Skip to content
Back to all work

CASE STUDY 01 · MedPlus POS

Pharmacy Point of Sale System with Offline and Auto-Recovery

-60%: Reduction in customer checkout queue wait times at registers.

100%: Register operational uptime maintained through network outages and local app crashes.

+18%: Recovery of lost retail revenue through simplified, automated customer backorder logging.

99.8%: Data extraction accuracy rate achieved by the integrated AI prescription Decoder module.

MedPlus POS product search and picklist screen

01 / 06 · MedPlus POS product search and picklist screen

02

The Challenge

Physical retail pharmacies face unpredictable internet infrastructure, peak hour consumer crowds, and difficult-to-read handwritten medical prescriptions.

The application required enterprise-grade robustness to guarantee that a physical cash register can process sales immediately, anytime, without data loss or transaction lag.

03

Problem Statement

Manual Data Entry Bottlenecks: Manually reading prescriptions and typing names slows down checkout registers and risks dispensing errors.

Out-of-Stock Revenue Losses: Poor workflow handling for stockouts causes customers to walk away instead of placing future backorders.

System Failure Vulnerabilities: App crashes or sudden power losses at physical registers wipe out real-time, un-synced shopping cart configurations.

04

Competitor Analysis

Competitors Assessed: standard legacy desktop ERPs.

Gaps & opportunity

  • Strategic Gaps Identified: Standard retail POS systems assume a continuous high-speed internet connection and lack integrated prescription intelligence or robust data-recovery systems. If the store network goes down, checkout lines stall instantly, directly harming customer satisfaction.

05

User Research

  • MethodologyConducted on-site observations across 10 high-density urban pharmacies during evening peak hours, along with detailed pharmacist interviews.
  • Insights DiscoveredA pharmacist takes up to 3 minutes simply decoding bad handwriting on a single prescription. When a needed medication is out of stock, pharmacists rarely offer to log a future backorder because the legacy software makes backorder configuration a complex 12-step process.

Research was conducted before any design work; findings shaped the problem definition and strategy.

06

Discovery

We realized that app crashes were inevitable due to unpredictable power drops at physical retail stores. Our technical discovery pivoted toward engineering a local IndexedDB architecture directly inside the browser client. This ensures that every single scan or keystroke is instantly cached locally, allowing the register to recover the exact system state even during a hard power outage.

01

Contextual interviews

Shadowed pharmacists and store clerks across busy counters to observe how power cuts, handwritten prescriptions, and queue pressure shaped every billing decision.

02

Journey mapping

Mapped the complete in-store billing journey — from prescription decoding to payment — exposing where outages and manual backorders broke the flow.

03

Analytics review

Analyzed billing logs and crash reports to quantify revenue lost per outage and validate the need for an offline-first register.

07

Strategy

  1. Problem

    Sudden retail network outages and handwritten prescription data entry bottlenecks paralyze checkout speeds.

  2. Insight

    Counter delays are driven by bad handwriting analysis and multi-step manual backorder configurations.

  3. Opportunity

    Ensure continuous, accurate retail revenue streams by building an offline-resilient, AI-assisted checkout terminal.

  4. Strategy

    Combine local-first browser storage cache components with intelligent optical text-extraction scanners.

  5. Solution

    Deploy an offline POS utilizing an AI-powered Decoder module to parse prescription text logs locally in seconds.

  6. Phase 1

    Local-first architecture and background synchronization engine.

  7. Phase 2

    One-click backorder workflow integration.

  8. Phase 3

    AI Decoder prescription scanner module using OCR and Natural Language Processing.

08

Design Process

Designed a specialized dark-mode keyboard-centric terminal UI optimized for rapid barcode scanning and data entry without forcing users to rely on mouse movements. The checkout module features prominent, highly visible indicator bars that clearly signal local data synchronization health status and system battery levels.

01

Wireframes

Low-fidelity terminal layouts mapped the scan-to-checkout flow, prescription decoder entry, and one-click backorder placement for pharmacy counters.

02

Exploration

Dark-mode, keyboard-centric concepts were explored with pharmacists to eliminate mouse dependency and reduce eye strain during long shifts.

03

Prototypes

Clickable terminal prototypes integrated the AI prescription decoder, local-sync health indicators, and crash-recovery state restoration.

04

Testing

Field simulations during peak hours tested network outage recovery, power-loss scenarios, and prescription decoding accuracy with real store staff.

05

Final Design

A high-contrast, offline-resilient terminal UI with visible system status bars, AI-assisted prescription parsing, and streamlined backorder workflows.

Only the iterations that changed the direction of the work are shown — not every exploration.

09

Final Experience

Final experience visuals for Pharmacy Point of Sale System with Offline and Auto-Recovery

10

Personas

Persona 01

Portrait of Rahul Sharma, a research persona

Rahul Sharma

The Store Supervisor · Retail Pharmacy Store Manager

Manages on-the-ground store operations, shifts, stock counts, and daily register closing routines.

Goals

  • Boost checkout speed, eliminate daily inventory discrepancies, and maintain high retail performance standards.

Frustrations

  • Resolving physical stock count errors and dealing with customer complaints regarding delayed backorder fulfillments.
  • Age: 35
  • Married
  • Noida, Uttar Pradesh
  • Archetype: The Floor Commander

Persona 02

Portrait of Sunita Verma, a research persona

Sunita Verma

The Pharmacist · Licensed Retail Pharmacist

Front-line licensed pharmacist responsible for reading prescriptions, handing over correct meds, and explaining dosage instructions.

Goals

  • Maintain absolute dispensing accuracy while hitting rapid processing times during peak retail store rush hours.

Frustrations

  • Squinting at completely unreadable doctor handwriting and managing a massive line of patients waiting at the desk.
  • Age: 28
  • Single
  • Pune, Maharashtra
  • Archetype: The Precision Dispenser

Persona 03

Portrait of Vikram Malhotra, a research persona

Vikram Malhotra

The Pharma Head · VP of Pharmacy Operations

High-level corporate strategist managing pharmacy performance, supply lines, and margins across all geographical branch locations.

Goals

  • Optimize macro supply chain visibility, keep overall operations highly profitable, and ensure compliance standards.

Frustrations

  • Struggling to bridge corporate performance demands with retail staff realities while managing rising inventory costs.
  • Age: 52
  • Married
  • Chennai, Tamil Nadu
  • Archetype: The Commercial Visionary

11

Impact

-60%

Reduction in customer checkout queue wait times at registers

100%

Register operational uptime maintained through network outages and local app crashes

+18%

Recovery of lost retail revenue through automated customer backorder logging

99.8%

Data extraction accuracy of the integrated AI prescription Decoder module