GO BACK

From Monolithic Chaos to a Fully Automated Lead-to-Alumni Lifecycle

Overview

A leading private university had been running its entire operation on a single, monolithic database where applicants and enrolled students shared the same environment with no separation, no lifecycle logic, and no enforcement of university policies. The Admissions team struggled to track lead conversions. The Registrar was overwhelmed with manual processes. Grading was managed manually, and document management, internally referred to as the "honeybee hive," had become an unstructured collection of files with little organization or context. The University needed a complete transformation: a modern University Management System built on KodeFast, replacing the monolithic setup with a structured Lead-to-Alumni lifecycle platform. The objective was clear: ensure data integrity, implement role-based access controls, automate critical workflows, and establish a clear separation between Applicants (Leads) and Enrolled Students.

Challenges

No Lifecycle Separation

  • Applicants and enrolled students lived in the same database. There was no concept of a Lead graduating into a student: they were all just 'managed students,' making it impossible to track conversion rates or clean up dead records
  • The Admissions team couldn't run document verification workflows without touching live student data, and accreditation reports required manually pulling from a system that wasn't designed to generate them

Unstructured Document Storage

  • All files: passports, MCAT scores, letters of recommendation, clinical documents: were dumped into a central repository with no structure. Finding a specific file meant manually searching through thousands of records 
  • Multi-file uploads weren't supported, so staff had to merge PDFs externally or route documents over personal email, leaving no traceable record inside the system

Manual, Unverifiable Grading

  • The system stored scores but never calculated them. Whether a final grade correctly reflected the configured weighting of Block Exams, Mid-Modules, and Internal Assessments depended entirely on human math: with no audit trail
  • There was no automated enforcement of the university's 80% attendance requirement for exam eligibility. Compliance was checked manually and inconsistently

Disconnected Workflows & Manual Administration

  • Leave of Absence requests and NBME registrations were handled via external Google Forms: no status tracking, no automated approvals, no reminders, and no record inside the portal
  • Administrative memos required staff to manually copy-paste content from emails into the system, creating unnecessary friction and a high risk of error
  • There was no role-based access control. Faculty, students, preceptors, and admissions staff all operated without clearly defined permission boundaries, creating both security gaps and data privacy concerns

