Prepared October 15, 2026. Evidence: Microsoft's four FY2026 earnings calls, October 29, 2025–July 29, 2026, and FY2026 financial filings. This is a proposed owner/operator plan, not an announced transaction. It uses the same transcript set as Xbox: The Reset. It does not claim to incorporate subsequent operating developments.
The decision
Carve out the entire gaming business into an independently managed Xbox company, prove its operating economics, then pursue a full distribution to Microsoft shareholders. Include the studios and franchises, Activision/Blizzard/King, Minecraft, Game Pass, gaming commerce, player services, and the hardware operation. Keep the portfolio together initially. Give the new board authority over product availability, subscriptions, studio funding, pricing, hardware and capital allocation.
My preferred sequence is an operational separation immediately, a separation decision after the first 100 days, and a legal spin targeted for months 12–18 only if the readiness gates below pass. These are proposed planning intervals, not estimates supplied by Microsoft. A spin should create an accountable games company with a viable balance sheet. It should not be a way to transfer liabilities, finance a parent dividend, or hide weak projects behind a new reporting entity.
The operating mandate: turn valuable player relationships and intellectual property into repeatable cash returns. Sell games where customers want to play; preserve the benefits of the owned platform; make Game Pass earn its content investment; ship fewer simultaneous bets when production capacity is the constraint; protect release quality and live-service reliability.
The reason to separate is a hypothesis about better decisions and accountability. The calls establish an operating problem, but do not prove Microsoft caused it or that independent ownership will fix it. First test these changes inside a ring-fenced subsidiary. If Microsoft grants durable autonomy and the quantified separation costs exceed the incremental benefits, retain that structure and stop the legal spin. A valuation rerating is not the operating case.
What the transcripts actually tell us
| Evidence | What it establishes | Operating response proposed here |
|---|---|---|
| October 29, 2025 — Nadella: “focused on our high-margin content and services.” Minecraft reached 155 million monthly active users; PC players set a record. Hood reported content/services growth of 1%, with third-party content better than expected. T1 | Valuable engagement and a broad distribution strategy. No disclosed standalone Xbox margin or customer lifetime value. | Separate publishing economics from store economics. Track spending retained when players change devices. Protect third-party developer participation. |
| January 28, 2026 — Hood: content/services was “below expectations driven by first-party content”; revenue fell 5%. T2 | An execution miss against management's own expectations, not merely an outside investor's disappointment. | Require a title-by-title budget-versus-actual bridge; install production gates and a launch-quality veto. The call does not identify a culpable franchise. |
| April 29, 2026 — Nadella: “win back fans.” He described Game Pass changes and record Xbox activity/streaming; Hood tied part of the next-quarter decline to recent price changes. T3 | Customer value is an explicit management priority; engagement alone is insufficient evidence of economic recovery. | Stabilize entitlements and billing, fix high-friction customer journeys, and test subscription retention and contribution together. Exact prices and churn are not disclosed here. |
| July 29, 2026 — Nadella: “reset the business for long-term growth”; expected growth in FY2027. Hood reported content/services down 10%, severance and impairment charges, and a further mid-single-digit decline expected next quarter. T4 | A portfolio/platform/operations reset, with recovery still ahead. No quantified savings plan or standalone Xbox cash flow. | Make each reset action accountable to a cash bridge and delivery milestone. Demand a definition of FY2027 growth and a downside plan if releases slip. |
FY2026 Xbox revenue was $21.790 billion versus $23.455 billion in FY2025: a calculated $1.665 billion / 7.1% decline. Annual content/services revenue fell 5%; hardware revenue fell 29%. These figures establish scale and pressure, not profitability. F4
Three analytical guardrails govern this plan: More Personal Computing profit is not Xbox profit; impairment charges do not by themselves establish an Activision goodwill write-down; and record users do not establish paying users or incremental cash generation. The public sources leave title returns, Game Pass cohorts, cloud delivery costs and standalone overhead unresolved. T1–T4, F4
1. Draw the perimeter before promising the savings
One company; distinct economic accounts. Start with three operating divisions: Studios & Publishing, Player Platform, and Devices. Within Studios & Publishing, maintain separate franchise/business accounts for Activision, Blizzard, King, Minecraft, Bethesda and the remaining studio portfolio. This is a proposed management structure, subject to the legal entity and rights inventory; it is not a claim about today's reporting lines.
