Solution Atlas
EverydayUser storyConsultative playbook

Engineering teams are not accountable for their cloud spend

The CFO has asked why product-team cloud costs are flat to the org rather than charged back to product P&Ls. Today there is no team-level allocation, no internal billing, no monthly cadence with product teams reviewing their spend against budget.

Trigger
CFO pressure on cost accountability; product P&L pressure.
Good outcome
Tag taxonomy enforced, monthly showback live to BU finance partners, cost dashboard per product team.
Diagnostic discovery

Signals this story fits

Observable cues that confirm the conversation belongs here.

  • ·CFO has asked why cloud cost is flat to the org rather than charged to product P&Ls
  • ·No team-level cloud allocation today
  • ·Tag taxonomy absent or inconsistent across subscriptions
  • ·Product teams view cloud cost as "central IT problem"
  • ·No monthly cadence with BU finance partners reviewing spend

Questions to ask

Open-ended, SPIN-style — each one has a reason it matters.

  1. 1.How is cloud cost currently allocated — central IT, by subscription, by tag, or not allocated?

    WhySurfaces the current allocation maturity.

    Listen for: “it all sits in central IT” · “we allocate by subscription which does not match teams” · “we have no tags”

  2. 2.Who sponsors cost accountability — CFO, CIO, both?

    WhyDetermines the political shape of the engagement. CFO-sponsored is faster.

  3. 3.What does the BU finance partner do today when cloud cost is mentioned?

    WhySurfaces the engagement model. The BU finance partner is the day-to-day operating relationship.

  4. 4.What tag taxonomy exists today — product, environment, owner, cost-centre?

    WhyDrives the policy and showback design. Without tags, allocation is impossible.

  5. 5.Is the goal showback (visibility), chargeback (real billing), or pseudo-chargeback (allocation on the P&L)?

    WhyCritical scope decision. Real chargeback needs finance system integration; showback is faster.

  6. 6.Are there shared services (network, security, platform) that need allocation alongside team consumption?

    WhyShared-services allocation is often the hardest part. Surfaces phase-two scope.

Baseline → target architecture

TOGAF-style gap framing — what we typically see today, and what the proposed end state looks like. The gap between them is the engagement.

Baseline architecture

Cloud cost sits in central IT with no team-level allocation. Tags are absent or inconsistent — some subscriptions are clean, others have tag drift. Product teams see cloud as a central cost, not their own. No monthly cadence with BU finance. Cost Management exists at tenant level but not exposed to product owners. CFO sees the total bill but cannot attribute it.

Typical concerns

  • ·Product teams have no cost accountability
  • ·Cloud spend grows without business pushback
  • ·Tag taxonomy missing — allocation impossible
  • ·No monthly cost cadence with BU finance
  • ·CFO cannot answer "which team caused the increase?"

Capability gaps

  • ·Tag taxonomy enforced via Azure Policy
  • ·Cost Management views per product team
  • ·Monthly showback artefact for BU finance
  • ·Anomaly detection on team-level spend
  • ·Shared-services allocation model
Target architecture

Tag taxonomy enforced via Azure Policy at subscription creation — every resource carries product, environment, owner, and cost-centre. Microsoft Cost Management produces team-level views consumed via Power BI by BU finance partners. Monthly showback pack distributed automatically with variance commentary. Shared-services allocation runs on an agreed cost-driver basis (typically headcount or compute share). Quarterly chargeback cadence with named product owners. CFO sees the breakdown live.

Key capabilities

  • Enforced tag taxonomy via Azure Policy
  • Cost Management views per team
  • Power BI showback dashboards
  • Monthly showback pack with variance commentary
  • Shared-services allocation model

Enabling SKUs

Resolved in the ‘Recommended cards’ section below.

Architecture decisions

Each decision is offered as explicit options with trade-offs — Hohpe's “selling options” principle. A safe default is noted where one exists.

