Operations Consulting: Process Improvement, Efficiency and Operational Excellence

Operations consulting helps organizations improve how work actually moves from demand to delivery.

A company can have a sound strategy, capable employees and growing demand yet still struggle because orders take too long, teams repeatedly enter the same information, approval queues slow decisions, quality varies, or each increase in volume requires a similar increase in headcount.

Those are operational problems.

Operations consulting focuses on understanding why those problems occur and redesigning the systems, processes, roles and management routines behind them.

The first step should not be choosing Lean, buying automation software or creating another dashboard. It should be finding the constraint that is preventing the operation from delivering the required speed, cost, quality, capacity or reliability.

That distinction matters because improving the wrong part of a process can make one department look more efficient without improving the end-to-end result.

This guide explains what operations consulting involves, how operational problems are diagnosed, where process improvement and operating-model design fit, how automation and AI should be used, and how organizations can measure whether an operational change actually worked.

Operations Consulting at a Glance

Question	Practical answer
What is operations consulting?	Advisory and implementation work focused on improving how an organization produces and delivers products or services.
What problems does it solve?	Bottlenecks, high costs, long cycle times, capacity constraints, quality issues, unreliable handoffs and weak operating routines.
What should an operations consultant do first?	Establish the current baseline and identify the main operational constraint before recommending solutions.
Is it only for manufacturing?	No. Operations consulting applies to service businesses, healthcare, technology, financial services, logistics and other process-driven organizations.
Does operations consulting include automation?	It can, but automation should normally follow process diagnosis and simplification.
What is operational excellence?	The ability to improve performance continuously through effective processes, management systems and employee capability.
How should results be measured?	With problem-specific metrics such as cycle time, throughput, cost, quality, service level, backlog and reliability.

What Is Operations Consulting?

Operations consulting is a branch of management consulting focused on improving the systems through which an organization performs and delivers work.

Depending on the business and the problem, an operations consultant may examine:

• process flow;
• capacity;
• operating costs;
• supply chains;
• procurement;
• service delivery;
• quality;
• inventory;
• workforce deployment;
• decision rights;
• performance management;
• technology;
• automation;
• operational risk.

The scope can range from one problematic workflow to a broader operating-model redesign.

The important distinction is that operations consulting focuses primarily on how the organization executes.

A strategy consultant might help a company decide to enter a new market.

An operations consultant might then determine whether fulfillment, staffing, systems and capacity can support that expansion.

McKinsey currently describes operations work as connecting strategic priorities with frontline execution, while BCG and Bain frame the discipline across areas such as operational excellence, service operations, supply chains and procurement.

1. Diagnose the Operational Constraint Before Redesigning the Process

Operational problems often appear in one place while originating somewhere else.

A growing backlog may look like a staffing shortage.

The real problem could be an approval queue.

Rising costs may look like a supplier issue.

The real cause could be rework, poor scheduling or excessive process variation.

Before changing the process, separate the symptom from the constraint.

What you observe	What may actually be causing it
Orders ship late	Capacity, scheduling, inventory, approvals or handoffs
Employees work excessive overtime	Workload imbalance, rework, poor flow or genuine capacity shortage
Customer complaints increase	Quality variation, handoffs or expectation mismatch
Cost rises quickly as volume increases	Poor scalability or excessive manual work
Work waits between departments	Queue design or unclear ownership
Managers constantly intervene	Weak decision rights or unstable processes
Errors repeatedly return	Process variation, unclear standards or system design
New software fails to improve performance	The original workflow was never simplified

A useful diagnosis starts with five questions:

Where does work wait?

Where does work repeat?

Where do defects or errors enter?

Which resource limits the total output?

Which decisions require unnecessary escalation?

The goal is not to identify every imperfection.

It is to determine which problem has the greatest effect on end-to-end performance.

Establish a Baseline First

Improvement becomes difficult to prove when nobody knows how the process performed before it changed.

