Project Friction: How Delivery Capacity Quietly Disappears
Large projects rarely lose capacity in one dramatic event. They lose it in small, respectable ways.
A discipline lead waits three days for a model review that adds no new decision. A contractor produces the same progress data in three formats because the client, the joint venture, and the parent company each want a different dashboard. A shutdown team holds daily coordination meetings, then spends the afternoon updating actions from the meetings. A commissioning engineer cannot close a punch item because two systems disagree on who owns the tag.
None of this looks like failure. It looks like governance, assurance, coordination, control, and professionalism. Some of it is necessary. In high-risk industries, weak control can kill people, damage assets, breach regulations, and destroy value. But some control reduces risk, and some control simply consumes delivery capacity.
That difference is the heart of project friction.
Project friction is the non-value-adding effort created by poorly designed management work. It includes duplicated reporting, excessive approvals, administrative handoffs, conflicting governance demands, overlapping digital systems, unnecessary meetings, interface churn, low-quality information, rework, and unclear accountability.
Experienced project teams recognize it. Fewer measure it. Fewer still manage it with the same discipline they apply to cost, schedule, quality, and safety.
That is a problem, because friction does not sit outside delivery. It uses the same people, attention, time, and decision bandwidth needed to deliver the job.
Project Friction Builds Across The Whole Lifecycle
Friction starts early. It rarely appears first during construction, even if construction is where it becomes most visible.
In concept selection, friction can begin with unclear decision rights. Teams produce multiple option papers without knowing who has authority to choose. Commercial, technical, operations, finance, and risk functions each ask for assurance, but the decision criteria remain unsettled. The project burns effort creating confidence rather than creating clarity.
In FEED, friction often grows through interface complexity. Engineering teams refine layouts, process design, brownfield tie-ins, long-lead equipment packages, permitting inputs, and cost estimates. If design maturity, estimate class, risk allowance, and contracting strategy do not move together, work loops back. A package may be “ready” from an engineering view but not ready for procurement, operations, constructability, or regulatory review.
Professional bodies have long warned against weak definition at the front end. The Construction Industry Institute’s work on front-end planning, AACE International’s guidance on cost engineering and estimate classification, and major project guidance from institutions such as PMI, APM, ICE, and RICS all point to the same principle: early clarity matters. Poor definition creates downstream waste.
Yet friction is not only caused by too little definition. It can also come from too much process wrapped around immature information. A FEED team may spend weeks formatting review packs, reconciling assurance comments, and building governance slides while major assumptions remain unresolved. The project appears controlled, but the team is processing uncertainty rather than reducing it.
During detailed engineering, friction shows up as review congestion. Documents move through discipline checks, inter-discipline checks, client reviews, third-party reviews, vendor reviews, regulator comments, and classified comment responses. Some reviews catch real technical risk. Others repeat the same checks because no party trusts the previous one.
In EPC and construction programs, friction often sits at the contractor-client boundary. A contractor submits RFIs, technical queries, method statements, inspection test plans, change notices, and payment applications. The client asks for more detail, more substantiation, or more alignment with reporting templates. The contractor responds defensively because commercial exposure sits behind every exchange.
Both sides then build parallel organizations to manage the interface. The project gains expediters, coordinators, document controllers, change analysts, claims support, reporting teams, and approval trackers. Some roles are vital. Others exist because the process itself has become hard to navigate.
In procurement, friction appears as duplicated vendor data requests, late technical bid evaluations, unclear approval thresholds, and misaligned engineering release dates. A long-lead compressor package, switchgear order, marine spread, or specialist fabrication contract can sit idle while teams debate exceptions that should have been resolved in the contracting strategy.
Construction brings a different pattern. The visible constraint might be labor productivity, but the hidden constraint may be access, permits, workface planning, scaffold availability, isolations, missing materials, late drawings, incomplete quality packs, or slow responses to field changes. On a brownfield modification or offshore upgrade, a crew can be fully staffed and still lose the day because the system around them cannot release work.