| Asset or obligation | Proposed treatment | Day-one deliverable and accountable owner |
|---|---|---|
| Studios, owned game IP, trademarks, production tooling and staff | Transfer together with relevant development, licensing and service obligations. Audit ownership, encumbrances and joint rights title by title. | General counsel + chief studios officer: signed asset/rights register and named franchise GMs. |
| Game Pass, gaming storefront, player commerce and Xbox brand | Transfer the gaming economics, contracts and product decision rights. Preserve purchased entitlements and existing commitments. | Platform president + CFO: product/contract map reconciled to revenue, refunds, deferred revenue and partner payables. |
| Existing consoles, accessories, inventory and warranties | Transfer the operation with channel inventory, manufacturing commitments, support and warranty reserves. | Devices GM: installed-base support plan, inventory aging and contractual cash-out schedule. |
| Xbox accounts, saves, achievements, social graph, family settings and payments | Separate the gaming domain while initially federating with Microsoft identity under contract. Transfer only data and rights legally available; do not copy unrelated Microsoft account data. | CTO + privacy lead: dependency map, consent requirements, entitlement reconciliation and tested rollback. |
| Azure/cloud gaming, security services, Windows interfaces and shared tooling | Microsoft retains the underlying enterprise platforms; Xbox buys required services under explicit commercial agreements. Negotiate gaming-specific licenses and access rights. | CTO + procurement: capacity, pricing, support, portability, incident and termination schedules. |
| Treasury, payroll, finance, tax, HR, procurement, facilities and public-company controls | Build minimum viable standalone functions; use time-limited transition service agreements (TSAs) where necessary. | CFO + separation lead: service catalog, staffed replacement plan, monthly exit milestones. |
| Pre-close litigation, employee obligations, royalties, tax exposures and regulatory commitments | Allocate explicitly in negotiated agreements; do not assume liabilities or commitments disappear in a spin. | General counsel + CFO: liability schedule, consent/approval register, reserve and indemnity proposal. |
Microsoft relationship: commercial partner and initial owner during separation, then independent shareholder bases after the preferred full spin. Xbox needs authority to choose platforms and suppliers; Microsoft should not retain a veto over releases to support Windows or a right to impose unrelated AI spending. Conversely, Xbox must pay an economically supportable price for services it consumes. Neither perpetual subsidized Azure capacity nor an inflated transfer charge is a sound operating model.
Agreements to negotiate first: identity and entitlements; Azure capacity and service levels; game/engine/brand IP licenses where transfer is incomplete; distribution and content commitments; employees and benefits; and the balance-sheet/liability allocation. Existing platform, content and cloud-rights arrangements require contract-by-contract review. This report does not establish their assignability, current restrictions, or tax treatment.
Target TSAs for 12 months with a priced extension option, and an 18-month exit backstop where feasible—proposed negotiating objectives, with identity/security exceptions allowed only against a funded migration plan. Long-term commercial Azure or identity federation contracts are not automatically failed TSAs; temporary operational dependence without decision rights is the problem.
2. First 100 days: create the facts and change the decisions
Day 0 means board authorization of the operational carve-out, not a forecast closing date. All deadlines and thresholds in this plan are proposed management requirements. “Complete” means an accepted deliverable with evidence, not a presentation describing work underway.
