Flagship case studyPhase 1 of 18 · Discovery

Hospital OS · Enterprise SaaS

Designing a unified operating system for hospitals.

A modular, cloud-native platform that unifies clinical, operational, financial and administrative workflows into a single connected ecosystem — reducing fragmentation for hospitals operating in the Indian healthcare market.

Role
Product Designer
Scope
Discovery research · Product strategy · Systems design
Status
Phase 1 of 18 · Discovery
Year
2026

This is a self-initiated project grounded in real Indian healthcare-industry research, regulatory analysis and public competitor documentation. It is not presented as commissioned client work.

11

operational domains

from Patient Access through Administration & Compliance — the architectural pillars the platform is organised around.

42

departments

coordinated across 4 functional blocks: front-of-house, clinical core, ancillary/diagnostic, and back-office.

443

workflows mapped

each one specified down to its trigger, actors and system integrations before any interface was considered.

19

vendors researched

spanning enterprise EHR, India-focused HIS/EMR, and specialised point solutions (billing, patient engagement, RCM).

01 / Problem

Hospitals run on a patchwork of disconnected systems.

Modern hospitals rely on separate software for appointments, records, billing, pharmacy, labs, inventory, HR and CRM. As hospitals grow, these specialised applications rarely operate as one cohesive system.

Staff spend significant time coordinating between systems instead of delivering care, duplicate data entry is routine, and leadership lacks a single view of operations across departments.

Core design question

How might we design a unified Hospital Operating System that simplifies healthcare operations while supporting the needs of every department through a single, scalable platform?

02 / Context

An Indian hospital is a portfolio of tightly coupled businesses, not a single building.

Before designing any screen, the research grounded the platform in how Indian hospitals actually operate as regulated, financially complex enterprises.

A layered market, not one buyer

Government/public hospitals (AIIMS, district hospitals), corporate chains (Apollo, Max, Fortis), trust and not-for-profit hospitals, private standalone facilities, and single-specialty day-care clinics all operate under different economics and constraints.

Labour is the largest cost line

Salaries and benefits account for roughly 43–56% of hospital revenue, which is why bed occupancy and staff utilisation — not just software features — drive whether a hospital survives.

Three payer types, one patient

A single hospital concurrently manages out-of-pocket cash patients, private insurance (TPA) claims with strict pre-authorization workflows, and government schemes like PM-JAY at standardised, discounted rates.

Compliance is infrastructure, not a checkbox

ABDM/ABHA interoperability mandates, the DPDP Act's consent and access-control requirements, NMC's biometric attendance rules, and IRDAI's 1-hour cashless pre-authorization timeline all shape what the platform must guarantee, not just offer.

03 / System model

Reverse-engineering the hospital before designing a single screen.

Rather than starting from software features, the research mapped the hospital itself: every domain, department, workflow, actor and dependency — the foundation a modular platform has to be built on.

Select a domain to see its operations, actors and dependencies

11 operational domains, decomposed further

Each domain breaks down into departments (e.g. Patient Access → Registration, Scheduling, Admission Desk).

A hospital coordinates roughly 42 departments across these 11 domains — mapping them first is what makes a modular platform possible.

04 / Workflow depth

Domains and departments are the map. 443 workflows are the terrain.

Every one of the 42 departments was decomposed into its actual workflows — each specified with its trigger, its actors and the systems it has to talk to. Four examples, spanning very different parts of the hospital:

Workflow 058Emergency & Trauma Services

Emergency Triage

Rapidly assesses an arriving patient's condition, assigns an evidence-based urgency category, and routes them to the right treatment area within minutes.

Walk-in patientAmbulance arrivalMass casualty incident

Actors

Triage Nurse, Emergency Physician, Trauma Team

Workflow 100Outpatient & Inpatient Care

Clinical Decision Support

Analyses clinical data in real time and delivers evidence-based, patient-specific guidance to clinicians — while keeping the clinician the final decision-maker.

Medication orderingLab result availabilityClinical deterioration

Actors

Physician, Nurse, Pharmacist

Workflow 108Inpatient Care & Ward Management

Inpatient Discharge Management

Governs the safe transition out of inpatient care — clinical sign-off, medication handover, patient education and encounter closure — so no step gets skipped to save time.

