Back / Content Delivery Automation
B. Sivarami Reddy
Principal Designer & UX Design Manager
Profile

Content Delivery Automation

Week-Based Learning System — Reducing PM Workload from 1 Day to 0 Hours Per Week

Project Overview

  • Role — Lead Product Designer (IC)
  • Company — Great Learning
  • Timeline — 6 months (May 2024 – Nov 2025)
  • Team — 1 Designer, 3 Engineers, 1 PM, Academic Ops stakeholders
  • Tools — Figma, User interviews, Workflow mapping, Stakeholder workshops

The Problem

Program Managers were spending 1 full day every week manually releasing content, coordinating with Academic Ops about curriculum changes, and dealing with version mismatches across batches.

The Solution

Automated delivery schedule system where Academic Ops creates a master plan once, and it automatically syncs to all batches with smart conflict resolution.

Impact at a Glance

1 day → 0 hrs

PM time on content release per week, per program manager

500+ PMs

Across programs benefiting from full automation

Zero errors

Content release errors from version mismatches (was 15%)

₹1.2Cr+

Annual savings in PM productivity across the organisation

340 → 0

Monthly support tickets for content release issues

100% adoption

Across all domestic programs within 6 months of launch

The Challenge

Through shadowing sessions with 15 PMs and 5 Academic Ops team members, I mapped their exhausting weekly routine. The problem wasn't just inefficiency — it was a system misaligned with how people actually think about content.

A PM's Weekly Schedule

  • Monday — Follow-up: learner attendance, feedback scores, missing submissions
  • Tuesday — Spillover from Monday: escalations, learner support
  • Wednesday — Team meetings: reports, batch health, plan interventions
  • ThursdayContent Release Day: 6–8 hours of manual work
  • Friday — Plan next week's live sessions, coordinate with faculty

The Thursday Nightmare — 8 Steps Every Week

  1. Check with Academic Ops: "What content for this week?"
  2. Go to template batch, find modules for Week X
  3. Delete old modules from actual batch
  4. Import new modules from template (one by one)
  5. Set unlock dates (Thursday/Friday release)
  6. Set due dates (following Sunday)
  7. Release content to learners
  8. Post announcement in course

6–8 hours per PM managing 3–4 batches. Every. Single. Week.

Why Thursday/Friday release for Monday content?

Early finishers start over the weekend. Motivated learners don't wait until Monday. Reduces Monday support load. But managing this manually across programs with different release days created constant errors.

The Numbers Behind the Problem

Survey of 50+ PMs across a 500+ PM organisation:

PM Pain Points

  • 85% — Thursday content release was their most time-consuming weekly task
  • 92% — Had released wrong content at least once
  • 68% — Spent additional time fixing version mismatch issues
  • 73% — Struggled managing different release day configs across programs
  • 100% — Wanted automation

Academic Ops Pain Points

  • Every curriculum change required manually notifying 20+ PMs
  • No visibility into which batches had received updates
  • 2–3 hours/week answering PM coordination questions
  • Excel-based delivery schedule had no system enforcement
Infographic: Old Workflow — 8-step manual process, 6–8 hours, high error risk

Research & Discovery

Method 1: PM Shadowing (15 sessions)

Sat with PMs during their content release ritual. Observed, timed, and documented every step.

Insight 1: Week-Based Mental Model

"I think in weeks. Week 1 is Python basics, Week 2 is data structures. But the system doesn't work that way — I have to remember which modules are in which week."
— PM, Data Science

Insight 2: Template ≠ Reality

"The template changes, but I don't know when or what changed. I just copy everything on Thursday and hope it's right."
— Senior PM

Insight 3: Release Day Chaos

"My DSBA batch releases Thursday, AIML releases Friday. I have to remember which is which. One mistake and learners get confused."
— PM, multiple programs

Method 2: Academic Ops Workshops (5 sessions)

AcadOps had a clear mental model of the curriculum — programs by weeks, courses spanning weeks, modules grouped by week. But this structure existed only in Excel, not in the system.

Their pain: creating delivery schedules in spreadsheets, sharing with PMs manually, no system enforcement, curriculum changes = notification nightmare to 20+ PMs.

Method 3: Data Analysis (3 months of batch data)

2,400+

Manual content imports across 200 batches

340

Support tickets related to content release issues

15%

Of content releases had errors — wrong week, wrong modules

Design Process

Phase 1: Information Architecture — Week-Based Hierarchy (Weeks 1–3)

The core structural decision: introduce time as a first-class dimension in the content hierarchy.

Old: Flat Structure

Program → Course → Module

No time dimension. PMs had to mentally map modules to weeks.