| Deadline | Accountable owner | Deliverable | Acceptance test / decision |
|---|---|---|---|
| Day 7 | CEO + board | Operating charter; authority matrix; named CFO, platform, studios, devices and separation leads | No shared veto on approved gaming budgets; named escalation route for Microsoft dependencies. |
| Day 14 | CFO | Weekly 13-week cash forecast, commitment register and separation cost ledger | Opening cash reconciled; payroll, royalties, content milestones and hardware purchases included; forecast versus actual reviewed weekly. |
| Day 30 | CFO + data lead | Standalone economic accounts; subscriber/customer cohorts; franchise cost ledgers | Every material revenue stream reconciles to the ledger; internal transactions eliminated; unresolved allocations shown explicitly. |
| Day 30 | CTO + platform president | Reliability baseline; top customer-failure backlog; separation dependency map | Purchase, download, launch, save, refund and recovery journeys measured by platform; critical dependencies have owners and fallbacks. |
| Day 45 | Chief studios officer | Review of every funded title and live service, with cost-to-complete and contractual commitments | Each project gets a signed continue / rescope / partner / pause / cancel decision; no automatic cancellation quota. |
| Day 60 | Platform president + CFO | Game Pass and distribution experiment designs; first eligible tests launched | Predefined control, cohort eligibility, contribution metric and harm guardrails; no change to already promised entitlements. |
| Day 60 | CFO + procurement | Standalone services bids, Azure economic quote, duplicate overhead inventory | Savings have an actual cash cost, removal date, replacement cost and customer/production risk assessment. |
| Day 90 | Devices GM + CTO | Console and cloud investment cases | Hardware case includes retained store economics; cloud case includes metered delivery and capacity costs. Board can defer either commitment. |
| Day 100 | CEO + board | Separation investment case; funded release slate; customer repair results; downside liquidity plan | Approve, defer or stop the legal spin. Unpriced dependencies, unfunded obligations or customer continuity failures prevent approval. |
At Day 100, subscription tests may still lack enough renewal cycles and a title may not have shipped. Report interim evidence and the date of the next decision; do not declare long-term economics proven. The first 100 days should improve control of the business, not manufacture a recovery quarter.
3. Studios: finance the next milestone, not the original promise
Owner: chief studios officer; each franchise GM owns its complete cash account. The January miss is the direct trigger. It does not support identifying a particular studio as the problem. T2
Within 30 days, build one project card per title: executive producer, player proposition, playable build, target platforms, remaining headcount by discipline, external vendors, monthly burn, milestone dates, remaining development/marketing/live-service cash, existing contracts, expected receipts by channel, and downside demand. Show sunk spend separately from remaining spend.
Use four production gates: concept/prototype, validated playable slice, full production, and release readiness. Approve only the resources to reach the next gate. A greenlight committee comprising studios, finance and publishing assesses commercial assumptions; an independent production/quality lead signs off readiness. Creative decisions remain with accountable teams inside the agreed scope.
At each gate ask: does the updated, risk-adjusted value of remaining receipts exceed remaining cash costs, including launch support and the opportunity cost of scarce staff? Compare continuation with rescoping, licensing, external partnership and orderly closure. Prior investment is not a reason to continue; a lower forward return is not a reason to abandon an otherwise positive project without comparing alternatives.
Practical interventions: cut features before extending the calendar reflexively; stop carrying multiple unstaffed “priority” projects; move scarce engine/build specialists to funded milestones; consolidate vendor procurement and repeatable build/QA infrastructure where compatibility is proven. Do not force every studio onto one engine or replace production expertise simply to meet a headcount target.
Trigger a mandatory re-review if cost-to-complete rises more than 10% or a critical milestone slips more than 30 days—proposed governance thresholds, not evidence about current slippage. The response is a revised scope and investment decision, not an automatic cancellation. Publish externally only dates backed by a validated production plan. A launch with unresolved critical stability, payments, security or progression defects fails the release gate, even when delaying it hurts a quarter.
Franchise playbooks: for premium releases, measure paid unit receipts, refunds, channel fees and post-launch conversion; for recurring/live-service businesses, measure retained payers and contribution after the content calendar and server/support costs; for King/mobile, demand cohort-based paid-acquisition payback before scaling spend; for Minecraft, test marketplace/subscription/creator improvements against retained participation and player trust. These are distinct economic models to examine, not claims that any named franchise is currently failing. Minecraft's disclosed audience supplies a reason to protect the relationship, not a license to assume unearned monetization. T1
Pay producers partly on milestone predictability, product quality and post-launch returns. Awarding engagement credit without funding accountability would reproduce the problem the carve-out is meant to solve.
