PRO-FOCP

Final check — FinOps Certified Practitioner (FOCP) — Exam-Prep

Answer the 50 questions. You need 80% or more to pass. You can retry as many times as you like.

Question 1A newly formed FinOps team is struggling because engineers ignore cost reports and finance distrusts the cloud bill. According to the FinOps principles, which approach best resolves this tension?Mandate that finance approves every cloud resource before it is provisionedFoster cross-functional collaboration and shared accountability where teams take ownership of their cloud usageTransfer all cost decisions to a centralized cost-control officeFreeze all new spending until a full audit is complete

Question 2Which statement best reflects the FinOps principle regarding the value of cloud spend?The lowest possible cloud bill is always the goalCloud spend should be evaluated purely against last year's budgetTeams should make trade-off decisions weighing cost against speed, quality, and business valueCost optimization must always take priority over feature delivery

Question 3A FinOps practitioner notes that the cloud's variable, consumption-based cost model is a fundamental cultural shift from traditional IT. What is the primary implication of this shift?Spending decisions become decentralized and happen in near real time as engineers provision resourcesCapacity planning is done once per year by a procurement committeeCosts are fixed and predictable like an on-premises depreciation scheduleOnly the finance department can influence the monthly cost

Question 4In the FinOps lifecycle, which phase focuses primarily on creating cost visibility, allocation, and shared understanding of cloud spend?OptimizeOperateInformBenchmark

Question 5An organization can allocate only 62% of its cloud bill to specific teams because the rest sits in untagged resources and shared services. Which Inform-phase action will most directly improve allocation coverage?Purchase additional Savings Plans to lower the rateEstablish and enforce a tagging strategy plus rules to split shared costsRightsize the largest virtual machinesSwitch the untagged resources to spot instances

Question 6What is the key difference between showback and chargeback?Showback applies only to reserved instances; chargeback applies only to on-demandShowback reports cost back to teams for visibility without billing them, while chargeback financially bills the consuming teamsShowback is performed by finance; chargeback is performed by engineeringChargeback is always more accurate because it ignores shared costs

Question 7A team has tagged 90% of resources, but a report still shows large 'unallocated' costs from data transfer and a shared Kubernetes cluster. What is the most appropriate next step?Define proportional or usage-based allocation rules for shared and untaggable costsDelete the shared cluster to remove the ambiguityReport only the tagged 90% and ignore the remainderMove all shared costs into a single team's budget at random

Question 8Why is a consistent tagging (or labeling) taxonomy considered foundational to the Inform phase?It automatically reduces the per-unit price of computeIt enables accurate cost allocation, reporting, and accountability by mapping spend to teams, products, and environmentsIt replaces the need for budgets and forecastsIt is only required for compliance audits, not for FinOps

Question 9A chargeback model is being rolled out, but engineering leaders complain it creates friction and disputes over shared platform costs. Which practice best preserves accountability while reducing friction?Abandon allocation entirely and report only the total billAgree on transparent, mutually understood allocation rules and start with showback before full chargebackBill all shared costs to the FinOps team's cost centerCharge each team an equal flat fee regardless of usage

Question 10A cost report shows a development environment running production-sized instances at full capacity 24/7, yet utilization metrics show CPU rarely exceeds 8%. Which usage-optimization action is most appropriate first?Buy a 3-year Reserved Instance for the oversized instancesRightsize the instances to a smaller type and schedule them to stop outside working hoursIncrease the instance size to avoid future bottlenecksAdd a Savings Plan to discount the current rate

Question 11Which scenario is the clearest example of usage optimization rather than rate optimization?Committing to a 1-year Savings Plan to lower the hourly priceNegotiating an enterprise discount agreement with the providerDeleting orphaned disks and idle load balancers that are no longer usedPurchasing Reserved Instances for a stable production workload

Question 12When rightsizing a workload, why is it important to analyze utilization metrics over a representative time window (e.g., 2–4 weeks) rather than a single peak hour?Because providers only bill based on peak usageTo capture cyclical patterns and peaks so the new size meets performance needs without overprovisioningBecause a single hour always represents the true averageTo ensure the instance is always downsized to the smallest available type

Question 13A FinOps practitioner finds several non-production environments idle every night and weekend. Which optimization technique addresses this most directly?Workload scheduling (start/stop automation) to turn resources off when not neededPurchasing a Compute Savings Plan for those environmentsMigrating the environments to a more expensive regionIncreasing the storage tier to a higher-performance class

Question 14Which of the following is the best definition of FinOps as commonly described by the FinOps Foundation?A tool that automatically deletes unused cloud resourcesAn operational framework and cultural practice that maximizes the business value of cloud by enabling collaboration between engineering, finance, and business teamsA finance-only discipline for auditing cloud invoicesA procurement process for negotiating hardware contracts

Question 15Engineers complain that cost data arrives 30 days after spend occurs, making it useless for decisions. Which FinOps principle most directly supports fixing this?Cost data should be timely and accessible to drive better real-time decisionsOnly month-end reconciled data is trustworthy enough to useFinance should be the sole owner of all cost dataCost decisions should wait until the annual budget cycle

