Coverage is going everywhere. The money won't follow. · Strategist edition

In April 2026, Amazon announced it was buying Globalstar. Globalstar is the satellite company Apple pays to carry Emergency SOS calls when iPhones go off-grid. With its own device ecosystem, Prime logistics network, and AWS infrastructure platform, Amazon absorbed a satellite operator and the Apple supply relationship it carries. A hyperscaler now sits one layer below one of the most consequential user experiences in consumer electronics. The strategic implications cascade upward from there.

The weighted answer:

The panel's opening position: coverage commoditizes, and in three of the four weighted futures, the margin above it migrates to device makers or vertically integrated launch operators, not to carriers. The direction is high-confidence. The timing is medium-confidence. The path is contested.

How we got here

The direct-to-device satellite technology was validated in 2020 by BlueWalker 3, AST SpaceMobile's prototype that proved a low-earth-orbit satellite with phased-array antennas could talk directly to an unmodified smartphone using standard 4G/5G spectrum. No special device, no extra hardware.

Two commercial models were born within three months of each other in late 2022. SpaceX and T-Mobile announced a partnership: Starlink satellites would use T-Mobile's mid-band licensed spectrum to deliver basic connectivity in dead zones. Satellite operator supplies coverage; carrier supplies the customer. Then Apple launched Emergency SOS via Satellite on the iPhone 14, contracted directly with Globalstar, and bundled the feature into the device at no charge. No carrier. OEM owns the experience. Two models, two structural templates, one unresolved tension.

The deeper pattern

T-Mobile, AT&T, and Verizon have all signed satellite coverage deals. Each made tactical sense at the time. The collective result is that all three are jointly commoditizing geographic coverage, the single most defensible pricing advantage US carriers have maintained for decades. It justified thirty years of tower builds, spectrum auctions, and capex cycles.

Two mechanisms lock them in. Disruptive Innovation (Christensen) identifies the profit paradox: reported metrics improve while the structural basis of pricing power degrades. And Wardley Mapping (Wardley) identifies the Red Queen Effect: once T-Mobile adds satellite backup, AT&T and Verizon must match it or lose subscribers. Once all three offer it, none earns a dollar more. The treadmill runs faster; nobody gets ahead.

The carriers cannot win this game by participating. They cannot afford to sit it out. Understanding why that is structurally locked in, not merely unfortunate, is the frame for everything else in this analysis.

The five frameworks: full reading

Wardley Mapping (Wardley): The coverage layer sits at approximately 0.45 on the evolution axis as of mid-2026, confirmed mid-Product by the absence of published commodity tariffs. The value-chain apex, Device OEM OS Integration, is the margin concentration point the framework tracks. The Innovate-Leverage-Commoditise pattern is already operating: OEMs pioneer applications above the coverage layer, leverage their installed base to capture value, and commoditize the infrastructure beneath. The Red Queen Effect and Commoditization Conflict (the tension between a component's market-pricing trajectory and any single player's desire to retain it as a differentiator) operate independently of who owns the coverage layer.

Scenario stances: Favors Fragmented Multi-Stable State (40%), pricing opacity confirms mid-Product; favors OEM Vertical Capture (28%), Innovate-Leverage-Commoditise already in operation at the apex. Opposes SpaceX Direct (spectrum dependency on T-Mobile's FCC SCS license is structural, not a business preference). Identifies the one genuine benefit of Carrier Vertical Integration: MNO capital injection could elevate the phased-array manufacturing constraint faster than AST could achieve alone.

Disruptive Innovation (Christensen): The satellite D2D technology S-curve (improvement rate over time) is in its steepest phase since 2022. Performance adequacy ("good enough for routine, non-emergency use") arrives in 2026-27. Once a technology reaches performance adequacy, Christensen's framework predicts that fragmented equilibria tend to break, value migrates upmarket to whoever controls the application relationship, and incumbents who participated in commoditizing their own differentiator are structurally exposed.

Scenario stances: Favors OEM Vertical Capture (28%) and SpaceX Direct-to-Consumer (17%) at high confidence. Gives Carrier Vertical Integration (15%) medium confidence as a predicted defensive attempt (rational but fraught). Opposes Fragmented Multi-Stable State (40%) at the structural level: the S-curve inflection tends to break fragmented equilibria, and stable fragmentation requires the inflection to stall past 2028.

Complexity Economics (Arthur): The coverage layer has no self-reinforcing adoption dynamics. Additional satellites past full geographic coverage add nothing for the buyer. The OEM layer has strong increasing returns: Apple's 100 million-plus US iPhones represent a platform where satellite capability can be delivered at near-zero marginal cost per user. Each new feature makes the ecosystem stickier and strengthens Apple's negotiating position against any satellite provider. Once coverage is interchangeable, the satellite provider needs Apple's distribution. Apple needs no specific satellite provider. Amazon's Globalstar acquisition creates a second node targeting enterprise logistics and agricultural IoT, the niches that commoditizing coverage enables.

