Understanding Indicator Systems: Definitions, Lifecycle, and Construction Methodology

1. What Is an Indicator System

An indicator system is a collection of metrics used to evaluate a company's business status from multiple dimensions; it combines indicators (quantified measurements of business units) with a structured framework.

Indicators make business goals describable, measurable, and decomposable, serving as the bridge between business and data.

Indicators are mainly divided into result‑type (measured after a user action, often lagging) and process‑type (measured during the action and can be influenced by operational strategies).

The system consists of dimensions, which are the perspectives or "thinking angles" for observing and describing phenomena; without dimensions, indicators lack meaning.

Dimensions are categorized as qualitative (textual descriptors such as city, gender, occupation) and quantitative (numeric descriptors such as income, age, requiring grouping).

2. Indicator System Lifecycle

The lifecycle includes definition, production, consumption, and deprecation, requiring continuous operation, quality assurance, and data‑operation work to improve reuse and reduce user cost.

2. Why Build an Indicator System

1. Understand Business Status

It provides objective data for different departments and managers, aligning their understanding of the current business situation.

2. Identify Business Pain Points

Using models like the AARRR pirate metrics, it reveals issues across acquisition, activation, retention, revenue, and referral stages.

3. Guide and Drive Business

Indicators enable result prediction, early warning, and anomaly attribution to uncover business flaws.

4. Optimize Product or Business Logic

Funnel analysis helps spot major losses in processes and target optimization efforts.

3. How to Build an Indicator System

1. Indicator Value

Technical goal: unify indicator and dimension management, standardize naming and calculation, ensure consistency.

Business goal: provide unified data outputs for multiple scenarios.

Product goal: deliver a tool that streamlines indicator definition, production, and consumption.

2. Model Architecture

1) Business Lines

Define business segments by abstracting logical layers and physically organizing structures, typically up to three hierarchical levels.

2) Standard Definitions

Data domain: abstract collection of business processes or dimensions; processes are indivisible events, dimensions are measurement contexts.

Business process: indivisible activity events such as order placement or payment.

Time period: specifies the statistical range (e.g., last 30 days, natural week).

Modifiers: non‑dimensional qualifiers like APP or PC.

Metric/Atomic indicator: indivisible measurement tied to a business event (e.g., transaction amount).

Dimension: attribute set reflecting a business property (e.g., geographic or time dimension).

Dimension attribute: specific fields within a dimension (e.g., region name, ID).

Indicator classification: atomic, derived, or composite.

Atomic indicator: basic metric such as transaction count or amount.

Derived indicator: atomic indicator + modifiers + time period.

Derived indicators further split into transaction‑type (measuring processes) and stock‑type (measuring entity states). Stock‑type examples use a time period like historical up to a specific date .

Composite indicator: combines transaction‑type and stock‑type into ratios, growth rates, or averages.

3. Indicator System Construction Process

1) Collaboration Workflow

2) Naming Convention

Atomic indicator = business process + metric. Example: use trd_pay_amt for transaction payment amount to avoid duplication with generic pay_amt .

Derived indicator = atomic indicator + modifiers + time period. Modifiers follow the atomic part to preserve inheritance and readability; undefined modifiers receive a three‑digit code. Example: trd_pay_amt_7d for payment amount over the last 7 days.

3) Indicator Model Construction

Modeling abstracts business requirements into thematic groups, standardizes terminology, and reduces duplication. The data warehouse follows Kimball dimensional modeling, comprising dimensions and fact tables.

Data Model

Model Definition Standards

4. Productization of the Indicator System

1) Indicator Product Overview

The product suite aligns with the lifecycle, providing tools that automate, standardize, and streamline indicator management, thereby turning data into business value.

Key product capabilities include generating standardized indicators to eliminate naming conflicts and providing a unified API for external consumption.

2) Tool Design Process

1) Metadata Management

2) Indicator Definition

3) Indicator Production

4) Indicator Consumption

Conclusion

The article presented a comprehensive methodology for building indicator systems and the associated product tools, noting that data indicators and development tools are already integrated internally, with future data services to be offered via DataAPI.

Indicator system tools have been deployed within the company, covering core business units, and will continue to evolve through practice.