Question 16A platform team offers a shared logging service used by many product teams. To distribute its cost fairly in the Inform phase, which allocation approach is generally most defensible?Allocate the entire cost to the platform team that owns the serviceSplit the cost evenly across all teams regardless of how much they logAllocate cost based on each team's measured consumption (e.g., volume of logs ingested)Leave the cost unallocated since it is a shared service

Question 17During the cultural shift to FinOps, a leader insists every cost decision must be escalated to a central committee. Why does this conflict with FinOps best practice?Because FinOps forbids any governance or guardrailsBecause FinOps pushes decentralized ownership so teams who incur the cost can act quickly within agreed guardrailsBecause only engineers are allowed to make cost decisions in FinOpsBecause central committees are required to approve every rightsizing change

Question 18A FinOps practitioner is struggling to get an engineering team to act on a cost-optimization recommendation. Which persona is BEST positioned to provide the executive air-cover and mandate needed to drive accountability for cloud spend across teams?The procurement analystLeadership / executives (e.g., CTO, CFO)An individual software developerThe cloud vendor's account manager

Question 19In the FinOps Framework, where is the FinOps team most effectively positioned within an organization to fulfill its mandate?Embedded exclusively inside the central finance departmentAs a sub-team reporting only to procurementAs a central function that connects and enables finance, engineering, product and leadershipInside each engineering squad with no central coordination

Question 20A product manager wants the FinOps team to help them understand the cost of a specific feature so they can decide on its pricing. Which FinOps stakeholder responsibility does this primarily reflect?Negotiating enterprise discount agreements with the vendorConnecting cloud cost data to business value and unit economicsPatching security vulnerabilities in the workloadApproving the corporate annual capital budget

Question 21An organization's monthly cloud bill suddenly spikes 40% over the daily run-rate without any planned launch. Within the FinOps lifecycle, this is BEST handled as part of which capability?Anomaly managementCommitment-based discount purchasingRightsizing workloadsChargeback configuration

Question 22A FinOps team builds a forecast by taking historical spend and projecting it forward using growth rates and known seasonality, with little input from engineering roadmaps. Which forecasting approach does this describe?Driver-based (bottom-up) forecastingTrend-based (top-down) forecastingZero-based budgetingChargeback reconciliation

Question 23A finance leader complains that variance reports compare actual cloud spend only to a static annual budget set 11 months ago, making them useless for early action. What is the MOST effective FinOps improvement?Stop producing variance reports entirelyAdopt rolling, continuously updated forecasts and compare actuals against themSwitch all accounts to chargebackIncrease the annual budget by the variance amount

Question 24An anomaly-detection alert fires every time a nightly batch job runs, even though that spike is expected and recurring. What is the appropriate FinOps response to reduce alert fatigue?Disable all anomaly alerts across the organizationTune thresholds/baselines so expected, recurring patterns are not flaggedShut down the batch job to stop the costMove the workload to a different cloud provider

Question 25A SaaS company reports that its 'cost per active customer' fell from $0.42 to $0.31 while total cloud spend rose. Which FinOps concept does 'cost per active customer' represent, and what does the trend signal?A unit economics metric; improving efficiency despite growing spendA tagging coverage metric; worsening allocationA commitment coverage metric; under-utilized reservationsAn anomaly score; an active cost spike

Question 26An organization can produce cost reports and has some tagging, but allocation is manual, optimization is reactive, and there are no formal targets or governance. Which FinOps maturity stage best describes this?RunCrawlWalkSprint

Question 27Leadership asks the FinOps team to recommend a single KPI that best reflects whether cloud commitment purchases are being used effectively. Which KPI is most appropriate?Tagging coverage percentageCommitment (reservation/Savings Plan) utilization rateMean time to detect anomaliesNumber of cloud accounts

Question 28A FinOps team wants governance that scales without becoming a bottleneck. Which approach is most aligned with FinOps governance principles?Require central FinOps approval before any resource is provisionedDefine policies, guardrails and automated controls that empower teams to self-serve responsiblyEliminate all spending limits to maximize engineering velocityLet each engineer set their own untracked budget

Question 29During a monthly business review, the FinOps team must explain why spend rose. Which framing BEST embodies the FinOps cultural shift away from pure cost-cutting?'Spend grew 20%, but revenue per dollar of cloud spend improved 15% — this growth is efficient.''Spend grew 20%; we must cut it back to last month regardless of impact.''Spend grew 20%; this is finance's problem, not engineering's.''Spend grew 20%; we will freeze all provisioning immediately.'