Shutdowns and turnarounds make friction especially clear. The clock is unforgiving. When work packs, isolations, SIMOPS controls, permit-to-work, materials, inspection resources, and decision authority do not align, crews wait. A single approval delay can affect critical path work. A single unclear boundary between operations and projects can ripple through nightshift productivity.
Commissioning, start-up, and handover then expose all earlier friction. Tags do not reconcile across systems. Mechanical completion dossiers are incomplete. Vendor documentation is late. Punch categories are inconsistently applied. Operations teams inherit procedures that do not match the asset configuration. The project may be “physically complete” but not operationally ready.
This is why project friction is a systems problem. It is not just bad admin. It is the result of interactions among governance, contracts, risk perception, digital tools, human behavior, commercial incentives, and organizational design.
Conventional Controls Often Miss The Capacity Drain
Most major projects have extensive controls. They track progress, cost, earned value, risks, issues, changes, quality, safety, and procurement. They report to boards, sponsors, lenders, regulators, joint venture partners, and internal committees.
So why does friction remain hard to see?
The first reason is that traditional measures focus on outputs and variances, not the effort required to produce control data. A weekly report might show engineering progress, construction quantities, procurement status, safety performance, cost forecast, risk movement, and change exposure. It rarely shows how many hours were spent creating, reconciling, reviewing, challenging, and reworking the report.
The second reason is that friction hides inside chargeable work. Contractor coordination hours, engineering review hours, document control hours, claims administration, expediting, planning updates, and management meetings all look like legitimate effort. They may be legitimate. The issue is whether their volume is proportionate to the risk they reduce.
The third reason is that schedules often model physical work better than management work. A Primavera or equivalent schedule can show engineering deliverables, procurement milestones, construction activities, completions systems, and start-up sequences. It may include approvals as activities, but it rarely captures the lived reality of comment cycles, decision latency, repeated reviews, competing approvers, or time lost to weak information.
The fourth reason is cultural. Senior teams often ask for more reporting when confidence falls. This feels sensible. If a project is under pressure, leadership wants visibility. The risk is that each new report adds extraction cost. Frontline managers become reporters of work rather than removers of constraints.
PMI and APM both emphasize tailoring governance and management effort to context. ISO 21502 also points toward project management practices that should suit the project environment rather than run as rigid bureaucracy. In high-risk sectors, that idea matters. A subsea installation campaign, an LNG train, a transmission upgrade, a hospital infrastructure program, and a mining process plant do not need identical governance rituals.
Yet many organizations layer corporate requirements on top of project requirements. A capital project may answer to a business unit, central projects function, asset owner, investment committee, finance group, HSE function, regulator, insurer, lender, and joint venture. Each group has rational needs. The total load may become irrational.
The test is not whether a report, approval, or meeting has a purpose. The test is whether it improves decisions, reduces risk, or protects value enough to justify the capacity it consumes.
Assurance can suffer from the same problem. Technical assurance, process safety reviews, constructability reviews, HAZOPs, model reviews, independent cost reviews, stage gates, peer assists, and readiness reviews can all add value. In oil and gas, energy, utilities, marine, and chemical processing, many are essential. Guidance from bodies such as IOGP, the Energy Institute, regulators, and professional institutions reflects hard-earned lessons from major incidents and asset failures.
But assurance loses strength when every review becomes a broad search for comfort. If reviewers reopen settled decisions without clear thresholds, or if comments carry equal weight regardless of risk, assurance becomes a rework engine. Good assurance focuses attention. Poor assurance spreads it thin.
The Main Ways Friction Consumes Delivery Capacity
Friction has many sources, but several patterns repeat across capital projects.
Duplicated Systems Create Hidden Reconciliation Work
Digital tools should reduce effort. Too often, they multiply it.
