Background

Ruangguru is Indonesia’s leading educational technology company, offering a wide range of online learning services for students and professionals. During my time at Ruangguru, one of the critical business goals was to boost customer acquisition, retention, and engagement through effective promotional strategies. To achieve this, I led the design and development of a promo engine—a scalable, flexible system that could handle various discount, referral, and loyalty-based promotions.

Problem Statement

<aside> 💡

The goal was to create an automated promo engine that allowed the marketing team to define and launch promotions quickly, with real-time tracking of performance metrics.

</aside>

The previous promotional management system at Ruangguru had significant limitations:

• Promotions were hard-coded and required developer support for each new offer.
• Lack of flexibility in applying dynamic conditions (e.g., user segmentation, product bundles, time-based triggers).
• Inability to measure the success of different promotions effectively, which made A/B testing difficult.
• Manual setup led to human errors, creating inefficiencies and delays in launching new campaigns.

Hypothesis

Implementing a granular and product-specific promo engine tailored to each product line's unique characteristics will lead to increased sales, better user engagement, and higher conversion rates, particularly in targeted product segments.

Approach

Market Research and Competitive Analysis

To understand how other companies manage promo engines, we analyzed various competitors in the edtech and e-commerce sectors.

<aside> 💡

Key Insight: Companies with high product diversity (e.g., Amazon) tend to have more granular, product-specific promo engines, which inspired Ruangguru's approach to tailoring promotions for different product lines.

</aside>

Audience Analysis

Design and Technical Architecture

graph TD
    A[Client] --> B[API Gateway]
    B --> C[Promo Creation Service]
    B --> D[Rule Engine Service]
    B --> E[Promo Application Service]
    B --> F[Analytics and Reporting Service]
    B --> G[User Segmentation Service]
    B --> H[Product Catalog Service]

    C --> I[(Promo DB)]
    D --> J[(Rule DB)]
    E --> K[(Transaction DB)]
    F --> L[(Analytics DB)]
    G --> M[(User DB)]
    H --> N[(Product DB)]

    O[Admin Dashboard] --> B

    P[Message Queue] --> C
    P --> D
    P --> E
    P --> F
    P --> G
    P --> H

    subgraph Data Stores
        I
        J
        K
        L
        M
        N
    end

    subgraph Core Services
        C
        D
        E
        F
        G
        H
    end

Challenges

• High Granularity Requirements: Due to Ruangguru’s emphasis on achieving sales targets, the promo engine needed to account for various detailed promotion types (e.g., user-based, course-based, seasonal discounts). This introduced complexity in managing different variables, rules, and conditions.
• Product-Specific Focus: The promo engine was not applied uniformly across all products, which added another layer of complexity. Instead, it was tailored for certain key products, making it necessary to create product-specific logic and conditions for the engine.
• Scalability: With more products being developed, ensuring the promo engine could scale and adapt to future product launches was a critical factor to consider during the design.

Feature Prioritization & MVP Definitions

<aside> 💡

• As a student, I want to find relevant course promotions, so that I can afford supplementary learning materials.
• As a parent, I want to apply a discount on educational bundles, so that I can enhance my child’s learning within a budget.
• As a professional, I want a discount on certification courses, so that I can upskill for career advancement. </aside>

MVP Features:

1. Promo engine for specific products (e.g., course-based or bundle-based).
2. User segmentation to ensure personalized promos.
3. Rules engine for granularity in conditions (eligibility, timing, etc.).
4. Backend dashboard for marketing to launch and adjust promotions.

Solution Explored

Solution Implemented