Depending on the issue, the baseline might include:

• average cycle time;
• order lead time;
• backlog;
• cost per transaction;
• defect rate;
• rework;
• service level;
• throughput;
• downtime;
• customer waiting time.

Measure before redesigning.

Otherwise, the organization may implement a large change and still be unable to determine whether it made the operation better.

2. Map the Process to Understand Flow and Handoffs

Process mapping is useful because operational problems often hide between teams rather than inside one individual task.

A practical process map should show:

• what triggers the process;
• major activities;
• who owns each step;
• handoffs;
• decision points;
• queues;
• rework;
• systems used;
• output.

The goal is not to create an impressive diagram.

It is to make the flow visible enough that people can see where work stops, repeats or becomes unnecessarily complicated.

For each step, ask:

Does this step need to exist?

Does it create customer or operational value?

Is it required for safety, risk or compliance?

Could it be eliminated, combined or simplified?

A process map is therefore a diagnostic tool.

It is not the improvement itself.

Pay Particular Attention to Handoffs

Many operating problems appear when responsibility moves from one person or department to another.

Typical questions include:

• What information must be transferred?
• Who confirms the handoff?
• What happens when information is incomplete?
• Does the receiving team understand the priority?
• Does the work enter a queue?
• Who owns exceptions?

A process can have individually efficient departments and still perform poorly if the handoffs between them are weak.

3. Use Lean Principles Where Waste and Flow Are the Problem

Lean methods can help when the operation suffers from unnecessary movement, waiting, excessive inventory, defects, overprocessing or inconsistent flow.

Useful tools can include:

• value-stream mapping;
• root-cause analysis;
• standard work;
• visual management;
• continuous-improvement routines.

But “apply Lean” should not become the diagnosis.

A company with weak customer demand does not solve that problem by reducing process waste.

A company with the wrong strategy does not become strategically stronger because its internal workflows become faster.

Operational frameworks should fit the problem rather than become the objective.

McKinsey’s operational-excellence work treats continuous improvement as an organizational capability rather than a temporary cost-reduction program.

4. Find the Bottleneck Before Adding More Capacity

One of the easiest responses to operational strain is adding more people, equipment or technology.

That can be expensive when the real constraint has not been identified.

Consider a service process with five stages.

Four teams have spare capacity, but one specialist review stage is permanently overloaded.

Adding employees to the other four stages will not meaningfully improve total throughput.

It may simply increase the queue waiting for the bottleneck.

Look for signals such as:

Signal	Question to investigate
Growing backlog	Where does work accumulate?
Long cycle time	Which stage creates most of the delay?
High overtime	Is workload genuinely above capacity or distributed poorly?
Idle resources upstream	Is another step restricting flow?
Quality drops as volume rises	Which control or capability stops scaling?
Constant executive escalation	Has a decision become the real bottleneck?

Improving a non-constraining step may improve a departmental KPI without improving the system.

Operations should therefore be evaluated end to end, not only by local productivity.

5. Fix the Operating Model When Process Changes Are Not Enough

Some operational problems keep returning even after the workflow has been redesigned.

That often means the process sits inside a weak operating model.

An operating model describes how the organization organizes work, authority, resources and management routines to execute its strategy.

An operations consultant may need to examine:

• process ownership;
• roles and responsibilities;
• decision rights;
• organizational structure;
• management routines;
• performance metrics;
• technology dependencies;
• cross-functional governance.

For example, a company might redesign customer onboarding from twelve steps to seven.

But if every exception still requires approval from three senior managers, the operation may remain slow.

In that situation, process design is only part of the problem.

The organization also needs to reconsider where authority sits.

Decision Rights Matter

Operational delays are often treated as workflow problems when they are actually decision problems.

Clarify:

Who owns the decision?

What can that person decide independently?

When is consultation required?

What genuinely needs escalation?

Clear decision rights can reduce unnecessary management involvement without removing appropriate controls.