Solution Approach

  • Six-Role RBAC Hierarchy
      • Designed and implemented a structured six-tier permission model: Owner (full system control), Super Admin/Registrar (user management and override capabilities), Faculty Admin (department-level academic oversight), Faculty/Professor (gradebook and attendance), Preceptor (clinical evaluation only), and Student (read-only grades and write access to specific forms)
      • Each role maps directly to the university's actual operational structure, ensuring that a student cannot access the admin panel, a preceptor cannot see unrelated course data, and the Registrar has override tools without needing developer support
  • Admissions & CRM: The Leads Engine
      • Architected a strict separation between the Leads Master (applicants) and the Students Master (enrolled), treating them as fundamentally different data entities with different fields, workflows, and access rules
      • Built an API connection from the public website to auto-create Lead records on inquiry submission, feeding directly into an admissions pipeline with document upload, interview scheduling, and committee voting: all inside the CRM
      • The Conversion Event is the centerpiece: when Admissions marks a lead as 'Accepted' and uploads the acceptance letter, the system automatically migrates the record to Students Master and provisions a university email address. No manual data re-entry, no duplicate records
  • Student Information System (SIS)
      • Built a self-service student portal covering personal, immigration, emergency contact, and academic profile management: with a profile verification step required before registration can proceed
      • Implemented the Financial Gatekeeper: registration forms are hidden entirely for any student whose financial status is not marked 'Cleared' by admin. A 'Contact Finance' alert is displayed in its place, enforcing tuition compliance without any manual gating
      • Added an Auto-Progression Engine that suggests the next logical course or semester (e.g., MD1 → MD2) based on the student's current stage, reducing registration errors and advisor overhead
      • Replaced the central document hive with Contextual Storage on AWS S3: every file is anchored to the record it belongs to, with support for up to 10 uploads per field. Clinical documents live inside the rotation record. Admissions documents live inside the application. Nothing is floating loose
  • Academic Management & Gradebook Engine
      • Built a seven-phase grading engine covering assessment setup, weightage configuration, score entry, curve application, rounding, submission, and data management: each phase enforced by system logic, not by trust
      • Professors define percentage weightages per assessment type (Block 1, 2, 3, Mid-Module, Internal Assessment). The system validates that combined weightages never exceed 100% before locking the configuration 
      • Weighted scores are calculated automatically using the formula: (Student Score / 100) × Assessment Weightage. Curve marks (up to 10 points) are additive, and Standard Round Half Up rounding is applied to reach the final integer grade
      • Grade thresholds are enforced system-wide: A (90to100), B (80to89), C (70to79), F (<69). Any student receiving an F triggers an orange Add (+) icon for the professor to enter a remedial score, which re-runs the full calculation chain automatically
      • 'Submit to Registrar' creates a hard lock: no edits to types, weightages, or scores are possible after submission. Grades are only published to students once the Registrar completes their review, preserving the approval chain
      • Attendance compliance is handled separately: professors upload four-week attendance and promotion lists, and the system auto-calculates each student's percentage, flagging anyone below the 80% threshold without any manual checking
  • Forms, Workflows & Automations
      • LOA Workflow: Replaced the Google Form with a native multi-step approval chain: Registrar reviews, then Dean of Basic or Clinical Science approves. Leave days are auto-calculated, and the system sends automated reminders five days before a student's scheduled return
      • NBME Registration: A native form with a daily summary email sent to Admin listing all new registrations: replacing the previous Google Form and manual tracking process
      • Memos Parser: An inbound email parser that converts incoming emails into Draft Memos. Admin categorizes and publishes in a single click: no copy-pasting
      • Clinical Contract Form: Time-bound (active only in December and June) and automatically visible only to MD4 students, preventing form clutter for everyone else
      • Student Dormitory Agreement: A digital form replacing paper-based housing agreements, allowing students to review terms, select preferences, and submit a legally binding agreement through the portal
      • Instructor Evaluation Form: Student-facing, with dynamic course filtering: selecting an instructor automatically shows only the courses they taught in the current semester
      • All form submissions use server-side timestamps for digital signatures. Users cannot manually type or alter submission dates, ensuring a forensic record
  • Reporting & Accreditation
      • Built standard accreditation reports covering Census data (student count, gender/ethnicity breakdown), USMLE Pass Rates, Clinical Weeks Completed, and Instructor Performance: all generated automatically from live data
      • Survey and evaluation data feeds into an integrated analytics engine that generates histograms and pie charts for administrative review, removing the manual effort of compiling accreditation submissions
      • Historical data from legacy 'Track Management' and 'Exam' modules was migrated to read-only, searchable archive tables: preserving institutional history without rebuilding functionality that's no longer needed

Results

Data Integrity Across the Full Lifecycle

  • The separation of Leads and Students eliminated the dead-record problem and gave both Admissions and the Registrar a database they could actually trust for reporting and accreditation submissions
  • Server-side timestamps, hard-locked grade submissions, and a full audit trail on every role action mean that every record is forensic: traceable, timestamped, and tamper-proof

Operational Efficiency at Every Level

  • The Financial Gatekeeper, Auto-Progression Engine, LOA Workflow, and NBME form collectively eliminated the largest manual bottlenecks in the Registrar's and Admissions team's daily operations
  • The seven-phase grading engine replaced a process that was previously manual and unverifiable. Grades are now calculated, standardized, and published through a defined approval chain: eliminating grading disputes
  • The 80% attendance compliance monitor removed a class of checks that previously relied on a staff member remembering to look. The system flags proactively; staff act on exceptions

Security & Role Clarity

  • The six-role RBAC model means every user sees exactly what they should: nothing more. Students can't access admin tools. Preceptors can't see unrelated student data. The Registrar has override power without needing to call a developer
  • The Orange Checkmark override tool alone has eliminated a recurring category of support requests, letting the Super Admin unlock and re-open submitted forms independently

Accreditation-Ready Reporting

  • Leadership gained real-time visibility into USMLE pass rates, clinical completion, census data, and instructor performance from a single dashboard: without exporting anything manually or consolidating spreadsheets 
  • Automated histogram and pie chart generation from survey and evaluation data means accreditation report preparation no longer requires a dedicated manual effort before submission deadlines

Conclusion

By leveraging KodeFast's enterprise platform, the university replaced a fragmented and outdated system with a unified operational backbone designed around the way the institution actually works. The new University Management System enforces policies through intelligent workflows, from financial clearance and registration controls to attendance monitoring, compliance tracking, grading, and document management. Processes that once depended on manual intervention are now automated, governed, and auditable. The result is a modern, scalable, and efficient system that reflects the university's unique operational requirements rather than forcing users to adapt to generic software.