Physician authorises dischargeTransfer to another facility approvedDischarge against medical advice

Actors

Attending Physician, Discharge Nurse, Case Manager, Billing Team

Workflow 224Financial Operations & Revenue Cycle

Denial Management

Investigates, corrects and appeals insurance claims a payer has denied after adjudication — the workflow that decides whether the hospital actually gets paid for care already delivered.

Medical necessity denialAuthorization denialCoding denial

Actors

Denial Management Specialist, Medical Coder, Appeals Coordinator

05 / Competitive research

Understanding what already exists before proposing something new.

19 hospital and healthcare software vendors were researched across three tiers — global enterprise EHR, India-focused HIS/EMR, and specialised point solutions — deliberately before evaluating any interface.

Enterprise EHR / HIS

  • Epic Systems
  • Oracle Health (formerly Cerner)
  • athenaOne
  • NextGen Enterprise
  • SAP S/4HANA Healthcare

India-focused HIS / EMR

  • Akhil Systems – Miracle HIS
  • Attune HIS
  • Ezovion HIMS
  • KareXpert HIMS
  • MocDoc HMS
  • Practo Ray
  • HealthPlix EMR
  • Cliniqwise
  • CrelioHealth LIMS & RIS

Specialised point solutions

  • Availity Essentials
  • LeadSquared Health (CRM)
  • Luma Health
  • Phreesia
  • Waystar RCM / AltitudeAI
  • Hayati AI Nexus

Questions asked of every vendor — deliberately not about UI

What modules does this vendor actually provide?

What real operational problems do they solve?

What do they explicitly ignore or leave to other vendors?

What integrations does adopting them require?

06 / Product scope & principles

A modular ecosystem, not a monolith with everything bolted on.

Clinical Operations

Patient registration, EMR, appointments, consultation, admissions, discharge, nursing workflows.

Diagnostic Services

Laboratory, radiology, report management.

Pharmacy

Prescription management, drug inventory, dispensing.

Business Operations

Billing, insurance, CRM, sales pipeline, customer support.

Hospital Administration

HR, attendance, shift scheduling, asset management, procurement.

Patient Experience

Mobile portal, appointment booking, digital consent, payments, notifications.

Design principles guiding every later phase

  • One source of truth — data is entered once and accessible everywhere.
  • Role-based by default — interfaces adapt to the specific needs of the user.
  • Automation over repetition — the system handles handoffs and notifications.
  • Configurable, not hardcoded — workflows adapt to the hospital, not the reverse.
  • Modular before monolithic — built to scale and be adopted incrementally.
  • Secure by design — privacy and compliance built into the foundation.

07 / Where this stands

This follows an 18-phase process. I am at the end of Phase 1.

I'd rather show exactly what's done and what's next than round the project up to a finished product.

Phase 1 — Discovery

1.1Design Challenge Brief
Done
1.2Healthcare Ecosystem Overview (India)
Done
1.3Hospital Operations Analysis (11 domains)
Done
1.4Competitive Landscape (19 vendors)
Done
1.5Stakeholder Inventory
Next
1.6Problem Inventory
Next
1.7Affinity Mapping
Next
1.8Root Cause Analysis
Next
1.9Opportunity Mapping
Next
1.10Final Problem Statement
Done

Phases 2–18 · Ahead

Stakeholder & User ResearchCurrent System AnalysisBusiness Goals & Product VisionInformation ArchitectureService BlueprintUser Journey MappingFeature PrioritizationWorkflow DesignSystem ArchitectureInteraction DesignDesign SystemPrototypeUsability TestingIterationFinal Portfolio Case Study

Ahead of the process — an early prototype

The 18-phase process above is still in discovery, but I built Nadi — an authenticated, deployed OPD workspace — to pressure- test whether the Spine design system holds up once it’s wired to real routes and real auth, rather than waiting for Phase 15 to find out.

Sign in to Nadi ↗

Continue exploring

See more systems, products and design thinking.

Back to selected work ↗