New: Week-Based Hierarchy

Program → Delivery (Week 1, Week 2…) → Course → Module

Matches exactly how PMs and AcadOps think.

Infographic: IA — Old Flat vs New Week-Based Hierarchy

Why it worked: Aligns with how people think ("Week 3 content"). Makes delivery schedule visual and scannable. Enables bulk operations — shift all weeks, add a break week.

Phase 2: Template → Batch Sync Logic (Weeks 4–8)

The hardest design problem: how do you sync schedules when batches start at different times and have different break weeks?

❌ Version 1: Full plan sync

Copy entire delivery plan from template → batch. Problem: Batch-specific breaks got overwritten.

❌ Version 2: Manual merge

Show diff, let PM choose what to update. Problem: Too complex — PMs didn't understand the differences.

✅ Version 3: Additive sync with smart rules

Sync upcoming deliveries automatically. Published deliveries stay frozen. New content from template gets added to existing content. PM reviews before publishing.

Breakthrough from PM testing

"Oh! So it only updates the weeks I haven't released yet. That makes sense — I don't want to change what students already saw."
— PM during prototype testing
Wireframe: Conflict Resolution Dialog — View 1 Wireframe: Conflict Resolution Dialog — View 2 Wireframe: Conflict Resolution Dialog — View 3

Phase 3: Break Week Management (Weeks 9–11)

PMs needed flexibility for local holidays without breaking the template sync. Two distinct break types solved this cleanly:

Learning Break (in template)

Built into the program structure by AcadOps. Part of the master plan — syncs to all batches.

Break Week (in batch)

Added locally by PM for Diwali, long weekends, etc. When added, system auto-adjusts all subsequent delivery dates.

Screenshot: Break Week Dialog

Phase 4: Content Import Automation (Weeks 12–16)

Old flow — 7 steps

  1. Manually delete old modules
  2. Import from template (one by one)
  3. Set unlock dates
  4. Set due dates
  5. Publish to learners
  6. Write announcement
  7. Post in course

New flow — 1 step

Click "Sync & Publish" → done.

  • Import content from template
  • Set unlock date automatically
  • Publish to learners
  • Generate announcement (PM can edit)
Screenshot: Error State — Missing Modules

Phase 5: Version Management (Weeks 17–20)

AcadOps creates multiple versions of the delivery plan in the template. Batches needed to know which version they were on and see exactly what changed.

Template Batch (AcadOps)

  • Publish v2 of delivery plan
  • System shows: "24 batches will be affected"
  • Review each batch, resolve conflicts
  • Confirm sync

Actual Batch (PM)

  • Notification: "New version available from template"
  • Click "Sync" on each upcoming delivery
  • Visual diff: what changed?
  • Confirm or keep current version
Infographic: Version Sync Flow — Template v2 → Review batches → Resolve conflicts → Sync

Key Design Decisions

1. Week-Based Organisation

Introduced Delivery as a first-class entity — a named week of learning with courses and modules inside it. This single structural decision unlocked everything else.

Batch Delivery Planning — View 1 Batch Delivery Planning — View 2 Batch Delivery Planning — View 3 Batch Delivery Planning — View 4 Batch Delivery Planning — View 5 Batch Delivery Planning — View 6 Batch Delivery Planning — View 7 Batch Delivery Planning — View 8 Batch Delivery Planning — View 9 Batch Delivery Planning — View 10

Template Batch

Template Delivery Planning — View 1 Template Delivery Planning — View 2 Template Delivery Planning — View 3

Impact: PMs think in weeks — UI matches. Easy to communicate dates. Bulk operations become possible (shift all weeks by 1, add break).

2. Additive Sync with Smart Rules

Rules:

  • ✅ Sync upcoming deliveries automatically
  • 🔒 Published deliveries stay frozen
  • ➕ New content from template added to existing
  • 👤 PM reviews before publishing

Visual language:

  • Blue: New modules being added from template
  • Existing modules remain untouched
  • Single action: "Sync & Publish"

Why additive, not overwrite?

Safer — never removes content students may have started. Clearer — PMs see exactly what's new. Flexible — PMs can manually remove if needed.

Impact: Zero accidental content removal. Clear visibility. PMs stay in control of final publish.

Template Sync Preview — View 1 Template Sync Preview — View 2

3. Program-Level Release Day Configuration

Set once at program level (Thursday or Friday). All batches inherit this. System auto-calculates unlock dates, start dates, and due dates.

How it works

Release day

Thursday or Friday

Early access window for fast learners

Content start

Monday

Official start for all learners

Due date

Sunday

End of the learning week

Impact: No more manual date calculations per program. PMs don't need to remember which program releases which day.

