-up

more often

Turnover of technical staff and supervisors

rising

From Urban and Lourens, 2016.

YESTERDAY’S SOLUTIONS

How each solution turned into a risk

This is how we connect the history to EY’s list. It is our argument, drawn from practice and from the sources below, and we offer it for you to test against your own operation.

EY NO. 1: OPERATIONAL COMPLEXITY

Budgets built back from the target, each area resourced to its average share

What it removed: spare capacity at every step.

Today’s problem: a bottleneck that moves from shift to shift, and a plan that fails by midweek.

Read: Operating Executives

EY NO. 2: RISING COSTS AND PRODUCTIVITY

Utilisation and cost targets for every department

What it removed: any reason to run an area below its maximum to keep the bottleneck fed.

Today’s problem: departments on target while the mine misses, so cost per tonne rises because the tonnes fall.

Read: Lean and OEE

EY NO. 3: CAPITAL

Approval rules and variance reporting

What it removed: local authority to hold a buffer, or to spend to protect output.

Today’s problem: output that swings, so forecasts are discounted on the way to the board.

Read: Capital and the Bottleneck

EY NO. 5: LICENCE TO OPERATE

More controls and verification asked of a stretched operation

What it removed: supervisors’ time and attention.

Today’s problem: control checks and production competing for the same people. Our working hypothesis is that this is part of why the fall in Australian mining fatality rates has stalled.

Read: Fatality Plateau

EY NO. 6: WORKFORCE

Cost-out after the boom

What it removed: the people who connect the work, and spares close to it.

Today’s problem: fewer people carrying more conflicting demands, with the experienced ones leaving first.

Read: Let them play golf

EY NO. 8: DIGITAL AND INNOVATION

Technology laid over the same operating model

What it removed: no buffers restored.

Today’s problem: a better-measured version of the same shortfall.

Read: Why digital is a risk factor

The other four risks on the list sit mostly outside the gate. The next section deals with them.

THE LIMITS

What the operating model does not explain

Grade decline and depth are geology. Commodity prices and geopolitics are set outside the gate. The energy transition is changing demand, and reserves deplete however well a mine is run. No change in how a mine is managed reverses any of that, and we do not claim it does.

What the operating model does decide, we think, is how well a mine copes. A mine that loses part of its capacity to its own design feels a falling grade sooner and harder, and a mine that can commit to a number and keep it goes to the capital market from a stronger position. No mine controls the weather, though how it is run decides how seaworthy the vessel is.

THE WAY BACK

Put the buffers back where they protect output

The way back keeps a century of good management in place. The MOS and the ERP both stay, and the change is in where the spare capacity sits and who can act on it.

The mine names its bottleneck and protects it, with stock in front of it and spare capacity in the areas that feed it. The people closest to the work meet for 30 minutes a day to coordinate around it. We call that meeting the Flow Room. Through many interventions we have found that in the absence of a Flow Room, the gains fade.

At an Australian longwall mine, annual production rose from a 2.4 Mt average (2018 to 2023) to 3.2 Mt in 2024, a 33 per cent increase, with no additional people and very little capital. Read the longwall case

FOR OPERATIONS

Operating executives

Run the mine around its bottleneck, inside the MOS you already have.

Go to the page

FOR OWNERS

Capital owners and boards

Test the capacity an asset already owns before funding more.

Go to the page

FOR SAFETY LEADERS

Capacity to be safe

Our working hypothesis on predictability and the attention supervisors need for safety.

Go to the page

TRY IT

Mine simulation

Five minutes with one small mine and two scoreboards. Most people predict the wrong number.

Play the simulation

START WITH YOUR OWN DATA

See what your mine has already shown it can do

The Best-Day Diagnostic compares everyday output with your mine’s own best days. In our experience a gap above 25 per cent is worth a closer look. It is a screen, never a target.

Talk it through in 30 minutes

Run the Best-Day Diagnostic