  1. Decision 1.Allocation model — showback only vs pseudo-chargeback vs real chargeback

    Showback only

    When it fitsFirst phase; behaviour change is the goal; teams see their numbers but no internal billing.

    Trade-offsNo financial consequence; behaviour change requires CFO follow-through.

    Pseudo-chargeback (P&L allocation)

    When it fitsP&Ls run by BU; cloud cost shows on the P&L without internal invoicing.

    Trade-offsAllocation rules become contentious; needs CFO arbitration.

    Real chargeback (internal invoicing)

    When it fitsMature finance organisation; internal billing systems already in place.

    Trade-offsIntegration with the GL needed; allocation disputes need formal resolution.

    Default recommendationShowback in phase one; pseudo-chargeback in phase two once the allocation rules are agreed. Real chargeback only where the finance organisation is mature.

  2. Decision 2.Tag enforcement model — Azure Policy hard deny vs audit-and-remediate vs deployment template-enforced

    Azure Policy hard deny

    When it fitsGreenfield estate; teams accept policy guardrails.

    Trade-offsBreaks legacy deployments; needs an exception process.

    Audit-and-remediate

    When it fitsBrownfield estate; non-compliant resources need remediation rather than denial.

    Trade-offsSlower behaviour change; remediation backlog.

    Deployment-template-enforced

    When it fitsLanding zones with mandatory templates; tags applied at template level.

    Trade-offsOut-of-band deployments bypass; needs platform team buy-in.

    Default recommendationAudit-and-remediate for brownfield; Azure Policy hard deny on new subscriptions once the taxonomy is stable.

  3. Decision 3.Shared-services allocation basis — headcount vs compute share vs flat-rate per team

    Headcount-based

    When it fitsNetwork and platform services serve teams roughly proportional to size.

    Trade-offsPenalises growth; ignores actual consumption.

    Compute share (% of team consumption)

    When it fitsShared services scale with consumption (network egress, security scanning).

    Trade-offsPunishes legitimate scaling.

    Flat-rate per team

    When it fitsPredictability matters more than precision; small number of teams.

    Trade-offsSubsidises heavy users.

    Default recommendationCompute share for elastic shared services; flat-rate per team for platform services where the value is roughly equal.

Low-risk trial — proof of value

60-day showback rollout for top 5 product teams

8 weeks

Tag taxonomy agreed and policy-enforced for the trial subscriptions. Cost Management views configured per product team. Power BI showback dashboard live for the top 5 product teams (typically the highest spenders). Monthly showback pack template designed and distributed for the trial month with variance commentary drafted from Cost Management anomaly data. BU finance partners onboarded as the named recipients. Cadence and ownership documented.

Success criteria

  • Tag taxonomy enforced for the trial subscriptions
  • 5 product teams have a Power BI cost dashboard they can read
  • Monthly showback pack distributed and reviewed by BU finance
  • CFO accepts the showback pack as the standing artefact

InvestmentCost Management is included with Azure. Power BI Pro for the BU finance partners + Tagging Policy enablement effort. Estimated ~€2–4k/month for the trial scope plus a one-off ~€10–20k for the tag remediation work.

Proof metrics

  • ·Tag coverage >90% for trial subscriptions
  • ·Showback pack reviewed by BU finance for two consecutive months
  • ·At least one product team makes a measurable cost-driven decision
  • ·CFO endorses the showback model as the standing artefact

Recommended cards

The SKUs and capabilities most likely to be part of the solution, with the editorial rationale for each in the context of this story. Add the ones that fit your situation.

Why for this story

The source of truth for Azure cost. The team-level views, anomaly signals, and budget alerts all flow from Cost Management — the substrate the showback dashboards consume.

Why for this story

The presentation surface for BU finance partners. Power BI turns the Cost Management feed into the team-level dashboards and the monthly showback pack — the artefact that actually changes behaviour.

Back to Showback & chargeback