Title: Choosing the Right Analysis Model
Locale: en
URL: https://sensorswave.com/en/docs/analytics/choosing-analysis-model/
Description: Understand the different analysis models and choose the right analysis method for your business scenario
Sensors Wave provides multiple analysis models to help you gain insights into user behavior and business performance from different perspectives. Choosing the right analysis model is the first step to efficient data analysis. Each analysis model is designed for specific analysis scenarios and question types — understanding their core capabilities and applicable scenarios will help you quickly find solutions.

## Analysis Model Overview

Sensors Wave supports the following four core analysis models:

### Event Analysis

**Core capability**: Track and measure user behavior events, with multi-dimensional analysis and comparison support.

**Applicable scenarios**:
- Analyze the usage frequency and trends of specific features
- Compare behavioral differences across user groups
- Monitor changes in key business metrics
- Identify user behavior patterns and high-frequency actions

**Typical questions**:
- How many users used the search feature each day?
- What are the differences in activity across users from different channels?
- What is the trend in product favorites over the past 30 days?

### Retention Analysis

**Core capability**: Measure users' continued return visits after initial use, analyzing user stickiness and product value.

**Applicable scenarios**:
- Assess product stickiness and long-term value
- Identify key factors that impact user retention
- Compare the impact of different features or releases on retention
- Analyze user lifetime value

**Typical questions**:
- What is the Day-7 retention rate for new users after registration?
- Do users who used coupons have higher retention?
- Which version of the onboarding flow leads to better retention?

### Funnel Analysis

**Core capability**: Analyze user conversion in multi-step processes, identifying drop-off points and optimization opportunities.

**Applicable scenarios**:
- Analyze the conversion efficiency of key business processes
- Identify the steps and reasons for user drop-off
- Evaluate the impact of product optimizations on conversion rates
- Compare conversion performance across different user groups

**Typical questions**:
- What is the overall conversion rate from browsing products to completing payment?
- At which step do users drop off the most?
- Did the new checkout flow improve conversion rates?

### SQL Query

**Core capability**: Use SQL for flexible, custom data queries and analysis.

**Applicable scenarios**:
- Execute complex multi-table join analysis
- Perform custom data aggregation and calculations
- Exploratory data analysis and hypothesis validation
- Address special needs that standard analysis models cannot cover

**Typical questions**:
- Need to join multiple data tables for combined analysis
- Need to use custom calculation logic
- Need to perform ad-hoc data exploration

## How to Choose an Analysis Model?

Based on the type of question you want to answer, you can quickly determine the suitable analysis model:

### Decision Flow

Follow these questions in order:

1. **Do you want to analyze conversion across a multi-step process?**
   - Yes → Use **Funnel Analysis**
   - No → Continue to the next step

2. **Do you want to analyze user return visits and retention?**
   - Yes → Use **Retention Analysis**
   - No → Continue to the next step

3. **Can standard analysis models meet your needs?**
   - Yes → Use **Event Analysis**
   - No → Use **SQL Query**

### Analysis Model Comparison

| Analysis Model | Primary Use | Difficulty | Typical Use Cases |
|---------|---------|---------|------------|
| **Event Analysis** | Measure behavior frequency and trends | Simple | Feature usage analysis, user activity analysis |
| **Retention Analysis** | Analyze user stickiness and return visits | Medium | New user retention, feature value validation |
| **Funnel Analysis** | Analyze process conversion efficiency | Medium | Purchase conversion, registration flow optimization |
| **SQL Query** | Custom complex analysis | Complex | Multi-table joins, custom calculation logic |

## Recommendations

### Start Simple

If you're just getting started with Sensors Wave, we recommend learning and using the analysis models in the following order:

1. **Event Analysis**: The most basic analysis model and the starting point for understanding user behavior
2. **Funnel Analysis**: Analyze business process conversion to quickly discover optimization opportunities
3. **Retention Analysis**: Gain a deeper understanding of user value and product stickiness
4. **SQL Query**: Handle complex, customized analysis needs

### Combine Multiple Models

In practice, you'll often need to combine multiple analysis models:

- **Event Analysis + Retention Analysis**: First use Event Analysis to identify key behaviors, then use Retention Analysis to validate their impact on user stickiness
- **Funnel Analysis + Event Analysis**: Use funnels to discover drop-off points, then use Event Analysis to investigate user behavior at those points
- **Standard Models + SQL Query**: Use standard models for routine analysis, and SQL for special customized needs

### Focus on Core Questions

When choosing an analysis model, always be guided by the core question you want to answer:

- **"How many" questions** → Event Analysis
- **"Do they come back" questions** → Retention Analysis
- **"How's the conversion" questions** → Funnel Analysis
- **"Custom needs" questions** → SQL Query

## Prerequisites

Before using any analysis model, make sure you have:
- Completed SDK integration and started receiving data, see [Data Integration](../data-integration/tracking-strategy.mdx)
- Understood the data model concepts of Sensors Wave, see [Data Model](../data-integration/data-model.mdx)
- Have the appropriate project access permissions, see [Roles and User Permissions](../10-组织与项目/03-角色和用户权限.md)

## Next Steps

Choose the analysis model that suits you and start your Insights journey:

1. **[Common Analysis Features](common-analysis-features.mdx)**: Learn about features shared across all analysis models
2. **[Event Analysis](event-analysis.mdx)**: Learn how to use Event Analysis in depth
3. **[Retention Analysis](retention-analysis.mdx)**: Master Retention Analysis techniques
4. **[Funnel Analysis](funnel-analysis.mdx)**: Learn Funnel Analysis methods
5. **[SQL Query](sql-query.mdx)**: Learn how to use SQL for custom analysis

If you encounter issues during use, check the [FAQ](faq.mdx) or refer to the [Best Practices](best-practices.mdx).