法國擁有歐洲數位化程度最高的基層醫療（Primary Care）預約基礎設施，但目前仍缺乏任何觀測工具，能夠捕捉就醫意圖（care intent）──亦即病患在成功完成預約之前，試圖從醫療體系獲得何種醫療服務或協助。
現有資料來源，包括 SNDS、Doctolib、IQVIA、Cegedim、Réseau Sentinelles 以及 OSCOUR／SI-DEP，皆只能觀測醫療流程下游（downstream）的最終結果事件，例如完成預約、接受診療或其他已發生的醫療服務；然而，沒有任何系統能夠掌握上游（upstream）的需求訊號。事實上，法國的醫療遠距秘書服務（medical telesecretariats）每月處理約五千萬通來電，其中估計有三千五百萬至四千萬通最終形成的處理結果，並未被任何現有基礎設施記錄。
SILLAGE 是全球首個將這些電話流量轉換為**結構化上游就醫意圖情報（structured upstream-intent intelligence）**的商業化基礎設施，開創了一種目前市場上尚無任何業者能夠提供的全新醫療資料類別。此類資料正是四大機構型客戶族群所迫切需要的資訊資產，並可在 36 個月的營運期間內持續累積，建立六項競爭護城河（six-moat defensibility），形成難以複製的市場優勢。
在財務表現方面，SILLAGE 預計於第三年達成 430 萬歐元營收，EBITDA 利潤率達 65%；整體發展僅依靠 22 萬至 28 萬歐元的非股權稀釋（non-dilutive）、自籌資金（bootstrapped financing）即可完成，並自第一年起實現獲利。

France runs the most digitised primary-care booking infrastructure in Europe, yet no observational instrument captures care intent — what patients try to obtain from the healthcare system before any successful booking is made. Existing data sources (SNDS, Doctolib, IQVIA, Cegedim, Réseau Sentinelles, OSCOUR/SI-DEP) each observe a downstream resolution event; none observes the upstream demand signal, even though French medical telesecretariats process approximately fifty million calls per month, of which an estimated thirty-five to forty million resolve into outcomes no existing infrastructure records. SILLAGE is the first commercial infrastructure to convert this call traffic into structured upstream-intent intelligence — a category of healthcare data that no existing actor records, that four institutional buyer segments need, and that compounds into a six-moat defensibility position over a 36-month operating horizon — anchored by €4.3M Year-3 revenue at a 65% EBITDA margin, achieved on €220–280K of non-dilutive bootstrapped financing and profitable from Year 1.

[1] Snds.gouv.fr, Système National des Données de Santé, Guide d'utilisation, accessed January 2026. Coverage: reimbursed acts (ville + hôpital), 12–24 month latency, infra-departmental geographic resolution for authorised users.
[2] Doctolib, Corporate communications and platform documentation, accessed April 2026. Approximately 50% of liberal physicians, one third of public hospitals; Assistant téléphonique product launched 2 December 2025 at €99/month per practitioner.
[3] IQVIA, Real World Evidence solutions overview and PharMetrics Plus dataset characterisation; "Pharma 2026: How Digital Transformation and AI Will Redefine Drug Safety," February 2026; "Nine for 2026" global webinar series. 1.2 billion+ de-identified patient lives across claims, EMR, hospital, lab and specialty pharmacy data.
[4] Réseau Sentinelles, Sentiweb, Sorbonne Université / Inserm, accessed February 2026. Sentinel-physician network for primary-care surveillance.
[5] Santé Publique France, OSCOUR (Organisation de la Surveillance Coordonnée des URgences) and SI-DEP (Système d'Information de DEPistage), public documentation 2024–2025.
[6] French medical inbound-call volume — estimates derived from public data. Base sources: OECD, Health at a Glance (5.5 physician consultations per capita per year, 2022); Assurance Maladie (CNAM), liberal-physician activity and general-practice consultation volumes (>1 million GP consultations per day; 106.6 million general-practice consultations in H1 2022); Conseil National de l'Ordre des Médecins (CNOM), Atlas de la démographie médicale 2024 (~233,000 active physicians; ~100,100 in exclusive liberal practice). Telesecretariat outsourcing rate among physicians ≈ 38% in 2022 (up from 30% in 2019). Method: ~380 million consultations/year × 3–5 inbound calls per booked consultation ≈ 1.1–1.9 billion inbound calls/year to French medical practices; the ~38% routed through outsourced telesecretariats represents on the order of 40–60 million calls/month.
[7] Conseil d'État, decision of 13 February 2026, Reinforcement of the re-identifiability standard for health data processors. Summaries via Covington Inside Privacy and Lexology, February–March 2026.
[8] About.doctolib.fr, Press release, "Doctolib lance un Assistant téléphonique fondé sur l'IA capable de dialoguer naturellement avec les patients," 2 December 2025.
[9] Egora "Doctolib lance un assistant IA pour automatiser la prise de rendez-vous par téléphone," November 2025. Pricing details: €99/month for GPs and dentists, €49/month for osteopaths/podologists.