6. Operations Consulting Also Applies to Service Businesses

Operations consulting is sometimes associated mainly with factories and supply chains.

The same logic applies to service businesses.

A service organization still has:

• demand;
• capacity;
• queues;
• handoffs;
• quality;
• variation;
• constraints.

The difference is that inventory may appear as waiting customers, unresolved cases or unfinished work rather than physical goods.

Examples include:

Customer Support

How long do customers wait?

Where do cases escalate?

Which issues create repeat contacts?

Healthcare

Where do patients wait?

How are appointments scheduled?

Where do clinical and administrative handoffs fail?

Professional Services

Which work requires specialist expertise?

Where does project work become delayed?

How much capacity is tied up in low-value administration?

Financial Services

Where do applications or claims queue?

Which controls are necessary?

Which reviews duplicate one another?

SaaS and Technology

Where does customer onboarding stall?

Which support cases consume disproportionate resources?

Where do engineering, product and customer teams struggle to coordinate?

BCG and Bain both maintain dedicated service-operations capabilities, reinforcing that operational excellence extends well beyond manufacturing.

7. Simplify Before Automating

Automation is valuable when it removes repetitive work from a process that already makes sense.

It is much less valuable when it preserves unnecessary complexity.

Use this sequence:

Eliminate → Simplify → Standardize → Automate

Before automating, ask:

1. Does the activity need to exist?
2. Can steps be combined?
3. Are exception cases understood?
4. Is the workflow stable enough to automate?
5. Which systems and data are involved?
6. What happens when the automation fails?
7. Who owns the process afterward?

Consider an approval workflow involving six manual steps.

Automating all six may make the workflow faster.

Removing three unnecessary approvals before automation may create a much larger improvement.

Automating an unnecessary process simply allows unnecessary work to happen faster.

If the issue becomes primarily one of system architecture, APIs, cloud platforms or technical integration, the work should move into dedicated IT consulting rather than stretching the operations engagement beyond its expertise.

8. Standardize What Needs Consistency—Not Everything

Standard operating procedures can make repeatable work easier to train, measure and scale.

They are especially useful when variation creates:

• quality problems;
• safety risk;
• compliance issues;
• avoidable errors;
• inconsistent customer outcomes.

But excessive standardization creates another problem.

Employees may be forced to follow detailed procedures even when the task requires professional judgment.

A useful rule is:

Standardize the parts of the work where consistency matters. Preserve judgment where expertise matters.

A customer-support team may need standard verification requirements while still allowing experienced employees discretion in resolving unusual cases.

The purpose of an SOP is reliable execution not removing all human judgment from the process.

9. Build Scalability Into the Operating System

Scalability does not mean that every cost remains fixed while revenue increases.

A scalable operation understands which resources become constraints as demand grows and can increase capacity without unacceptable deterioration in:

• cost;
• quality;
• delivery;
• customer experience.

Before scaling, evaluate:

• process capacity;
• bottleneck stages;
• staffing model;
• training time;
• system limits;
• management workload;
• supplier capacity;
• inventory;
• quality controls.

A process that works at 100 transactions per month may behave very differently at 1,000.

That is why business growth strategy should be tested against operating capacity before a company commits heavily to expansion.

Growth is not operationally sustainable if each new customer creates disproportionate cost, delay or management effort.

10. Use Operational KPIs That Help Someone Make a Decision

Dashboards are useful only when they help managers act.

A dashboard filled with dozens of metrics can create visibility without clarity.

Start with the operating question.

Operational question	Possible KPI
How fast does work move?	Cycle time / lead time
How much can the system deliver?	Throughput
How reliably are commitments met?	Service level / on-time delivery
How often does work fail?	Defect or error rate
How much work must be repeated?	Rework
How much capacity is being used?	Utilization
What does delivery cost?	Cost per unit / transaction
Where is work accumulating?	Backlog / work in progress
How stable is the process?	Variation / exception rate
How often does the operation stop?	Downtime / availability

Then ask:

What decision changes if this metric changes?

