Why Chargeback Matters for Microsoft Licensing
Most enterprises manage Microsoft licensing as a central IT cost — negotiated centrally, paid centrally, and allocated to business units as a flat overhead. This model has a structural flaw: it eliminates the financial incentive for business units to optimise their licence consumption. When a business unit pays £30.10/user/month for M365 E3 regardless of whether each user actively uses the licenced capabilities, and when reducing that count saves the IT department's budget rather than the business unit's budget, the person closest to the usage data has no motivation to act on it.
Internal chargeback — allocating actual Microsoft licensing costs to the consuming business unit — changes this incentive structure fundamentally. Organisations with well-designed chargeback programmes consistently achieve 25–40% higher licence utilisation rates and require 15–20% fewer EA renewals before reaching right-sized licence counts compared to organisations managing Microsoft costs purely centrally. The mechanism is straightforward: when a business unit's budget is directly charged for licences it holds, that business unit becomes a motivated participant in licence harvest, right-sizing, and optimisation programmes.
Chargeback vs showback: Chargeback actually transfers cost to business unit budgets — the business unit's P&L is charged for its Microsoft licence consumption. Showback provides visibility into what costs would be allocated without actually transferring the charge. Showback is a useful first step for organisations building data maturity and stakeholder awareness. But the behavioural change that drives optimisation requires chargeback — when there is no actual financial consequence, the visible data rarely generates the urgency needed to act on it.
The Four Chargeback Model Options
There is no single correct chargeback model for Microsoft licensing — the right choice depends on your organisational structure, licence complexity, and the behavioural outcomes you want to drive. The four primary models each have distinct strengths and weaknesses.
| Model | Allocation Basis | Strengths | Weaknesses |
|---|---|---|---|
| Headcount-based flat allocation | Each business unit is charged based on its headcount as a % of total employees | Simple to implement; no per-user tracking required; predictable for BU budget planning | Does not reflect actual licence consumption; no optimisation incentive; penalises lean teams |
| Assigned licence allocation | Each BU is charged for licences currently assigned to its users | Directly ties cost to assigned licences; creates incentive to harvest unused licences | Requires accurate user-to-BU mapping in M365 Admin; does not penalise low utilisation within assigned licences |
| Active utilisation-based allocation | Each BU is charged based on measured active usage of licenced products (active users, not just assigned) | Most accurate reflection of value received; strongest optimisation incentive; rewards high utilisation | Requires usage data collection; more complex to administer; creates instability in BU budget forecasting |
| Hybrid: assigned + utilisation premium | Base charge on assigned licences; additional charge (or credit) based on utilisation rate vs threshold | Balances predictability with optimisation incentive; graduated approach to behavioural change | More complex to design and communicate; requires both assignment and usage data |
The Assigned Licence Model: The Recommended Starting Point
For most enterprises beginning a Microsoft chargeback programme, the assigned licence model is the correct starting point. It is significantly more effective than headcount allocation at driving licence optimisation, requires only the data already available in the Microsoft 365 Admin Centre and Microsoft Entra ID, and is straightforward to communicate to business unit finance leaders.
The core mechanics are simple: at the end of each month, pull the assigned licence count from M365 Admin Centre by user, map each user to their business unit based on HR system data or Entra ID department attribute, and apply the per-unit EA price for each licence type to produce a per-BU monthly charge. For a mid-size enterprise running this process on a 2,000-user population with 4 licence tiers, the monthly run takes 2–3 hours with appropriate automation.
The critical design requirement is the billing trigger for leavers. Assigned licences on accounts that are disabled or deleted continue to consume EA commitment unless actively revoked. The assigned model should charge the business unit for licences on active accounts and apply a 30-day grace period for accounts disabled due to employee departures — creating a direct financial incentive for each business unit's HR and IT liaison to process leaver licence revocations promptly.
When to Use Utilisation-Based Allocation
Utilisation-based allocation is appropriate for organisations that have already achieved reasonably clean licence assignment governance and want to drive the next level of optimisation — moving from "right-sized assignment" to "actively consumed licences justify their cost." It is particularly effective for high-value add-ons like Microsoft 365 Copilot (~£24.70/user/month), where the question of whether each licenced user is actively extracting value is commercially critical.
The data source for utilisation-based allocation is the Microsoft 365 usage reports available in the M365 Admin Centre — providing per-product monthly active user data at the user level. For M365 E3 and E5, "active use" is typically defined as meaningful activity in at least two of the core workloads (Exchange, Teams, SharePoint, OneDrive) within the billing period. For Copilot, active use data is available through the Copilot Dashboard and Microsoft 365 Admin Centre reports.