Scenario stances: Favors Fragmented Multi-Stable State (40%), OEM Vertical Capture (28%), and SpaceX Direct (17%). Opposes Carrier Vertical Integration (15%): adding diminishing-returns infrastructure while increasing-returns competitors compound above you is structurally losing.

Modularity / Design Rules (Baldwin and Clark): The 3GPP NTN Release 17/18 standards created the visible interface (the published specification defining how modules connect) between LEO satellites and unmodified handsets. This is the event that modularized the D2D coverage market. Once a visible interface exists, the module can be substituted independently, and competition shifts to whoever sits at the architectural apex above the interface.

Modularity / Design Rules dissented from all four scenarios. Against fragmentation: the visible interface already enables faster multi-provider entry than stable fragmentation requires. Against OEM capture: two competing OEM apex nodes (Apple, Amazon) mean the apex is contested, not controlled by a single architecture owner. Against SpaceX Direct: 3GPP NR authentication requires a mobile network operator, and SpaceX cannot change that standard unilaterally. Against carrier integration: working against an already-modularized standard adds complexity without delivering architectural control.

Scenario stances: Partial dissent from all four. The dissent reflects a coherent position: the modular architecture is advancing faster than any of the four proposed end-states fully accounts for.

Throughput Accounting (Bragg): The binding constraint for AST SpaceMobile is phased-array manufacturing: the AST5000 ASIC production rate caps constellation growth regardless of spectrum or demand. SpaceX avoids this constraint through vertical integration of satellite and rocket production, giving it a durable cost-of-provision advantage. Throughput Accounting, Bragg's framework for scoring decisions by revenue minus truly variable cost against the binding constraint rather than by standard cost allocation, shows AST's throughput growth as constraint-limited through at least 2027.

Throughput Accounting also identifies two distinct margin pools: OEM above the coverage layer, efficient-scale operator (SpaceX) at the coverage layer. Carriers sit between them, squeezed on both sides. This is a partial dissent from the OEM-capture-only narrative.

Scenario stances: Favors fragmentation (manufacturing bottleneck sustains supply constraints for AST through at least 2027). Partially dissents from pure OEM capture: coverage layer retains its own margin pool. Opposes Carrier Vertical Integration at the unit-economics level: internalizing the constellation converts a variable wholesale fee into massive fixed operating expense at exactly the wrong point in the adoption curve.

The four futures: full depth

40%: Fragmented Multi-Stable State

Three parallel coverage tracks co-exist through end-2028: Starlink/T-Mobile, BlueBird/AT&T+Verizon, Amazon-Globalstar for enterprise. Wholesale pricing stays negotiated and carrier-specific. No open tariff appears. Coverage improves incrementally without completing the transition from Product to Commodity. The market enters 2029 still in mid-Product stage.

Supported by Complexity Economics (no self-reinforcing adoption threshold crossed), Throughput Accounting (AST manufacturing bottleneck persists through 2027), Wardley Mapping (mid-Product confirmed by pricing opacity).

Opposed by Disruptive Innovation (S-curve inflection will break fragmented equilibria in 2026-27); Modularity / Design Rules (visible interface enabling faster multi-provider entry than stable fragmentation requires).

The fragmentation scenario is stable, not robust. It requires the technology S-curve to stall and the manufacturing bottleneck to persist longer than the current trajectory implies.

28%: OEM Vertical Capture

Coverage commoditizes by 2027. Apple locks consumer safety and messaging above the commodity. Amazon locks enterprise logistics and IoT above the commodity through Globalstar. Satellite operators become structural analogs of tower companies: essential, thin-margin, priced as utilities. Carriers retain distribution but earn no new per-subscriber D2D margin; churn defense is the sole benefit.

Supported by Complexity Economics (increasing returns compound at OEM layer), Disruptive Innovation (Apple's Emergency SOS is the proof case; value migrated to OEM at performance adequacy, pattern repeating), Wardley Mapping (Innovate-Leverage-Commoditise pattern already in operation).

Partially opposed by Modularity / Design Rules (two-OEM apex is contested, not captured by a single architecture owner) and Throughput Accounting (efficient-scale satellite operator also retains margin at the coverage layer; this is not a complete OEM sweep).

17%: SpaceX Direct-to-Consumer Bypass

SpaceX migrates from carrier-distribution partner to direct consumer subscription once Starlink Direct achieves performance adequacy, likely 2026-27. The T-Mobile deal is renegotiated or terminated. SpaceX captures both infrastructure volume margin and the consumer relationship, collapsing two margin pools into one vertically integrated stack.

Supported by Complexity Economics (capital-subsidized threshold crossing before rivals is W. Brian Arthur's canonical playbook; SpaceX has the capital), Disruptive Innovation (up-market migration from a distribution foothold is the standard pattern), Throughput Accounting (near-zero variable cost per subscriber on a direct model means higher throughput than a revenue-share wholesale arrangement).

Opposed by Wardley Mapping (SpaceX's spectrum access depends on T-Mobile's FCC SCS license; exiting the partnership risks losing the spectrum that makes the service function on unmodified phones) and Modularity / Design Rules (3GPP NR authentication requires a mobile network operator; the interface standard prohibits carrier bypass without an international standards change SpaceX cannot engineer unilaterally).

