Skip to main content

Scroll

What is a Scroll​

A scroll in AskTable is a visual data analysis canvas. By dragging nodes and connecting data flows, you can build a complete process from data acquisition to analysis presentation.

Core Value: Make the data analysis process visual, reusable, and collaborative.

Why Do You Need a Scroll​

Traditional data analysis faces three challenges:

  1. Opaque Process - SQL queries, data processing, and chart generation are scattered, making it hard to trace
  2. No Reusability - Each analysis requires rewriting code, leading to redundant work
  3. Collaboration Difficulties - Analysis logic is hidden in code, making it hard for team members to understand and take over

The scroll solves these issues through node-based design:

  • Each node represents a data operation (query, transformation, visualization)
  • Connections between nodes show data flow paths
  • The entire scroll represents a complete analysis process, which can be saved, shared, and reused

Getting Started​

Create Your First Scroll​

  1. Click on "Scroll" in the left navigation bar
  2. Click the "Create Scroll" button
  3. Choose a data source type to start building

Data Source Types​

The scroll supports four types of data sources:

Query Database

  • Connect to internal enterprise databases
  • Query and analyze data using natural language
  • Use Cases: Business data analysis, operational reports

Query Feishu Spreadsheets

  • Connect to Feishu multi-dimensional spreadsheets
  • Directly query collaborative data
  • Use Cases: Team collaboration data, project management data

Online Search

  • Search for public market data, industry reports, and trend analysis
  • Example: 2024 China GDP growth rate
  • Use Cases: Market research, industry analysis

Import File

  • Upload Excel or CSV files
  • Instantly analyze table data
  • Use Cases: Temporary data analysis, external data import

Scroll Editor​

Interface Layout​

  • Top Toolbar - Scroll name, view switching, saving
  • Left Panel - Data source selection, node library
  • Central Canvas - Node arrangement area
  • Right Panel - Node configuration, attribute settings

Node Operations​

Add Node Drag a data source or analysis node from the left panel to the canvas

Connect Nodes Drag the output port of one node to the input port of another node

Configure Node Click the node and configure parameters in the right panel

Delete Node Select the node and press Delete key or right-click to delete

Use Cases​

Scenario One: Periodic Business Reports​

Build a scroll once and execute it automatically every week:

  1. Database node → Query sales data for this week
  2. Transformation node → Calculate year-over-year and month-over-month comparisons
  3. Chart node → Generate visual reports
  4. Share node → Automatically send to the team

Scenario Two: Market Trend Analysis​

Combine multiple data sources:

  1. Online search node → Get industry report data
  2. Database node → Query your own business data
  3. Comparative analysis node → Identify gaps and opportunities
  4. Report node → Generate an analytical document

Scenario Three: Temporary Data Exploration​

Quickly analyze uploaded files:

  1. Import file node → Upload Excel
  2. Data cleaning node → Handle outliers
  3. Statistical analysis node → Calculate key metrics
  4. Chart node → Visualize the results

Best Practices​

Naming Conventions​

  • Scroll Name: Describe the analysis objective, such as "Q1 2024 Sales Analysis"
  • Node Name: Explain the node's purpose, such as "Query Order Data"

Structural Design​

  • From left to right: Data source → Data processing → Result presentation
  • Avoid crossing connections: Keep the process clear
  • Modular design: Break complex processes into multiple sub-scrolls

Performance Optimization​

  • Data source node: Add time range restrictions to avoid full table queries
  • Transformation node: Try to perform calculations at the database level
  • Caching strategy: Enable caching for data that doesn't change often

Relationship with Other Features​

Scroll vs Chat

  • Chat: One-time Q&A, suitable for temporary queries
  • Scroll: Reusable process, suitable for periodic analysis

Scroll vs Report

  • Report: Static document, displays analysis results
  • Scroll: Dynamic process, continuously updates data

Scroll vs Quick Query

  • Quick Query: Quick query of a single metric
  • Scroll: Complex analysis process, multi-step processing

Frequently Asked Questions​

Q: Can scrolls be scheduled to run automatically? A: Yes. Configure a scheduled task in the scroll settings, and the system will automatically execute and update the results.

Q: How to share a scroll with team members? A: Click the share button in the top-right corner, select team members, and set permissions (view/edit).

Q: Will the data in a scroll be updated in real-time? A: It depends on the data source. Databases and Feishu spreadsheets will be queried in real-time, while online searches and imported files require manual refresh.

Q: How many nodes can a scroll have at most? A: It is recommended not to exceed 20 nodes. For overly complex processes, split them into multiple sub-scrolls.