IT & Consulting

Operations Cost Model Analysis: How to Build a Cost Model for Your Operations

By ITCM Team · Updated October 2026

An operations cost model analysis traces every cost in a business or department (people, IT, space, suppliers) to the activities that consume it, so you can see the true cost of each process, order, customer or service and how much capacity you are paying for but not using. It is the method an operations management consultant uses before recommending any cut, because it shows which savings are real and which only move cost around. This guide walks through the method step by step, with a template of cost categories, a fully worked (hypothetical) example, the KPIs to track and the mistakes that make cost models misleading.

What an operations cost model is (and how it differs from a budget)

A budget tells you how much each department is allowed to spend. A P&L tells you what was spent by account: salaries, rent, software, subcontractors. Neither tells you what it costs to process one invoice, onboard one customer, deliver one project phase or resolve one support ticket. An operations cost model fills that gap by re-arranging the same spending around activities and outputs.

The result is a cost per unit for each process, a view of which customers, products or services consume the most effort, and a clear figure for capacity you are paying for but not using. Those three outputs are what make cost reduction decisions safe. Without them, managers tend to cut the most visible lines, such as training or tools, while the real cost sits in rework, exceptions and manual handling.

Cost model analysis is not the same as a general cost-cutting exercise. If you want a quick list of common savings, read our guide on how to reduce IT and operational costs. This page goes deeper into the modelling method itself: how the numbers are built, tested and used.

The building blocks: how costs behave

Before allocating anything, classify each cost by how it behaves when activity goes up or down. This decides what can realistically be saved and over what time frame. A cost that is fixed this year may be variable over three years, so record the contract or notice period alongside the behaviour.

Choosing a costing method: traditional, ABC or time-driven ABC

Traditional costing spreads overhead using one broad base such as headcount, revenue or floor space. It is simple but it hides the fact that some processes, customers or products need far more support than others. Activity-based costing (ABC) assigns overhead to activities first, then to outputs, using a cost driver for each activity. It is more accurate but classic ABC relies on staff estimating what percentage of their time goes to each activity, which is slow to collect and tends to add up to 100 percent even when people are idle.

Time-driven activity-based costing (TDABC), described by Robert Kaplan and Steven Anderson in Harvard Business Review in 2004, simplifies this. You estimate only two things: the cost per minute of a resource group at practical capacity, and the minutes each activity takes. Multiply by volumes and you get both the cost per unit and the unused capacity. For most service, back-office and project-based operations, TDABC is the method we recommend because the inputs are easy to observe and update.

Step-by-step: how to build an operations cost model

The method below works for a whole company or a single department. Start with one department or one end-to-end process; a narrow, accurate model is far more useful than a broad, guessed one.

• Define the question and scope: which department or process, which period (usually the last 12 months or a recent quarter), and which decision the model must support.
• Collect the cost base from the general ledger: salaries and benefits, contractors, software and licences, hosting, devices, premises, suppliers. Reconcile the total to the P&L so nothing is missing or double counted.
• Group costs into resource pools: each team or shared function becomes a pool with its full monthly cost, including its share of IT, space and management.
• Calculate practical capacity for each pool: paid hours minus leave, training, meetings and breaks. Many models use around 80 to 85 percent of paid time as a starting assumption, then adjust from observation.
• Work out the capacity cost rate: pool cost divided by practical capacity, giving a cost per hour or per minute.
• Map activities and time per activity: interview the people doing the work, observe a sample, and use system timestamps from your ERP, CRM or ticketing tool where they exist. Record variations, such as exceptions or rework, as separate activities.
• Pull volumes for each activity from the systems for the same period and multiply: volume times minutes times cost rate gives the cost of each activity and the cost per unit.
• Reconcile and validate: compare minutes used with capacity available, explain any gap, and check results with the team leads before anyone acts on them.

Worked example: the cost of processing a supplier invoice

The numbers below are purely hypothetical, chosen to show the arithmetic. Imagine an accounts payable team of three people. Their fully loaded monthly cost, including salaries, benefits, their share of the ERP licence, devices and office space, is $14,400. After leave, meetings and breaks, their practical capacity is 24,000 minutes a month, so the capacity cost rate is $14,400 divided by 24,000, or $0.60 per minute.

The team performs four activities. Time estimates come from observation and ERP timestamps; volumes come from the ERP for the month.

Reading the results: utilisation, unused capacity and real savings

In the example, the blended cost per invoice is about $6.88 (the $13,080 of used capacity divided by 1,900 invoices), but that average hides the real story. An invoice without a purchase order costs two and a half times as much as a matched one, and 120 exceptions consume 4,800 minutes, more than half the time spent on all 1,500 matched invoices. The highest-value fixes are therefore upstream: enforcing purchase orders and cleaning supplier and price master data, not asking the team to work faster.