One system holds the schedule. Another holds cost. Another tracks documents. Another records completions. Another holds engineering tags. Another manages procurement. Another handles change. Contractors bring their own tools. Vendors submit through portals. Operations maintains the asset register in a separate platform.
The result is reconciliation work. Teams compare tag numbers, document revisions, quantities, dates, cost codes, discipline codes, contract packages, and handover boundaries. When systems disagree, people spend time deciding which version is true.
Poor information quality then drives poor decisions. A construction manager may not trust progress data. A commissioning manager may not trust system completion status. A project director may not trust the forecast. More meetings follow. More reports follow. The root problem remains data integrity and ownership.
This is not a technology problem alone. It is an information governance problem. ISO management system thinking, quality management principles, and asset management practices all rely on clear ownership, controlled information, and defined processes. If the organization has not defined what information matters, who owns it, and where the master record sits, software will not solve the issue.
Approval Chains Delay Decisions And Dilute Accountability
Approval stages often begin with good intent. They prevent unauthorized spend, unsafe work, scope drift, technical error, commercial leakage, and regulatory exposure.
The problem comes when approval chains grow without design. A field change may need endorsement from construction, engineering, contracts, planning, cost, quality, operations, HSE, and the client representative. Each approver looks at a narrow slice. No one owns the total decision cycle.
On a brownfield piping modification, for example, a small design change may affect stress analysis, materials, isolation planning, welding procedure, inspection scope, and commissioning sequence. That deserves proper review. But if the approval path treats every change with the same weight, low-risk decisions sit behind high-risk ones. The urgent and the trivial compete in one queue.
Decision latency carries real cost. Crews wait. Equipment stands idle. Specialist vessels burn day rates. Shutdown windows shrink. Contractors price delay risk into claims and future bids. The project pays for indecision, sometimes more than once.
Reporting Burdens Pull Leaders Away From Constraints
Reporting is necessary. Sponsors need visibility. Lenders and joint venture partners need confidence. Regulators may require evidence. Internal governance bodies need facts.
But reporting becomes friction when the same data is repackaged for different audiences, with different cut-off dates, definitions, and narratives. A project controls team might spend the first half of the week closing last week’s report and the second half preparing for next week’s review. Functional leaders then spend evenings validating slides rather than resolving blockers.
The hidden loss is management attention. Experienced construction managers, engineering managers, commissioning leads, package managers, and contract managers create the most value when they remove constraints and make trade-offs. If they are trapped in reporting cycles, the project loses experienced judgment at the point of need.
Meetings Become A Substitute For Clear Process
Meetings are often the visible symptom of unclear accountability. When roles, thresholds, and workflows are unclear, organizations add coordination.
A major infrastructure program can easily build a meeting architecture that looks rational on paper: executive steering, program controls, design coordination, interface management, risk review, commercial review, procurement review, possession planning, safety review, commissioning readiness, stakeholder review, utilities coordination, and contractor progress meetings.
Some are needed. The question is whether they produce decisions and remove constraints.
Meeting overload has recognizable signs. The same issue appears in several forums. Actions roll forward without owners. People attend “just in case.” Senior people join to protect their interests because the process does not protect them. Pre-meetings and side meetings appear to prepare for the formal meeting. The calendar becomes the control system.
Human factors research is relevant here. People have limited attention and working memory. High workload, interruptions, and unclear priorities increase error likelihood. In safety-critical settings, that matters. A fatigued construction superintendent or commissioning lead moving from meeting to meeting may miss weak signals in the field.
Interface Complexity Creates Effort That No Single Party Owns
Interfaces are where friction thrives. EPC contractor to client. Main contractor to subcontractor. Engineering to procurement. Construction to commissioning. Projects to operations. Marine contractor to offshore installation team. Utility provider to road authority. OEM to system integrator. Regulator to asset owner.
Interface management is a recognized discipline because unmanaged interfaces create cost and risk. The challenge is that many projects track interface registers without managing interface behavior.