Pricing the Internal Chargeback Rate
Determining what price to use in internal chargeback calculations is a design decision with commercial and political implications. The three options are: EA price (what you actually pay Microsoft), list price (Microsoft's published retail price), or a blended internal transfer price that includes overhead allocation.
| Pricing Basis | What It Is | When to Use | Implications |
|---|---|---|---|
| EA price (actual cost) | The net per-unit price in your current EA, including negotiated discounts | When the goal is accurate cost recovery — charge BUs what IT actually paid Microsoft | Accurate; requires EA price confidentiality management; gives BUs full benefit of negotiated discounts |
| List price | Microsoft's published price list prices | When IT wants to retain discount value as an IT department "saving" to reinvest | Overcharges BUs relative to actual cost; creates internal margin that funds IT optimisation; politically contentious |
| Blended internal rate | EA price + overhead allocation for IT licence management time and tooling | When IT wants to recover the full cost of running the licence governance programme | Most accurate total cost allocation; requires overhead cost modelling; complex to communicate |
The EA price basis is the most defensible from a finance and audit perspective — it ensures that the sum of all chargeback amounts equals the actual Microsoft invoice, with no internal margin accumulation. For organisations with strong governance programmes, using the blended internal rate (EA price + governance overhead) is a legitimate approach that recovers the cost of the programme that makes chargeback possible. Using list price as the chargeback basis is technically simpler but creates an internal margin that requires justification in most organisations' finance governance frameworks.
Product-Level Chargeback Design
Not all Microsoft products should be chargedback at the same granularity. For foundational products like M365 E3 (the standard productivity tier for most enterprise employees), a per-user assigned licence model is appropriate and administratively efficient. For high-value, discretionary add-ons — Copilot, Teams Phone, Project licences, Dynamics 365 — individual product-level chargeback creates the strongest optimisation incentive because these are the products where per-BU adoption decisions are most consequential.
Recommended Chargeback Treatment by Product Category
| Product Category | Recommended Chargeback Treatment | Key Data Source |
|---|---|---|
| M365 E3 / E5 (base productivity) | Per-user assigned, monthly — with leaver revocation trigger | M365 Admin Centre — Licences page |
| M365 Copilot | Per-user utilisation-based — charge only for users active in prior 30 days | Copilot Dashboard, M365 usage reports |
| Teams Phone | Per-user assigned — Teams Phone is a dedicated workload licence | M365 Admin Centre — Teams Phone licences |
| Project / Planner Premium | Per-user assigned, quarterly review — lower frequency sufficient | M365 Admin Centre — Project licences |
| Power Apps / Power Automate per-user | Per-user assigned, with monthly active use bonus/malus | Power Platform Admin Centre — user activity |
| Dynamics 365 | Per-user assigned — separate chargeback by D365 application tier | Power Platform Admin Centre — D365 licences |
| Azure consumption | Subscription/resource group tagging model — not user-based | Azure Cost Management — subscription tags |
| Security add-ons (Defender, Entra P2) | Incorporate into E5 or charge as flat security overhead per user | M365 Admin Centre — security licences |
Azure Chargeback: The Tagging Imperative
Azure cost allocation deserves separate treatment because the mechanism differs fundamentally from M365 licence-based chargeback. Azure consumption is charged at the subscription and resource level — not the user level. The correct chargeback architecture for Azure uses Azure resource tags to attribute cloud spending to business units, cost centres, and projects. Without a consistent tagging taxonomy enforced at resource deployment, Azure consumption cannot be allocated to consuming business units — and the entire Azure bill defaults to an unallocated IT overhead.
For organisations with a Microsoft Azure MACC commitment, the chargeback model should reflect the consumption draw-down from the MACC as a pro-rated allocation across the subscriptions consuming Azure resources. The MACC provides a predictable commitment value; tagging provides the per-BU attribution. Together they produce a defensible Azure chargeback model. See our guide to Azure MACC mechanics for the underlying commitment structure.
Connecting Chargeback to EA Negotiation
A well-run chargeback programme generates the most commercially valuable asset in an EA negotiation: accurate, business-unit-level data on actual licence consumption versus assigned licence count. This is the data that justifies count reductions at renewal, demonstrates utilisation efficiency to Microsoft, and enables construction of the optimised licence position before the renewal conversation begins.
The connection works in both directions. Pre-renewal, chargeback data reveals which business units are over-licenced (assigned count exceeds active use) and which are approaching saturation (active use justifies the assigned count). This analysis drives the right-sizing exercise that determines the renewal count position — see our guide to EA renewal preparation for the full methodology.