4. Game Pass: make the subscription and the games agree economically
Owner: platform president; CFO arbitrates portfolio economics. April's customer-value discussion and price-related guidance require joint measurement of retention and cash, rather than a reflex price increase. T3
Build monthly cohorts by plan, geography, acquisition channel, tenure, promotion, device and title-entry reason. Reconcile paying accounts, promotional accounts, recognized revenue, realized price, cancellations, involuntary payment failures, downgrades, reactivations and add-on spending. Link accounts using permitted identifiers; deduplicate cross-device customers.
Maintain two separate views:
- Accounting view: subscription receipts/revenue, external content minimum guarantees and royalties, allocated first-party content cost, customer support, payments, cloud delivery and platform overhead. Reconcile the consolidated result; an internal studio allocation is neither new revenue nor new cash.
- Decision view: expected portfolio cash with the proposed inclusion/price/benefit policy minus expected cash without it. Include incremental subscription and add-on receipts, displaced premium receipts, changes in retention and acquisition, incremental content/delivery costs, and longer-term franchise effects. Do not charge the entire development budget again if both alternatives incur it.
Proposed catalogue policy: keep existing customer promises. For future uncommitted releases, make day-one inclusion a title-level decision; evaluate premium launch windows, catalogue entry after the initial sales period, and simultaneous inclusion. Back-catalogue and discovery titles may fit a different policy from premium tentpoles. Avoid a universal answer before seeing the economics.
Start with eligible catalogue promotions and benefits where controls are feasible. Use randomized offer/marketing tests where lawful and fair; use matched cohorts and release analogues for availability decisions that cannot be randomized cleanly. Pre-register selection bias and uncertainty. Short-term subscriber spikes do not settle cannibalization or long-term retention. Observe multiple renewal cycles and revisit after six months before large policy expansion.
Decision rule: scale a change only when conservative incremental lifetime contribution is positive and cancellations, refunds, complaints and franchise retention remain within preapproved guardrails. If uncertainty straddles zero, narrow the experiment or retain the current policy. The CFO must show the whole-company result, so the platform cannot win its subscriber target by quietly transferring losses to publishing.
The first product repair should be comprehensible benefits, reliable billing, visible renewal terms, and working purchase/play/refund flows. Propose a 90-day pause on discretionary broad price/entitlement changes while auditing them, except necessary customer remediation or existing commitments. No claimed optimal price or subscriber count is available from this evidence set.
5. Distribution: sell widely and defend the customer relationship
Owners: publishing president + platform president; CFO resolves channel conflict. Microsoft's broader reach and PC records are evidence of access, not evidence that every new player has the same value. T1, T2
Default to releasing on commercially attractive platforms when rights, quality and port economics permit. Every new platform or exclusivity decision needs a one-page calculation: incremental receipts less platform fees, porting, certification, marketing, ongoing support and lost receipts elsewhere. Include the change in retained owned-store spending, Game Pass contribution and future hardware economics. Assess overlap across customers; do not count the same payer as incremental twice.
An exclusivity window is an investment. Charge its opportunity cost to the business asking for it, based on a documented counterfactual; eliminate that internal charge in consolidated reporting. If owned-platform retention cannot justify the foregone external contribution, widen distribution. If a port damages the experience or has negative incremental economics, delay or decline it. “Everywhere” is a customer strategy, not permission to ignore cost.
Create a console-to-PC/other-device migration dashboard: previously paying Xbox-console customers, subsequent activity, subsequent net spending captured by Xbox, retained subscription contribution, and third-party store contribution lost. Reach growth with falling captured contribution calls for a different intervention from outright customer loss.
Improve owned-store reasons to stay: reliable entitlement recognition, easy installs/updates, consistent cloud saves where supported, clear refunds, useful discovery and responsive developer support. Do not attempt to win through incompatible promises or by blocking purchased libraries.