The model also shows 2,200 minutes of unused capacity costing $1,320 a month. If 300 of the non-PO invoices moved to purchase orders, the team would free another 2,700 minutes. That is capacity released, not cash saved. Cash is only saved when the freed capacity is removed (a vacancy not refilled, a contractor released) or redeployed to work you would otherwise pay for. Good cost model analysis always states which of the two applies, and over what time frame.

Utilisation should be read with care. A team at 100 percent of practical capacity has no room for peaks and is likely building backlogs and errors. A useful target range depends on how variable demand is, which is why the model should be run on several months, not one.

Template: cost categories to include in your model

Use this as a checklist when collecting the cost base. Every row should be either assigned directly to a process or given a driver. Anything left unassigned should appear as a visible overhead line, never quietly spread across everything.

Where IT and process costs hide (cost-effective IT management)

Cost-effective IT management is less about the price of each tool and more about how much work each tool removes or creates. In most cost models, IT appears in two places: as a direct cost line, and as hidden time inside every operational activity. The second is usually larger and is invisible in the IT budget.

When the model maps activities, flag every minute spent on work that a system should already do. Those minutes are the real business case for automation, integration or a better-configured ERP, and they can be priced with the same capacity cost rate.

• Per-seat licences for leavers, inactive users or people who only need read access.
• Overlapping tools: two project tools, two file-sharing services, a CRM used alongside spreadsheets.
• Manual re-keying between systems that are not integrated, such as a website, CRM and ERP.
• Exports to spreadsheets for reports the system could produce directly.
• Rework caused by poor master data: duplicate suppliers, wrong prices, missing customer details.
• Downtime and slow systems, measured as minutes lost multiplied by the cost rate of the people affected.
• Customisations that must be retested and maintained at every upgrade.

KPIs to track once the model is built

A cost model is only valuable if it changes behaviour. Pick a small set of KPIs that come straight from the model and review them monthly with the process owners. Keep definitions fixed so trends are comparable; see our guide on IT performance management for how to set up regular reviews.

Common mistakes that make cost models misleading

• Allocating overhead by headcount or revenue and calling it a cost model.
• Using 100 percent of paid hours as capacity, which hides idle time and inflates every unit cost.
• Relying only on staff estimates of time with no observation or system data to check them.
• Modelling a single unusual month instead of a representative period.
• Treating released capacity as cash savings without a plan to remove or redeploy it.
• Ignoring exceptions and rework by modelling only the standard, happy-path process.
• Not reconciling the model total back to the ledger, so costs are missed or counted twice.
• Building the model once in a spreadsheet nobody owns, so it is out of date within a quarter.

When to hire an operations cost consultant, and how an engagement works

You can build a first model internally if you have clean ledger data, reliable system volumes and someone with time to observe the work. Bringing in an operations management consultant makes sense when costs are rising without a clear cause, when departments disagree about who drives cost, when you are preparing to scale, restructure or split a company, or when you need an independent view that staff will trust. IT operations management consulting is particularly useful when much of the cost sits inside systems and workarounds that finance cannot see on its own.

A typical engagement follows the steps above: agree the question and scope, collect ledger, payroll and system data, interview and observe the teams, build and validate the model, then deliver prioritised recommendations with the expected effect of each, whether it releases capacity or saves cash. Our operations and cost analysis service works this way, remotely, for a whole company or a single department, and can go on to implement automation or tooling where the model shows it pays. For an example of the outputs, see the operations and cost audit case study.

Every ITCM service starts from $100, and larger audits are quoted per project and delivered in stages. The price is agreed first and there is no advance payment: you pay after you have seen and reviewed the work. A free initial consultation is available to define the scope.

Frequently asked questions

What is an operations cost model?

It is a model that assigns all the costs of a business or department to the activities and outputs that consume them, so you can see the cost per unit of each process, which customers or services use the most resources, and how much capacity is unused.

What is the difference between activity-based costing and time-driven ABC?

Classic activity-based costing assigns overhead using staff estimates of the percentage of time spent on each activity. Time-driven ABC instead uses a cost per minute of practical capacity and the minutes each activity takes, which is quicker to build, easier to update and shows unused capacity explicitly.

How long does an operations cost analysis take?

It depends on scope and data quality. A single process or department with good system data can often be modelled in a few weeks, while a whole-company model with several departments takes longer and is usually delivered in stages.

What data do I need to build a cost model?

You need the general ledger for the period, payroll or team cost data, software and supplier contracts, activity volumes from your ERP, CRM or ticketing systems, and time estimates from observation or system timestamps.

Does freeing up capacity actually reduce costs?

Not by itself. Freed capacity becomes a cash saving only when it is removed, for example by not refilling a role or releasing a contractor, or when it is redeployed to work you would otherwise pay for.

How much does an operations cost consultant cost?

Fees vary widely with scope, region and depth. At ITCM, services start from $100, larger audits are quoted per project and delivered in stages, and you pay after reviewing the work, with no advance payment.

Related reading & services

• Operations and cost analysis service
• How to reduce IT and operational costs
• Case study: operations and cost audit
• Cost-effective IT solutions
• ERP services
• Custom operations software

← All articles