D2. Cost and FinOps for AI

Cost and FinOps for AI: eight cost fields, token and inference costs, unit economics, cost levers, and ongoing cost steering.

AI incurs costs in many places at once, from data preparation through training and operations to personnel. What sets it apart from classic software is inference, the productive model operation: every single request costs compute, so ongoing costs grow with usage. A FinOps approach applies proven cloud cost-management principles to AI: make spending visible, assign ownership, continuously reconcile cost against business benefit.

What this chapter delivers: ongoing cost control for the budget frame from chapter B1. First it introduces the eight cost fields in which AI costs arise. Two templates build on them: the cost model per use case and the unit-economics worksheet for the profitability of a single request. After that come the cost levers for reducing costs and the FinOps practice that keeps costs under control for good.

The eight cost fields

Every AI use case incurs costs in the same eight fields. Knowing all eight prevents hidden cost blocks from being overlooked and makes use cases properly comparable:

Cost fieldWhat belongs to itWhat the amount depends on
Data acquisition and preparationsourcing, cleaning, labeling, and legally clearing datadata volume, required quality
Development / trainingexperiments, model training, fine-tuning, and evaluation up to a deployable solutionnumber of experiments, GPU-hours
Inference / APIproductive operation: compute or API fees for every single requestrequests × tokens
Infrastructure / storagecompute capacity, data storage, network, and environments for development and operationsdata volume, required availability
Licenses / platformsoftware and model licenses, platform and tool subscriptionsnumber of users, booked modules
Operations / monitoringdeployment, monitoring, troubleshooting, and regular retrainingsolution complexity, committed service levels (SLA)
Security / complianceprotective measures, audits, and documentation dutiesprotection needs, number of audits
Personnelinternal roles and external support across all phasesrequired roles, effort

One field stands out: inference is the only field that grows directly with usage, because every request costs anew. A successful use case therefore becomes more expensive, not cheaper. This is exactly why a one-off budget estimate is not enough for AI.

The eight cost fields per use case with their cost drivers, inference grows with usage

Template: cost model per use case

For every use case, all eight cost fields are estimated, separated into one-time and ongoing. The cost-driver column names what the amount depends on; if the scope changes, the estimate can be adjusted traceably. Important: an empty field means unexamined costs, not zero costs.

Cost fieldOne-timeOngoing per yearCost driver
Data acquisition and preparationVolume, quality
Development / trainingExperiments, GPU-hours
Inference / APIRequests × tokens
Infrastructure / storageData volume, availability
Licenses / platformUsers, modules
Operations / monitoringComplexity, SLA
Security / complianceProtection needs, audits
PersonnelRoles, effort
Total

Template: unit-economics worksheet

Total costs alone do not show whether a use case pays off. The worksheet therefore breaks the economics down to a single request; that makes visible whether the use case gets better or worse as usage grows. The same unit metrics later serve ongoing steering: cost per request, per training run, and per user; especially telling is cost per unit of result quality (e.g., per percentage point of accuracy).

MetricValueCalculation
Cost per requestongoing cost ÷ requests
Benefit per requesttime and cost saved or additional revenue
Contribution margin per requestbenefit minus cost
Break-even volumeone-time cost ÷ contribution margin
Scaling effectcost trend at 2, 5, and 10 times the volume

Worked example for the use case “customer-service reply suggestions” (continued from chapter B2): 200,000 requests per year, ongoing cost of 30,000 euros per year, one-time cost of 60,000 euros.

MetricValueCalculation
Cost per request0.15 euros30,000 euros ÷ 200,000 requests
Benefit per request1.00 eurosaround 1.5 minutes of handling time saved
Contribution margin per request0.85 euros1.00 euros minus 0.15 euros
Break-even volumearound 71,000 requests60,000 euros ÷ 0.85 euros; reached after a good four months at this volume
Scaling effectcontribution margin stays positiveas volume grows, the fixed-cost share per request falls and caching further lowers inference cost

The case therefore pays off from around 71,000 requests and becomes more economical as usage grows. If the contribution margin were negative, the reverse logic of the following warning sign would apply.