Third-party partners need a named account owner, documented certification requirements, predictable submission/review status, payout reconciliation and escalation. Review certification cycle time, partner payment accuracy, support backlog and repeat launch commitments monthly. The positive third-party contribution in October argues for protecting this ecosystem while fixing internal content execution. T1
6. Hardware and cloud: preserve the installed base; make new capacity earn its keep
Hardware owner: devices GM. Maintain supported devices, purchases, peripherals, developer tools, security updates, warranty service and spare-parts obligations through separation. Publish a support policy grounded in actual contractual and technical capacity. The hardware decline argues for investment discipline; it does not establish that shutting hardware destroys less value than supporting it. F4
For each future device proposal compare: owned hardware, OEM partnership and continued support of the existing base. The investment case must include engineering/tooling, inventory, subsidies, returns, warranty, channel financing, customer acquisition, incremental subscriptions and retained third-party-store contribution. Stress volumes, component costs and release delays. An OEM device is not automatically capital-light once guarantees, marketing and support commitments are counted.
Pending the Day 90 review, pause new discretionary tooling commitments; honor existing contracts and required support. Approve new hardware only if a conservative lifecycle case is positive and can be funded after the content slate. If it fails, shift the next incremental hardware investment toward a viable partner model or defer it; do not strand existing players to improve a near-term cash number.
Cloud owner: CTO + platform president. Meter gaming workloads separately. Request Azure quotes by region, capacity type, peak reservation, egress and service support. Include idle reserved capacity, licensing, fraud and incident costs. Track cost per successful session and cost per streaming hour alongside queue time, latency, session failure and paid-cohort retention.
Classify cloud users by problem solved: no suitable device, trial before downloading, convenience on a second screen, or existing subscriber usage. Expand capacity where incremental retained contribution justifies its cost; optimize codecs, scheduling and regional footprint before adding fixed commitments. Any prospective fair-use or paid-capacity change must be tested and honor existing entitlements.
Treat Gaming Copilot or other AI features as bounded product experiments with adoption, customer benefit and full inference/development cost. The transcript mentions the initiative, but supplies no quantified Xbox return. Do not put assumed AI savings in the base plan. T1
7. Cash discipline: the separation must survive a bad release year
Owner: CFO, with independent board audit and capital-allocation oversight. Public revenue is sufficient for scale sensitivities, not for a valuation or debt package. Build the following bridge before negotiating transaction leverage:
| Bridge | Required treatment |
|---|---|
| Historical Xbox economic profit | Start with business ledgers; separate title costs, hardware economics, platform fees, royalties, cloud and support. Explain accounting policy and allocations. Never substitute More Personal Computing operating income. |
| Standalone recurring cost | Remove only parent charges for services genuinely not needed; replace necessary services with actual bids and staffing. Price insurance, audit, tax, treasury, payroll, security and employee benefits. |
| Operating repair | Quantify incremental title/distribution/Game Pass contribution and avoidable cash spend, net of overlap. Freeze baselines; prevent multiple teams claiming the same savings. |
| Sustainable reinvestment | Fund the release slate, live operations, security, maintenance capex and working capital. Include cash development investment whether expensed or capitalized. |
| Transition cash | Show severance, retention grants, duplicate systems, advisors, contractual exits, TSA charges and migration separately by month. Noncash impairments are not cash savings. |
| Owner economics | Subtract cash tax, interest and required funding. Show both Xbox and Microsoft's stranded costs; a cost left with Microsoft is not a combined-owner saving. |
Define operating free cash flow consistently as cash from operations less investing cash required for development and other capex, with an explicit reconciliation so development cash already in operations is not subtracted twice. Show recurring economics and actual cash including restructuring side by side. Existing impairment charges and lower acquisition amortization cannot be presented as proof of operating improvement. T4, F4
Cash waterfall: player/service commitments and payroll; funded content milestones and necessary maintenance; the separation reserve; then discretionary growth. Defer buybacks, acquisitions and extraordinary distributions until liquidity and operating returns are demonstrated. Preferred opening structure: no parent-funded dividend recapitalization and a conservative balance sheet. The evidence does not establish a safe debt amount.
Propose opening liquidity sufficient for the peak cumulative cash draw in an 18-month downside plan plus a board-set operating buffer; fund this with retained/allocated cash and committed facilities actually available at close. Test a six-month delay to the most cash-important release, subscription churn deterioration, a hardware shortfall and a higher Azure bill, with timing linked rather than assuming all costs instantly flex. Size scenarios from the data room; these are required tests, not forecasts.