A register may show open points, but it may not show the time spent chasing responses, clarifying responsibilities, resolving conflicting assumptions, or escalating decisions. Interface issues can sit between contract scopes because each party is commercially rational. No one wants to accept cost, delay, or liability without instruction.
Commercial design matters. Contract models, risk allocation, pain-gain mechanisms, change provisions, notice requirements, and payment terms shape behavior. If a project rewards defensive administration, it should not be surprised when teams produce defensive administration.
Change Processes Can Protect Value Or Create Rework
Change control is essential in capital projects. Scope growth, design change, latent conditions, regulatory changes, vendor changes, and operational requirements can move cost and schedule quickly.
AACE International’s total cost management principles and common industry practice emphasize disciplined change identification, evaluation, approval, and incorporation into forecasts. That discipline protects the owner and the contractor.
But change processes create friction when they confuse control with delay. If teams cannot distinguish a material compensation event from a minor field adjustment, the system clogs. If rejected changes reappear under different labels because the commercial route is unclear, trust falls. If approved changes do not flow into schedule, cost, procurement, and completions systems, the project creates a second layer of rework.
Good change control answers four questions fast:
What changed?
Who has authority to decide?
What is the impact on cost, time, risk, quality, safety, and operations?
How will the decision be reflected in the baseline and workface plan?
If the process cannot answer these without a long administrative chase, friction has entered the control system.
The Consequences Reach Beyond Cost And Schedule
Project friction is often discussed as inefficiency. That understates the risk.
Cost and schedule impacts are the most visible. Idle labor, extended preliminaries, additional management hours, vessel standby, delayed procurement, extended engineering, and claims support all cost money. Schedule slippage follows when decisions, information, access, materials, or approvals do not arrive when needed.
Productivity suffers because crews lose time to waiting, rework, travel, permits, unclear instructions, and missing constraints. Measures such as installed quantities per labor hour may show the symptom, but not the cause. Without a friction lens, leaders may blame contractor productivity while the project system prevents productive work.
Technical quality can decline too. Repeated comment cycles do not always improve design. They can create fatigue and version confusion. Late changes may be incorporated under pressure. Field teams may work from outdated documents. Commissioning teams may inherit inconsistent dossiers. Quality defects often have administrative roots.
Safety is affected in two ways. Directly, friction can create congestion, rushed work, unclear permits, weak handovers, and competing priorities. Indirectly, it drains attention from hazard recognition and field leadership. Guidance from regulators and bodies such as IOGP and the Energy Institute often stresses leadership, competence, risk control, and learning. Those require available attention, not just procedures.
Morale also suffers. Skilled people dislike waste. Engineers, planners, commercial managers, supervisors, commissioning leads, and project controls professionals can tolerate hard work when it matters. They become frustrated when effort goes into repetitive reporting, low-value approvals, or avoidable rework. Over time, people disengage or become cynical. The organization then loses discretionary effort, which is often what keeps difficult projects moving.
Contractor performance can deteriorate in predictable ways. If contractors experience slow decisions, unclear scope boundaries, inconsistent instructions, and heavy reporting demands, they protect themselves. They increase notices. They price risk. They assign stronger commercial resources. They become less willing to collaborate informally because informal collaboration may create unrecognized liability.
Operational readiness may be the final casualty. A project can focus so much effort on capital delivery controls that it underinvests in maintainability, spare parts, training, procedures, asset data, commissioning evidence, and operator acceptance. Handover becomes a documentation event rather than a readiness outcome.
The result is familiar: the asset is built, but operations is not ready to run it safely and efficiently.
Weak Signals Show When Friction Is Increasing
Friction rarely announces itself through a single KPI. It appears as weak signals across the system.
Common signs include:
Approval cycle times keep increasing, even for routine decisions.
The same issue appears in multiple reviews without resolution.