Warning sign: if the contribution margin per request is negative, the use case scales into a loss zone. Higher usage makes the problem worse instead of solving it.

Cost levers

If the worksheet comes out too expensive, the use case does not have to die. The following levers reduce costs without cutting the benefit; the right-hand column shows which cost field a lever mainly acts on. Cost optimization is an ongoing process, not a one-off project (terms such as RAG in chapter A2).

LeverHow it worksActs mainly on
Cachingreuse already computed answers instead of generating them againInference
Batchingbundle many requests and process them togetherInference
Prompt and context optimizationshorter inputs and a leaner context reduce the token volume per requestInference
RAGgrounded answers reduce costly wrong answers and reworkInference
More compact modelsquantization stores model weights at lower numerical precision, distillation lets a smaller model learn the behavior of a large one; both make models smaller and fasterInference, infrastructure
More efficient training methodsexperiment in a more targeted way, reuse already trained modelsDevelopment / training
Cheap compute windowsoff-peak training at times of low utilization; spot instances are short-notice, much cheaper cloud capacity that the provider reclaims when needed, suitable for interruptible training runsDevelopment / training
Specialized hardwareprocessors built for AI workloads compute more cheaply per resultDevelopment / training, inference
Cleaning upshut down unused resources, archive rarely used dataInfrastructure / storage

The FinOps practice: steering costs for the long run

A one-off estimate ages quickly because usage, prices, and models change continuously. The FinOps practice therefore anchors fixed habits in the organization:

PracticeHow it works
Cost transparencycapture all expenditure granularly by project, department, and technology; prevents hidden cloud or personnel costs from being underestimated
Cost allocationassign costs to the business units and products that cause them rather than blanket-charging IT. Showback shows each unit the costs it caused, chargeback actually bills them to it
Budgets and approvalsdefine who approves budgets, which thresholds apply, and from when approval is required; quotas per unit, unified standards for cost reports
Automated monitoringcapture costs in real time, trigger alerts on overruns, throttle automatically if necessary; especially important with usage-based cloud and API pricing
Forecasting and budget planningplan with scenarios (best, average, worst) and seasonal effects; makes investment plannable
Cost-benefit evaluationper use case, examine return on investment (ROI), total cost of ownership (the full cost over the lifetime, not just the purchase price), and break-even (the point from which the benefit covers the one-off costs); the basis for prioritizing between use cases
Cost awarenesstrain people that every training run and every query costs money; establish guidelines for cost-efficient development

The visible result of this practice is the monthly cost overview per use case with forecast and variance. It feeds the reporting from chapter D4.

Procurement, partners, and exit strategy

External dependencies are often underestimated. Clear criteria for selection and contracting (detailed vendor selection: chapter B3):

  • Data locations meet legal requirements
  • Customer data is not used to train third-party models
  • Interfaces are openly documented
  • Ownership of data and artifacts stays with the company
  • Price-adjustment clauses and audit evidence are contractually secured
  • Exit plan: export of data, features, and models regulated, replacement solution named, rollback organized

This keeps the risk of vendor lock-in (the situation where switching becomes practically unaffordable) manageable and secures long-term flexibility.

Checklist: FinOps maturity

The checklist follows the chapter’s path from estimating to lasting steering; open items are the next expansion steps:

  • All eight cost fields estimated per use case, separated into one-time and ongoing
  • Unit metrics defined (cost per request, training run, user)
  • Unit economics per use case positive or deliberately accepted
  • Cost levers anchored as an ongoing process: caching, batching, prompt and context optimization, more compact models, cheap compute windows
  • Costs assigned to those who cause them, showback or chargeback established
  • Budgets, quotas, and approval thresholds set
  • Automated cost monitoring with alerts active
  • Budget forecast produced with scenarios (best, average, worst) and seasonal effects
  • Cost-benefit assessment calculated with ROI, total cost of ownership, and break-even
  • Monthly cost overview with forecast and variance established
  • Cost awareness trained: every training run and every query incurs cost
  • Procurement and exit rules contractually secured