Question 30A FinOps practitioner needs engineers to feel ownership of the costs they generate. Which mechanism most directly drives this accountability?Hiding cost data from engineers to avoid distractionShowback/chargeback that surfaces each team's costs back to them in near-real timeSending a single consolidated bill only to the CFOReviewing costs once a year at budget time

Question 31An e-commerce company sets a target that its 'cloud cost per order processed' must not exceed a defined ceiling as transaction volume scales. This target is an example of using unit economics for what purpose?Establishing a scalable efficiency benchmark tied to a business driverReplacing the need for tagging entirelyDetermining the vendor's list priceEliminating the forecasting process

Question 32A company's untagged ('unallocated') spend is 35% of the bill, blocking accurate showback. Which Operate-phase action best addresses this while reinforcing governance?Accept the 35% as overhead and stop measuring itSet and enforce a tagging-policy KPI with automated compliance checks and remediationPurchase more reservations to lower the rateSwitch the unallocated workloads off

Question 33A FinOps team wants to measure how quickly the organization responds to optimization opportunities. Which operational KPI is most relevant?Total number of EC2 instancesTime-to-action / time-to-remediation on recommendationsAverage HTTP response latencyNumber of support tickets opened

Question 34Finance has built the annual cloud budget in isolation, and engineering teams routinely exceed it because they were never consulted. Which FinOps practice would MOST improve budget accuracy and buy-in?Collaborative, driver-based budgeting that incorporates engineering roadmaps and forecastsLocking the budget and penalizing any team that exceeds itLetting engineering set the budget with no finance reviewRemoving budgets and relying only on after-the-fact reports

Question 35A new cloud team has full cost visibility and accurate allocation, but spend is still growing faster than business value. Which FinOps lifecycle phase should they focus on next?Inform, to improve tagging coverage furtherOptimize, to reduce rate and usage wasteOperate, to publish more dashboardsRestart the lifecycle from scratch

Question 36The three iterative phases of the FinOps lifecycle, in order, are:Plan, Build, RunMeasure, Allocate, ReportInform, Optimize, OperateDiscover, Govern, Automate

Question 37An engineering manager says 'we can't optimize what we can't see.' Which lifecycle phase is a prerequisite to effective optimization?OperateInformOptimizeGovern

Question 38Which statement BEST describes how the FinOps lifecycle should be applied across an organization?It runs once as a one-time project, then stopsDifferent teams can be in different phases simultaneously and iterate continuouslyAll teams must complete Inform before any team may start OptimizeOnly the central FinOps team progresses through the phases

Question 39During the Operate phase, what is the primary goal?Define cost allocation rules and tagging taxonomyContinuously establish processes, governance and align cloud usage to business objectivesPurchase as many commitments as the budget allowsBuild the first showback report

Question 40A workload runs 24/7 on the same instance family and region for the foreseeable future. Which rate optimization mechanism typically yields the BEST discount for this steady-state, instance-specific demand?On-demand pricingA Standard Reserved InstanceSpot/preemptible instancesPaying with a credit balance

Question 41A team wants commitment-based discounts but needs flexibility to change instance families, regions, and even compute services over time. Which option fits BEST?Standard Reserved InstancesCompute Savings Plans (commit to $/hour of compute spend)Spot instances onlyZonal Reserved Instances

Question 42Rate optimization through commitments is MOST effective when:Usage is highly variable and unpredictableThere is a stable, predictable baseline of usage to coverThe workload will be decommissioned next monthThe team has no historical usage data

Question 43Best practice sequencing suggests teams should rightsize and eliminate waste BEFORE buying large commitments. Why?Commitments are cheaper if bought at nightOtherwise you risk committing to oversized or wasteful usage you would have removedRightsizing automatically cancels existing commitmentsProviders forbid commitments on rightsized resources

Question 44A FinOps analyst reports commitment coverage of 90% but utilization of only 60%. What does this MOST likely indicate?The organization is under-committed and should buy moreCommitments were over-purchased relative to actual eligible usage, wasting moneyCoverage and utilization always move togetherOn-demand spend is fully eliminated

Question 45Who should typically be involved in approving large multi-year commitment purchases in a mature FinOps practice?Only an individual engineer, to move fastA cross-functional group including Finance, FinOps and engineering ownersThe cloud provider's sales team aloneMarketing, since they own the budget

Question 46Spot/preemptible instances offer the deepest discounts but can be reclaimed by the provider. They are BEST suited for:A stateful production database with strict SLAsFault-tolerant, interruptible batch or stateless workloadsA workload requiring guaranteed long-term capacityAnything, since reclamation never happens in practice

Question 47A Convertible Reserved Instance differs from a Standard Reserved Instance primarily because it:Offers a higher discount than Standard RIsAllows exchanging for different instance attributes during the term, with a lower discountCannot be purchased for a 3-year termApplies only to storage, not compute

Question 48To decide how much commitment to purchase, a FinOps team should base the amount on:The maximum peak usage ever recordedThe stable usage baseline, often informed by forecastingA fixed percentage mandated by the providerWhatever maximizes the provider's revenue

Question 49In the FinOps lifecycle, anomaly detection and alerting most directly support which phase, and feed which other phase?They support Optimize and feed Inform's tagging rulesThey support Inform (visibility) and trigger Optimize/Operate actionsThey belong only to Operate and never inform other phasesThey are unrelated to the FinOps lifecycle

Question 50Rate optimization and usage optimization are distinct levers within Optimize. Which pairing is correct?Rate = buying RIs/Savings Plans; Usage = rightsizing and shutting down idle resourcesRate = rightsizing; Usage = buying RIsBoth refer to the same activityRate = tagging; Usage = chargeback

Final check — FinOps Certified Practitioner (FOCP) — Exam-Prep

Already paid?