15%: Carrier Vertical Integration

A major US MNO, most likely AT&T or Verizon, acquires controlling equity in AST SpaceMobile before capital-cycle distress. Coverage re-integrates into a proprietary carrier stack. Margin captured as churn reduction and premium rural plan pricing, not standalone D2D revenue.

Favors: Disruptive Innovation at medium confidence: the reading predicts one MNO will attempt this as a rational defensive move, even knowing execution is fraught.

Against every other lens. Complexity Economics: adding diminishing-returns infrastructure while increasing-returns competitors compound above you is structurally losing. Wardley Mapping: the Red Queen and Commoditization Conflict operate independently of ownership; buying the layer does not stop it from commoditizing. Throughput Accounting: internalizing the constellation converts a variable wholesale fee into massive fixed operating expense at exactly the wrong point in the adoption curve. Modularity / Design Rules: working against an already-modularized standard adds complexity without delivering architectural control.

The one genuine benefit: MNO capital injection could elevate the phased-array manufacturing constraint faster than AST could achieve alone. This is the only structural improvement Wardley Mapping identifies for this scenario.

Milestones and weight shifts

How the weights shift on milestone outcomes:

• Apple expands satellite by Sep 2026 (routine messaging or data): OEM Vertical Capture (28%) rises sharply, likely above 40%. Fragmentation drops. The proof case becomes undeniable.
• SpaceX renegotiates T-Mobile or launches direct D2D by Oct 2026: SpaceX Direct (17%) rises. The spectrum dependency argument weakens. Fragmentation and OEM capture both shift.
• MNO acquires AST by Oct 2026: Carrier Vertical Integration (15%) rises toward 30%. Other scenarios compress. The manufacturing constraint narrative shifts if MNO capital elevates it.
• First open D2D tariff published by Jan 2027: Fragmentation (40%) drops significantly. OEM capture and SpaceX Direct both gain. This is the single clearest commodity formation signal.
• AST Block 2 hits 20+ CONUS satellites by Jun 2027: Manufacturing bottleneck narrative weakens. Fragmentation loses its primary structural support. OEM capture and SpaceX Direct both benefit.
• ARPU compression confirmed in MNO earnings by Dec 2027: All non-carrier scenarios gain. Carrier Vertical Integration urgency increases (defensive rationale strengthens), but so does the opposing argument (commoditizing faster than expected).

The map

The value chain runs from deep infrastructure at the bottom (launch services, now effectively commodity) through spectrum licensing and phased-array manufacturing, through the coverage radio link, through carrier billing and distribution, through OEM device integration, and up to the emergency and consumer applications users actually interact with. Visibility increases going up the axis; evolution runs from Genesis (left, custom-built) to Commodity (right, interchangeable).

Component movements by scenario:

Scorecard

This is the panel's first analysis of this topic. The scorecard shows zero resolved calls across all five lenses. No hit rates, calibration scores, or predictive edge metrics are defined yet.

All five lenses have open exposure across all four scenarios. No call has resolved. These are opening positions.

What this means

The coverage race is not really about coverage. It is about who controls the user relationship when connectivity becomes universal, and that transition is under way regardless of whether 2028 closes the commoditization window.

For carriers: The structural position is uncomfortable in a precise way. Each satellite deal improves reported metrics today and erodes pricing power over time. The Red Queen treadmill has no exit: opting out loses subscribers in the short term; opting in accelerates the commoditization that costs margin over the long term. A carrier that acquires AST SpaceMobile buys the manufacturing constraint problem along with the asset, internalizes a variable cost as fixed capex, and still faces the same modularized interface standard. The honest summary: carriers are funding the infrastructure layer that will compress their own margins.

For builders above the coverage layer: Three of the five frameworks point to the same window. Emergency and safety applications are already staked by Apple. Agricultural IoT, maritime monitoring, industrial asset tracking, and cross-border logistics are the next tier. Cheap, universal coverage is the substrate; the applications that run on top of it are where the new margin will sit. This substrate is being built at enormous scale, mostly with other people's capital.

For investors: The coverage layer itself is a diminishing-returns infrastructure play. The strategic question is whether you can identify the application layer before coverage commoditization is complete. SpaceX's integrated model gives it the best unit economics at the coverage layer, but the consumer relationship play (17%) requires a regulatory and contractual maneuver that carries real structural risk. The OEM layer (Apple, Amazon) is where increasing-returns compounding happens, but those positions are already staked.

For strategists evaluating the milestones: The September 2026 Apple checkpoint is the single most informative near-term data point. If Apple expands satellite features beyond emergency SOS by that date, OEM Vertical Capture (28%) is likely the leading scenario by year-end, and fragmentation (40%) compresses substantially. The January 2027 first-open-tariff checkpoint is the clearest commodity formation signal. Watch those two.

The coverage layer is becoming plumbing. The margin is moving up, and the window to position above it is measured in months, not years.