What is direct-to-device connectivity worth, who will provide it, and what should decision-makers take from this study?
The short answer. direct-to-device (D2D), satellites connecting to ordinary mobile phones, has moved in four years from demonstration to a service used by millions of devices a month. Today it is a coverage product: messaging, emergency alerts and light data outdoors where there are no towers, used occasionally and priced at zero to about $10 a month. It becomes a capacity product only with very large satellite antennas and much more spectrum, and the industry is now spending heavily on both. SpaceX paid about $17 billion for 65 MHz of spectrum and in October 2026 won approval for 15,000 D2D satellites; Amazon agreed to buy Globalstar for about $11.6 billion; AST SpaceMobile has raised billions to build the largest commercial arrays in orbit; and Europe is redesigning its 2 GHz satellite band around sovereignty. Setcoin's base case puts attributable D2D revenue at about $9 billion in 2030 and $20 billion in 2035, concentrated among a few integrated providers.
Ten findings
| # | Finding | Chapter |
|---|---|---|
| 1 | Physics limits D2D: the phone's small antenna and power mean capacity depends on satellite antenna area and spectrum | Chapter 4, Chapter 9 |
| 2 | First-generation capacity is small: a few megahertz per beam yields a few megabits per second shared over tens of kilometres | Chapter 6 |
| 3 | Usage is real but light: ~7.4M monthly devices on SpaceX services; satellite traffic ~0.0002% of T-Mobile's network | Chapter 7 |
| 4 | Spectrum is the scarce asset: satellite spectrum now sells at ~$0.8 per MHz-pop, near terrestrial values | Chapter 26 |
| 5 | The industry is consolidating: EchoStar sold, Globalstar is being acquired, Lynk and Omnispace merged, Viasat and Space42 formed Equatys | Chapter 19 |
| 6 | Operators earn mainly through churn reduction and plan upgrades, not add-on fees | Chapter 28 |
| 7 | Satellites complement rather than replace towers: capacity density is orders of magnitude lower | Chapter 39 |
| 8 | SpaceX could become a mobile operator; this is the largest single uncertainty | Chapter 41 |
| 9 | WRC-27 (November 2027) and the EU 2 GHz selection (~2029) will set the international rules | Chapter 25 |
| 10 | Market structure is likely to be concentrated: in the base case SpaceX takes ~45% of revenue by 2035 | Chapter 42 |
Key numbers
| Measure | Value | Source |
|---|---|---|
| Starlink direct-to-cell satellites in orbit | 637 (663 launched) | Setcoin SATCAT analysis, 18 Oct 2026 |
| SpaceX next-generation D2D satellites authorised | 15,000 | FCC, 6 Oct 2026 |
| SpaceX monthly devices on mobile services | ~7.4M in ~30 countries | SpaceX S-1/A |
| SpaceX spectrum purchase | ~$17B for ~65 MHz | EchoStar/SpaceX |
| Amazon–Globalstar | ~$11.6B | Announced 14 Apr 2026 |
| AST SpaceMobile 2026 revenue guidance | $150–200M | AST Q2 2026 |
| AST pro forma cash | >$3.7B | AST Q2 2026 |
| Setcoin base case D2D revenue | ~$8.8B (2030); ~$19.8B (2035) | Chapter 43 |
The five interactive models
| Model | Question it answers |
|---|---|
| Link budget | Can a satellite with a given antenna close the link to an ordinary phone? (Chapter 4) |
| Beam capacity | How much capacity does a beam deliver, and how many users can share it? (Chapter 6) |
| Subscriber economics | What does D2D earn an operator and its satellite partner? (Chapter 28) |
| Capex per user | What does a constellation cost per paying user per month? (Chapter 30) |
| Spectrum value | What earnings justify a spectrum price? (Chapter 26) |
Each model can be downloaded as a spreadsheet with live formulas.
How the study is organised
| Part | Chapters | Content |
|---|---|---|
| I. Foundations | Chapter 2 to Chapter 7 | What D2D is, history, physics, protocols, capacity, usage evidence |
| II. The market and its providers | Chapter 8 to Chapter 22 | Worked beam example, market today, Starlink, SpaceX spectrum, AST and its financing, capital flows, Apple/Globalstar/Amazon, device gatekeepers, standards route, consolidation, China, defence use, 3GPP |
| III. Spectrum and regulation | Chapter 23 to Chapter 26 | US rules, interference and power limits, WRC-27, Europe and UK, spectrum value |
| IV. Economics | Chapter 27 to Chapter 33 | Wholesale models, operator economics, pricing, capex per user, ground segment, use cases, rural policy |
| V. Regions and resilience | Chapter 34 to Chapter 40 | Emergencies, maritime and remote industry, Europe, country profiles, Asia-Pacific and emerging markets, satellites vs towers, IoT |
| VI. Outlook | Chapter 41 to Chapter 49 | Starlink Mobile, forecasts and scenarios, market structure, capacity balance, supply chain, technology roadmap, three futures, sustainability, case studies |
| VII. Using the study | Chapter 50 to Chapter 55 | Questions answered, implications by reader, the case for and against, due-diligence checklist, lessons from adjacent markets, risks and method |
Appendices. A, reference table of D2D systems; B, glossary; C, Setcoin's satellite-catalogue dataset.
Reading paths
| Reader | Suggested path |
|---|---|
| Mobile operator executive | Chapters 1, 2, 28, 27, 29, 39, 41, 51 |
| Satellite or space company | Chapters 4, 6, 9, 12, 13, 26, 45, 42 |
| Government and regulator | Chapters 23, 25, 34, 36, 21, 33, 47 |
| Capital provider | Chapters 1, 26, 14, 43, 42, 53, 55 |
| Newcomer | Chapters 2, 3, 4, 8, 50 |
How this study relates to the series
This is the sixth study in Setcoin's space research programme. Study S5 (launch economics) explains the launch costs that make large constellations possible; Study S4 (defence space and Golden Dome) covers military demand for resilient communications. Facts are as of October 2026; Setcoin calculations and illustrations are labelled.