4. Automatic Date Calculation

System auto-calculates all delivery dates from three inputs: content start date (set once), delivery durations (from template), and break weeks (added by PM).

Infographic: Date Calculation Logic — Start date + Durations + Breaks = Auto-calculated timeline

Impact: Adding a break week = 1 click instead of 2 hours of recalculations. Date errors eliminated entirely.

5. "Sync & Publish" — One Action

Old flow had 7 steps across multiple screens. New flow: one button does everything — import, configure dates, publish, and generate announcement.

Sync & Publish Dialog — View 1 Sync & Publish Dialog — View 2

Impact: 6–8 hours → 5 minutes. Error-proof — can't forget a step. Announcement auto-generated from template.

6. Access Control Hierarchy

Clear permission model separating template (sacred) from batch (flexible).

Action AcadOps Lead AcadOps Admin PM
Edit template
Publish template (create version)
Edit batch & save
Publish batch (release to learners)

Why: Prevents accidental changes while giving PMs flexibility for local adjustments.

Impact & Results

6 months post-launch across 500+ PMs.

PM Efficiency

100% time saved — 1 day → 0 hours per PM per week

3,000+ hours saved across 500 PMs per month

₹1.2Cr+ annual savings in PM productivity

Quality Improvements

0 content release errors (was 15%)

0 version mismatch incidents

340 → 0 support tickets for content release issues

Adoption

100% adoption across all domestic programs

95% PM satisfaction (up from 35%)

9/10 NPS from PMs

Academic Ops

Curriculum changes propagate in minutes (was hours/days)

100% visibility into which batches use which version

2–3 hrs/week saved on PM coordination

"I used to dread Thursdays. 6–8 hours of manual content release. Now I just click Sync & Publish and I'm done in 5 minutes. Life-changing."
— PM, Data Science Program
"The best part? I don't have to remember which program releases Thursday vs Friday. The system knows and handles it."
— Senior PM, managing 5 programs
"We can now update curriculum without worrying about PM coordination. The system handles it."
— AcadOps Lead
"Adding break weeks used to take 2 hours of date recalculations. Now it's one click and everything auto-adjusts."
— PM, Business Analytics

Challenges & Learnings

1. Designing for Two User Types

Problem: PMs and AcadOps had different needs and mental models.

Solution: Same underlying system, different interfaces — template batch is planning-focused (AcadOps), actual batch is execution-focused (PMs).

Learning: Don't force one interface for two workflows. Optimise each separately.

2. Sync Logic Complexity

Problem: So many edge cases — batch started mid-program, break weeks, curriculum changes mid-run.

Solution: Prototyped with 10 real batches covering every sync edge case. Iterated on conflict resolution UI 5 times.

Learning: Complex logic needs extensive testing with real scenarios, not hypotheticals.

3. Change Management

Problem: PMs were used to manual control. Would they trust automation?

Solution: Progressive rollout — Phase 1: template creation only. Phase 2: manual sync (PM triggers). Phase 3: auto-sync with conflict resolution.

Learning: Introduce automation gradually. Let users build trust before removing control.

4. Handling Exceptions

Problem: Every program had unique quirks — prework, optional courses, capstone weeks.

Solution: Flexible rules engine — tag course types (Regular, Optional, Capstone), different sync behaviour per type, override option for edge cases.

Learning: Design for the 80% case, but always provide escape hatches for the 20%.

Future Enhancements

Predictive Scheduling

AI suggests optimal delivery dates based on historical completion rates. Auto-adjust for low engagement weeks.

Learner Analytics Integration

See learner progress per delivery. Flag deliveries with low completion so PMs can intervene proactively.

Bulk Operations

Shift all deliveries by X days. Duplicate delivery schedule across programs. Batch-apply template updates.

Assessment Integration

Auto-schedule assessments based on content delivery. Due dates calculated from content unlock automatically.

Reflection

This was the most complex workflow automation I've designed.

Key takeaway: Complex automation isn't about removing humans from the loop — it's about removing repetitive work while keeping them in control of exceptions.

What Made It Successful

  • User research depth — shadowed 20+ users before designing a single screen
  • Stakeholder alignment — weekly syncs with AcadOps, PMs, and Engineering
  • Iterative prototyping — tested 5 versions of sync logic with real data
  • Phased rollout — built user trust gradually before full automation

Personal Growth

  • Designing for multiple user types with competing needs
  • Managing technical complexity without overwhelming users
  • Change management for workflow automation
  • Balancing flexibility with guardrails at enterprise scale

All metrics from Great Learning internal data (Nov 2025). 500+ PMs, 100% adoption, 1 day → 0 hours saved per PM per week.