If nobody can answer, the KPI may not deserve management attention.

Do Not Optimize One Metric in Isolation

Operational measures interact.

Maximizing utilization, for example, may look efficient.

But a system operating permanently at maximum capacity can create longer queues and slower response when demand varies.

Similarly:

reducing inventory can weaken resilience;

reducing staffing can increase customer waiting time;

increasing speed can damage quality.

Operational excellence requires managing trade-offs, not maximizing every KPI independently.

11. Build Daily Management Around the Work

A redesigned process will deteriorate if nobody notices when performance begins to drift.

For each important operational priority, define:

Owner: Who is accountable?

Metric: What indicates performance?

Cadence: How often is it reviewed?

Threshold: When does intervention become necessary?

Escalation: Which issues need higher-level decisions?

The purpose of daily or weekly operational management is not creating more meetings.

It is helping teams identify deviations early enough to correct them before they become recurring problems.

A good operating rhythm should make the normal process easier to manage without requiring senior executives to resolve every exception.

12. Build Operational Resilience Around Critical Dependencies

Efficiency and resilience sometimes pull in opposite directions.

Reducing all spare capacity may lower short-term cost while leaving the operation unable to absorb unexpected demand.

Depending entirely on one supplier may simplify procurement while creating a serious point of failure.

An operational-risk review can examine:

• critical suppliers;
• specialist employees;
• important systems;
• inventory dependencies;
• recovery procedures;
• capacity buffers;
• alternative workflows;
• cross-functional dependencies.

The objective is not eliminating every risk.

That would often be prohibitively expensive.

The better question is:

Which failure would hurt the operation most, and what level of protection is justified?

Operational excellence increasingly includes resilience alongside productivity and efficiency, particularly when organizations face uncertain demand, supply disruptions and technology dependencies.

13. Use AI Against a Defined Operational Problem

AI is becoming part of operational improvement in areas such as:

• forecasting;
• scheduling;
• service workflows;
• knowledge retrieval;
• decision support;
• anomaly detection.

But “add AI” is not an operations strategy.

Before applying AI, define:

Which decision improves?

Which task becomes faster?

What data does the system require?

How will incorrect outputs be detected?

Which exceptions still need human judgment?

Which operating metric should change?

McKinsey’s June 2026 operational-excellence research argues that companies get more value from AI when it is embedded within strong management systems and continuous-improvement practices rather than treated as a standalone technology initiative.

The operational question should therefore come first.

The technology follows.

Operations Consulting vs Strategy Consulting

The two disciplines are related but should not be treated as the same service.

Strategy consulting	Operations consulting
Focuses on major strategic choices	Focuses on how work is executed
Evaluates markets, business models and competitive direction	Evaluates flow, capacity, cost, quality and reliability
Often asks “What should we do?”	Often asks “Why is execution underperforming?”
Defines strategic priorities	Helps build the operating system needed to deliver them

For example:

Choosing to expand into a new geographic market is primarily a strategy consulting question.

Determining how fulfillment, staffing and systems must change to serve that market is an operations consulting question.

The boundary can overlap, but maintaining the distinction prevents one consulting engagement from becoming unnecessarily broad.

Operations Consulting vs Business Transformation

Operations consulting is normally more focused.

A company does not need an enterprise transformation because one workflow has a bottleneck.

Operations consulting may be sufficient when the main problem involves:

• processes;
• capacity;
• supply chain;
• cost;
• service delivery;
• operating routines.

Business transformation becomes more appropriate when major changes must be coordinated across several areas at once, such as:

• strategy;
• operations;
• technology;
• organizational structure;
• workforce capability.

The size of the project should follow the size of the problem.

When Should a Business Hire an Operations Consultant?