Transparent sensitivities: at FY2026 revenue, 100 basis points of margin equals $217.9 million before tax (calculated: $21.790bn × 1%). A hypothetical 10% revenue decline at a hypothetical 60% decremental contribution rate costs $1.3074 billion before tax, before management responses. Neither is an Xbox margin estimate. [F4; arithmetic]
For a proposed efficiency package, calculate recurring net benefit = cash savings + incremental portfolio contribution − new standalone overhead − incremental external service costs − ongoing reinvestment. Separation payback uses that net annual benefit and transition cash, with actual timing and risk. If the net benefit is zero or negative, there is no positive payback to advertise.
8. The board's operating cadence and scorecard
Weekly: CFO cash/commitment review; CTO incident and failed-customer-journey review; studios production exceptions; a CEO decision log with owner and deadline. Monthly: integrated portfolio review, cohort economics, capacity commitments, savings realization and TSA exits. Quarterly: independent board re-underwrites the slate and separation readiness, with an explicit downside case.
Proposed leadership incentives: 40% sustainable consolidated cash improvement, 30% delivery and product quality, 20% retained paying-customer economics, 10% separation/control milestones. Calibrate after the baseline is established; defer a material portion across multiple years. Cash improvement receives no credit for deferring required maintenance or weakening the funded slate; material customer/control failures cap awards. These weights are a design proposal, not compensation disclosure.
| Metric / definition | Owner | Proposed decision standard |
|---|---|---|
| Forecast cash versus actual; minimum headroom under downside | CFO | Weekly review from Day 14; no unfunded contractual obligation or breach of board liquidity floor. |
| Titles with cost-to-complete, next gate, owner and risk-adjusted remaining return | Studios officer | 100% of funded projects by Day 45; exceptions explicitly approved by board. |
| Milestone and budget variance | Studios officer | Re-review at >30-day critical slip or >10% cost-to-complete increase; distinguish scope, quality and execution causes. |
| Critical customer journeys: success rate and p95 completion time, by device | CTO | Baseline by Day 30; proposed 20% reduction in failed transactions/journeys by Day 100 in prioritized flows, with no worsening of security or latency. This is a repair target, not current performance. |
| Game Pass contribution by acquisition/tenure cohort | Platform president | Reconciled by Day 60; scale experiments only on positive conservative incremental contribution and passed customer guardrails. |
| Captured contribution per migrating customer | Platform + publishing | Monthly migration analysis by Day 60; decline triggers channel/retention intervention, not an automatic exclusivity rule. |
| Cloud contribution by cohort, delivered-hour cost, queue/session quality | CTO + platform | Metered baseline by Day 60; no uncosted expansion commitment. |
| Realized cash savings net of replacement cost | CFO | Monthly actual ledger evidence; no credit for allocations, depreciation changes or unsustainable investment deferral. |
| Partner payment accuracy and certification turnaround | Platform president | Baseline by Day 30; every material overdue payment/blocked launch gets a named remedy and date. |
| Service migration, entitlements and finance controls | Separation lead | Two successful end-to-end rehearsals; zero unresolved critical continuity/control defects before legal close. |
At month 6, expect reconciled accounts, functioning gates, measured cohorts and completed high-priority customer repairs. At month 12, demand evidence from shipped releases and renewed cohorts plus an audited separation case. At month 24, assess sustained consolidated cash improvement against the pre-spin baseline after recurring standalone costs and adequate content investment. These are review points; the report supplies no invented revenue, margin or savings forecast.
9. Legal-close gates and the reasons to stop
Appoint a dedicated separation lead with a dependency register; do not make studio producers also run payroll and identity migrations. Protect key production, engine, live-ops, security and franchise personnel with targeted retention tied to delivery and handover. Consult employees and representatives as required; budget implementation obligations before recognizing savings.
Legal close requires: a completed rights/liability perimeter; required consents and approvals; counsel's transaction/tax analysis; audited carve-out accounts and opening balance sheet; funded downside liquidity; signed service contracts; tested player continuity; functioning treasury/payroll/tax/reporting controls; and no critical unresolved separation defects. The preferred legal form remains contingent on that diligence.