Teams maintain shadow trackers because official systems are not trusted.
Meeting attendance grows while decision quality falls.
Engineers spend more time responding to comments than closing design.
Construction supervisors escalate missing materials, access, permits, and drawings repeatedly.
Contractors submit more notices linked to late instructions or information.
Reports become longer while confidence in the data decreases.
Change logs grow, but approved changes are not reflected in the working plan.
Commissioning teams discover tag, system, and dossier gaps late.
Operations refuses handover because evidence is incomplete or hard to verify.
Senior leaders ask for “one more dashboard” because existing reports do not answer the question.
The project may still show acceptable headline progress. That is why friction matters. It often grows before conventional indicators show distress.
A useful test is to ask where experienced people spend their week. If senior delivery roles spend most of their time preparing reports, chasing approvals, reconciling systems, attending coordination forums, and restating decisions, the project has a capacity problem even if the cost report has not yet moved.
How To Identify And Quantify Friction Without Weakening Control
Friction can feel subjective, but it can be measured well enough to manage. The aim is not to remove all administration. The aim is to separate essential control from avoidable load.
Map The Real Workflow, Not The Procedure
Start with a live process that matters. Examples include technical query approval, document review, field change approval, work pack release, vendor data review, permit readiness, mechanical completion handover, or payment certification.
Map what actually happens:
Who starts the workflow?
How many systems are touched?
How many handoffs occur?
How many approvals are required?
How many times does work return for correction?
Where does the item wait?
What information is missing at each step?
Who can make the final decision?
This reveals queues, loops, duplicate checks, and unclear authority. It also shows whether the written procedure matches the project reality.
Measure Cycle Time And Waiting Time
Approval cycle time is one of the clearest friction indicators. Track median and outlier durations for RFIs, technical queries, design reviews, change approvals, procurement approvals, permits, completions sign-offs, and handover certificates.
Waiting time matters as much as processing time. A technical query may take two hours to answer but sit in queues for ten days. That delay consumes workface capacity and may create knock-on rework.
Track Rework By Cause, Not Only By Cost
Rework data is often underused. Projects may record nonconformance reports, design revisions, punch items, and field rework, but not connect them to root causes.
Useful categories include late design change, unclear scope, wrong revision, vendor data error, constructability issue, missed interface, client comment, operations change, material substitution, incomplete survey, or permitting constraint.
This shifts the discussion from blame to system design. If the same causes repeat, friction is structural.
Count Meeting Load And Reporting Effort
A simple meeting and reporting audit can be revealing. For two or three weeks, ask key roles to record time spent in recurring meetings, report preparation, data reconciliation, approvals, and administrative follow-up.
Do not turn this into a surveillance exercise. The goal is to understand capacity allocation. If delivery-critical roles spend a large share of time in management process rather than delivery problem-solving, leaders have evidence to act.
Compare Control Effort With Risk Reduction
Some high-effort controls are justified. A process safety review for a hazardous facility, a marine warranty approval for a heavy lift, a safety case requirement, a regulatory hold point, or an independent design check for a critical structure may consume time because the consequence of error is severe.
Other controls may not justify their load. A low-value procurement approval that passes through many senior people, a duplicated progress pack that no one reads, or a broad review meeting that produces no decisions may consume capacity without reducing risk.
Use a simple test:
Control activity | Risk it reduces | Effort consumed | Evidence it works |
Stage gate review | Poor investment decision or immature scope | Preparation, review time, actions | Clear decision, closed gaps, defined conditions |
Technical assurance | Design error or unsafe operation | Discipline reviews, comment resolution | Risk-based comments, fewer late defects |
Weekly progress report | Loss of visibility and weak forecasting | Data extraction, narrative, review | Better decisions, faster support |
Multi-level approval | Unauthorized spend or risk acceptance | Queue time, management attention | Faster, better decisions than delegated authority |
The final column matters. If no one can explain how a control improves outcomes, it may be bureaucracy wearing the clothes of governance.