[10] Salje, H., Tran Kiem, C., Lefrancq, N. et al. "Estimating the burden of SARS-CoV-2 in France," Science 369, 208–211, 2020. Reference for the timing of COVID-19 community-transmission detection in France, published by the Institut Pasteur.
[11] Internal bottom-up SOM analysis, named-prospect portfolio across Pharma RWE, AI training data, ARS, and State-tier segments. Internal commercial document, not externally distributed.
[12] Code de la Santé Publique, Article L.1110-4, Confidentiality of health data and conditions for processing; Législation française, accessed January 2026. Article L.1111-8, Code de la Santé Publique, statutory basis of the Hébergeur de Données de Santé certification regime; Ministère de la Santé documentation on the HDS certification regime.
[13] OVHcloud, Healthcare / HDS certification documentation (ovhcloud.com/fr/healthcare/, ovhcloud.com/fr/compliance/hds/), HDS certification since 2019, ISO 27001 / 27701 / CSA STAR / SOC II Type 2 coverage. Outscale and Scaleway HDS certification documentation considered substitutable French HDS providers.
[14] DREES, Direction de la Recherche, des Études, de l'Évaluation et des Statistiques, Ministère de la Santé. Reports on specialist access wait times in France, 2024–2025 editions. Used as the structural reference for the dermatology lighthouse use case.
[15] Cegedim, Real-world data and pharmaceutical panel documentation, accessed April 2026. Comparable for Pharma Strategic-tier price band benchmarking.
[16] DGOS / Ministère de la Santé, Fonds d'Intervention Régional (FIR), allocation framework and budget envelope (~€4.3–4.5B annually across the eighteen ARS), accessed February 2026.
[17] France 2030, Plan d'investissement, €54B total envelope, of which €7.5B for the healthcare innovation sub-programme (Santé Numérique). Documentation accessed February 2026.
[18] EU AI Act, Regulation (EU) 2024/1689, Annex IV technical documentation requirements for high-risk AI systems, enforceable from 2 August 2026. Substantiates the provenance premium on the Conversational Medical French Corpus pricing.
[19] Grand View Research / MarketsandMarkets, Healthcare AI training datasets market sizing (USD 520M in 2025, projected USD 4.1B by 2035, CAGR 22.9%). Used as the structural growth baseline for the AI Training Data Licensing component.
[20] Cafétech, Usine-Digitale, Glass Health, Coverage of Nabla Inc. and the French medical voice-AI ecosystem, December 2025 to February 2026. 85,000+ users, $120M raised, 130+ health systems, advisors include Yann LeCun and Tony Fadell.
[21] Kim, W. C. and Mauborgne, R. Blue Ocean Strategy: How to Create Uncontested Market Space and Make the Competition Irrelevant, Harvard Business Review Press, expanded edition 2015.
[22] Touchent and de la Clergerie (Inria / Sorbonne) CamemBERT-bio: Leveraging Continual Pre-training for Cost-Effective Models on French Biomedical Data, arXiv:2306.15550, 2023–2024.
[23] Labrak, Bazoge, Dufour et al. (LIA / Université d'Avignon) DrBERT: A Robust Pre-trained Model in French for Biomedical and Clinical Domains, ACL 2023.
[24] Barney, J. "Firm Resources and Sustained Competitive Advantage," Journal of Management 17(1), 99–120, 1991. Foundational article for the VRIO framework and the Resource-Based View of the firm. Used as the analytical reference for the assessment in Chapter 4.
[25] Whyte, A., with Dunkel, D. and Napoli, J. MEDDICC: The Ultimate Guide to Staying One Step Ahead in the Complex Sale, 2020 (ISBN 978-1-8382397-0-1). Canonical reference for the MEDDIC qualification framework, originally created at PTC (Parametric Technology Corporation) in 1996 by Dick Dunkel and Jack Napoli. Applied in Chapter 8 across the three buyer-side tracks.
[26] Kano, N., Seraku, N., Takahashi, F. and Tsuji, S. "Attractive Quality and Must-be Quality," Journal of the Japanese Society for Quality Control 14(2), 39–48, 1984. Foundational article for the Kano model of customer satisfaction. Applied in Chapter 5 to the three-track product roadmap with track-specific personas.
[27] PR Newswire, “Nabla Raises $70M Series C to Deliver Agentic AI to the Heart of Clinical Workflows, Bringing Total Funding to $120M,” June 2025; STAT News, “Nabla raises $70 million as ambient scribe market heats up,” 17 June 2025. Accessed June 2026.
[28] CNIL enforcement against IQVIA's French subsidiary, €5M sanction concerning officinal (pharmacy) data processing authorisations, May 2026. Press coverage: Le Moniteur des Pharmacies; Next.ink. Accessed June 2026.
[29] MarketIntelo / DataIntelo, Dataset Licensing for AI Training — Market Research Report: market sized at $4.8B (2025), projected $5.7B (2026), 18.8% CAGR to $22.6B (2034). Accessed June 2026.