For migration, shadow the new systems, reconcile purchases/entitlements and monetary balances, rehearse settlement and customer support, then cut over in controlled cohorts with monitoring and rollback. Run load and failure tests ahead of launches. A failed reconciliation blocks the affected cutover; it does not become a post-close customer-support problem.
Stop or defer the spin if: standalone services cost more than demonstrable operating improvement; rights or regulatory commitments prevent the intended operating model; the funded slate cannot survive the downside; key teams or partners are lost faster than the plan can replace capacity; or Microsoft will not grant viable access/portability terms for indispensable services. Continue the operational repair inside Microsoft when that is the better economic outcome.
Reject these shortcuts: a console-only orphan; a highly levered buyout justified by unverified EBITDA; a universal day-one Game Pass rule; an immediate withdrawal from other platforms; indiscriminate studio closures; or an AI productivity assumption used to fund cash commitments. Each skips a fact the transcript evidence says we still need.
10. The data-room request that makes this executable
| Priority | Data / owner | Decision unlocked |
|---|---|---|
| First week | Monthly business ledgers, cash flows, allocations, commitments and restructuring cash — CFO | Real starting earnings/cash; opening funding; cost ownership. |
| First week | Every title's original/current budget, remaining spend, milestones, platform receipts, rights and staff capacity — studios + publishing | Continue/rescope/partner/stop decisions; release concentration and delay risk. |
| First month | Anonymized permitted subscriber/payer cohorts, realized prices, renewals, content entry, add-ons and refunds — platform + data/privacy | Game Pass policy; customer repair; causal experiment design. |
| First month | Customer migration, channel fees, third-party settlements and port economics — commerce + publishing | Distribution and exclusivity investment decisions. |
| First month | Metered infrastructure usage, shared dependencies, incident data and vendor quotes — CTO + procurement | Azure agreement; cloud expansion; TSA cost and exit plan. |
| First month | IP/entity/contract/employee/liability map, regulatory obligations and change-of-control clauses — counsel + HR | Legally achievable perimeter, timing and one-time cash costs. |
| First month | Device installed base, inventory aging, warranty/repair exposure, component and tooling commitments — devices | Support reserve, new hardware hurdle, OEM alternatives. |
The first board instruction should be: within 30 days, reconcile where Xbox earns cash, where it consumes cash, and which decisions can change that; within 100 days, put that evidence behind a funded slate and a go/no-go separation decision. Independence matters only if it changes those decisions and protects the customers and teams that make the franchise valuable.
Sources and method
The gaming passages were checked directly against all four local call files, with original line numbers and hashes saved in MSFT_carveout_sources/evidence.json. Microsoft's official transcript pages were also opened to cross-check the evidence. The prior review is context; original transcripts and filings govern factual statements. Proposed actions, organization, deadlines, tests and targets are the author's operating recommendations. No confidential internal Xbox data is available here; no standalone EBITDA, valuation, subscriber count or quantified savings forecast is asserted.
- T1: Microsoft FY2026 Q1 call, October 29, 2025. Local gaming passages: lines 43, 45–46, 65–66, 78.
- T2: Microsoft FY2026 Q2 call, January 28, 2026. Local gaming passages: lines 40, 45, 56–57, 65.
- T3: Microsoft FY2026 Q3 call, April 29, 2026. Local gaming passages: lines 39–40, 59–60, 67.
- T4: Microsoft FY2026 Q4 call, July 29, 2026. Local gaming passages: lines 40, 46, 60, 71–72. MPC margins and amortization refer to the broader segment.
- F4: Microsoft FY2026 10-K, filed July 29, 2026. Xbox revenue disaggregation; More Personal Computing discussion; Notes 8 and 9. Local revenue table: lines 7460–7463.
- Arithmetic: Revenue change = 21,790 − 23,455 = −1,665 ($m); growth = −1,665 / 23,455 = −7.0987%; 100bps = 21,790 × 0.01 = 217.9 ($m); hypothetical downside = 21,790 × 0.10 × 0.60 = 1,307.4 ($m).