Outside operations expertise can be useful when:

• backlogs keep increasing;
• cycle times are too long;
• growth is creating delivery problems;
• costs rise disproportionately with volume;
• quality varies repeatedly;
• supply-chain or service reliability is weak;
• departments blame one another for handoff problems;
• managers spend too much time resolving routine exceptions;
• operating capacity is unclear;
• a new operating model is required;
• automation is being considered before the process is understood.

Another useful signal is recurrence.

If the organization has “fixed” the same operational problem several times and it keeps returning, the underlying system may need deeper diagnosis.

When Operations Consulting Is Not the Right Solution

Operations consulting is not automatically the best intervention for every performance problem.

It may add limited value when:

• the strategic direction itself is unclear;
• the real problem is insufficient customer demand;
• the issue is primarily marketing positioning;
• the company mainly needs a software implementation;
• a permanent operations leader is required;
• leadership already knows the solution and simply lacks execution capacity;
• several major business functions need coordinated transformation.

The correct conclusion may therefore be that another specialist is more appropriate.

That is a better outcome than forcing an operational solution onto a non-operational problem.

What Should an Operations Consulting Engagement Deliver?

The exact deliverables depend on the problem.

A typical engagement may include:

Deliverable	Purpose
Operational baseline	Documents current performance
Process map	Shows work, handoffs and queues
Bottleneck/root-cause analysis	Identifies the main constraint
Target-state process	Defines how work should operate
Capacity analysis	Tests resource and throughput requirements
Operating-model recommendations	Clarifies roles and decision rights
KPI framework	Defines how performance will be monitored
Implementation roadmap	Sequences improvements
SOPs where appropriate	Stabilizes repeatable work
Technology requirements	Defines what systems need to support
Risk assessment	Identifies critical dependencies
Handover plan	Transfers ownership internally

Not every engagement should produce every item.

A consultant should not create unnecessary deliverables simply because they are part of a standard template.

How Should Operations Consulting Results Be Measured?

Measure the result against the problem that justified the engagement.

If the problem was long customer waiting time, measure waiting time.

If the problem was excessive rework, measure rework.

If the problem was limited capacity, measure sustainable throughput.

Relevant operational measures can include:

• cycle time;
• throughput;
• cost per transaction;
• defect rate;
• rework;
• backlog;
• service level;
• customer waiting time;
• availability;
• downtime;
• capacity.

Establish the baseline before implementation.

Then separate three things:

Change implemented
Did the organization actually change the process?

Performance improved
Did the relevant metric improve afterward?

Attribution supported
Is there reasonable evidence that the intervention contributed to the change?

A redesigned process is not automatically evidence of better operations.

The outcome still has to be measured.

Build Internal Capability Before the Engagement Ends

The strongest operational improvement is one the organization can continue without permanent dependence on the consultant.

Before handover, clarify:

• process ownership;
• documentation;
• KPI responsibility;
• training;
• escalation rules;
• improvement routines;
• change-control responsibility.

This does not mean the operation will never need outside expertise again.

It means the internal team understands how the process works, how performance is monitored and how future problems should be diagnosed.

Operational excellence becomes more durable when improvement is treated as an internal capability rather than a one-time consulting project.

Conclusion

Operations consulting is most useful when an organization knows what it needs to deliver but the operating system is making execution slow, expensive, unreliable or difficult to scale.

The work should begin with the flow of work:

Where does it wait?

Where does it fail?

Which resource constrains it?

Which decisions create unnecessary friction?

From there, the answer may involve process redesign, operating-model changes, capacity planning, Lean methods, standardization, technology, automation or stronger management routines.

The goal is not maximum efficiency in every department.

It is a better end-to-end operating system one capable of delivering the required quality, speed, cost and resilience without unnecessary complexity.

Sometimes the answer is automation.

Sometimes it is another employee.

Sometimes it is clearer authority.

Sometimes it is fewer approvals.

And sometimes the strongest operational improvement is simply removing work that never needed to exist.

Frequently Asked Questions