Post-renewal, chargeback at the new EA prices cascades the negotiated discount to the business units consuming the licences. When IT negotiates a 15% improvement in M365 E3 pricing, business unit chargeback rates reduce by 15% from the renewal date. This creates visible value for the chargeback programme and reinforces IT's role as a commercial partner rather than a cost centre.
Chargeback data as negotiation anchor: The most powerful use of chargeback data in EA negotiations is constructing a utilisation-based renewal count position. If your chargeback programme shows that 2,300 of 3,000 assigned M365 E3 licences have 90-day active users, and you can document this with M365 Admin Centre usage reports, your renewal position for M365 E3 is 2,300 — not 3,000. The £7/user/month saving on 700 licences is £58,800/year, or £176,400 over the three-year EA term. Chargeback data is what makes that position defensible to both Microsoft and your internal finance function.
Implementation: The Four-Phase Approach
Phase 1: Data Foundation (Months 1–3)
Before implementing chargeback, establish the data infrastructure. Ensure that all M365 users have accurate department attributes in Microsoft Entra ID — this is the primary mechanism for mapping licences to business units. Validate the accuracy of the user-to-department mapping against your HR system. Establish the licence pricing table for all products in your EA. Export the M365 licence assignment report and produce a baseline allocation showing what chargeback amounts would be for the previous quarter under the proposed model.
Phase 2: Showback Launch (Months 4–6)
Launch as showback first — provide business unit finance leaders with monthly reports showing what their licence costs would be under chargeback, without actually charging. This educates stakeholders, surfaces data quality issues, and generates the first licence optimisation actions as business units identify unneeded assignments before they become actual charges. Use the showback period to refine the allocation model and resolve anomalies before real budget transfers begin.
Phase 3: Chargeback Activation (Month 7+)
Activate chargeback with actual budget transfers. Set a 90-day advance notice policy — business units should know their projected monthly charge at least 90 days before it is applied, allowing time to act on optimisation opportunities before the charge hits. Implement the leaver notification SLA (business units have 30 days to process leaver licence revocations before continued charges apply). Establish a monthly chargeback report distribution to BU finance leads and the IT licence governance team.
Phase 4: Continuous Optimisation Loop
The sustainable value of chargeback comes from the monthly optimisation loop it creates: BU finance leads review charges, identify high-cost users with low utilisation, escalate to IT for licence downgrade or revocation, IT processes changes in the M365 Admin Centre, and the following month's chargeback reflects the reduction. This loop — which does not operate without the financial incentive chargeback creates — is what produces the 25–40% utilisation improvement organisations achieve over 12–18 months of running a chargeback programme. For governance cadence design, see our guide to Microsoft licensing governance frameworks.
Common Chargeback Design Mistakes
Charging for the full EA commitment regardless of assignment. If your EA has 5,000 M365 E3 licences but only 4,200 are assigned, charging business units for 5,000/headcount-share includes 800 licences that no business unit is actually using. This creates an unallocated central IT cost that obscures the waste, rather than attributing it to the decision-making unit that should have reduced the count. Only charge for assigned (or active) licences — hold unassigned licence cost as a visible IT "waste" line item that targets renewal action.
Using list price without transparency. If business units are being charged Microsoft list prices when IT pays EA prices, the difference is an internal margin that business units will eventually discover and challenge. This undermines trust in the chargeback programme and creates political friction that can derail the optimisation agenda. Use EA prices and be transparent about the negotiated discount position.
No leaver process integration. Chargeback without an automated or semi-automated leaver licence revocation process creates ongoing charges for departed employees. The HR-to-IT leaver notification SLA must be part of the chargeback design — not an afterthought. A 30-day leaver revocation SLA with BU accountability for delayed revocations is the standard governance approach.
Charging for licences with no optimisation pathway. If a business unit is charged for Copilot licences but IT has no mechanism to revoke or reassign them mid-term, the charge creates frustration without an action path. Every chargeback trigger should have a corresponding optimisation action available — the financial signal only works if there is a process to act on it.
Get Microsoft Licensing Intelligence Weekly
Independent analysis on EA optimisation, chargeback, and cost governance — direct to your inbox every week.
No spam. Unsubscribe at any time.
Ready to implement Microsoft licensing chargeback?
We design and implement chargeback models that drive measurable licence reduction and integrate directly with your EA renewal strategy. 500+ engagements. $2.1B managed. 32% average cost reduction.
Book a Free Advisory SessionMicrosoft Cost Optimisation Framework
The systematic approach to reducing Microsoft EA spend — from licence harvest to chargeback to renewal strategy.
Download the framework →