Leaders Can Simplify Delivery While Protecting The Controls That Matter
Simplification does not mean weakening assurance. In high-risk industries, that would be reckless. The goal is better control with less waste.
Start by defining non-negotiables. Regulatory requirements, life safety, process safety, environmental controls, statutory approvals, critical technical assurance, financial authority, and material commercial decisions need clear protection. Make these controls visible and respected.
Then remove noise around them.
A practical approach is to segment decisions by risk and value. Low-risk, low-value decisions should sit close to the workface with clear boundaries. High-risk or high-value decisions deserve more governance, but with named decision owners and time limits. This reduces congestion and protects senior attention for decisions that merit it.
Next, rationalize reporting. Build one source of truth for core project data where possible. Align reporting calendars. Remove duplicate packs. Replace long narrative reports with exception-based reporting where the audience is mature enough to use it. Ask each report owner who uses the report, what decision it supports, and what would happen if it stopped.
Digital simplification requires discipline. Fewer systems can help, but only if data ownership is clear. Define master data for tags, documents, cost codes, work packs, completions status, change records, and asset information. Assign owners. Control changes. Reconcile early, not during handover.
Contractor interfaces need deliberate design. Set decision rights, escalation paths, information standards, review periods, and change rules early. Avoid pushing every uncertainty into contract administration. Where collaboration is needed, create forums that can decide, not just discuss. Where commercial protection is needed, make the route clear so parties do not create defensive workarounds.
For brownfield work, shutdowns, utilities, offshore campaigns, and live infrastructure, integrate operations early. Operations should not appear late as a reviewer of finished work. They should help define isolations, access, operating constraints, maintainability, start-up sequence, spare parts, competency needs, and handover evidence from the start.
Commissioning and completions should also shape earlier phases. A systems completion view can prevent tag chaos, dossier gaps, and late readiness disputes. Mechanical completion, pre-commissioning, commissioning, performance testing, training, spares, procedures, and asset data need to connect to the execution plan, not sit as end-stage administration.
Senior leaders play a specific role. They must stop equating more process with more control. More reports can reduce uncertainty, but they can also create the appearance of certainty. More approvals can reduce risk, but they can also delay the people best placed to manage it. More meetings can connect teams, but they can also hide weak accountability.
The better question is whether the project’s management system helps competent people make timely, risk-aware decisions with reliable information.
Project directors, sponsors, and senior leaders should ask sharper questions:
Which controls would we keep if we had to reduce management effort by 20 percent?
Which reports directly support decisions, and which exist because confidence is low?
Where are approval queues delaying workface progress?
Which decisions are escalated that should be delegated?
Where do contractors maintain shadow systems because they do not trust ours?
Which meetings make decisions, and which only circulate information?
What are the top causes of rework by volume and by consequence?
Where do operations, commissioning, and maintenance first see the asset information they must inherit?
Which governance requirements conflict across the client, contractor, joint venture, lender, or regulator?
What administrative work would the frontline stop tomorrow if leadership allowed it?
Where are we asking for assurance after the cost of change has already become high?
Which weak signals suggest people are managing the process instead of the project?
These questions can be uncomfortable. That is their value.
The Professional Takeaway
Project friction is not a soft issue. It is a delivery capacity issue with commercial, technical, safety, and operational consequences.
High-risk projects need governance, assurance, reporting, approvals, meetings, and disciplined change control. They also need leaders who can tell the difference between control that reduces risk and bureaucracy that consumes capacity.
The best project organizations do not remove control. They design it. They measure cycle times, rework, reporting effort, meeting load, interface delays, system reconciliation, and administrative workload with the same seriousness they apply to cost and schedule.
When leaders actively measure and manage friction, they recover scarce capacity for the work that matters: better decisions, safer execution, stronger contractor performance, cleaner handover, and assets that are truly ready to operate.
