How to analyze data in a chat
To analyze data within a Cassidy chat, follow these steps:
Start a new chat: Begin by clicking the "New Chat" button in the left sidebar or on the "All Chats" page.
Select an Assistant: Select the "Data Analyst" Assistant (which won't have context on your business) or a custom Assistant that has the Data Analysis tool enabled. For more information on enabling the Data Analysis tool, see Data Analysis.
Upload your data file: Upload the data file (or if using a custom Assistant it may be able to search your Knowledge Base for files or reference any that are attached to its instructions already). Click the "+" button in the bottom left corner of the text box, or drag and drop the CSV or Excel file containing the data you want to analyze. Make sure the file adheres to the formatting guidelines outlined in the next section.
Ask questions and request visualizations: Once your data is uploaded, you can ask the Data Analyst assistant questions related to your data. For example, you can request specific visualizations (e.g., "Create a bar chart showing sales by region"), gather insights (e.g., "What are the key trends in customer behavior?"), or ask specific data questions (e.g., "What was the total revenue for Q1?").
View the analysis and code: Click the "Analyzed sources" dropdown in the results panel, then click the code icon (
</>) next to any analysis step to open a modal showing the actual Python code the AI wrote to process your data.
Various use cases
The Data Analysis tool supports a range of tasks beyond traditional data analysis:
Manipulating branded template files: The tool can work with Word documents, PDFs, PowerPoint presentations, and other file types to create new files or edit existing ones. For example, you can ask it to "fill out this form" or "turn this report into a PowerPoint presentation."
Creating synthetic data: Generate sample datasets for testing, demonstrations, or modeling purposes without requiring existing data files.
How to properly format data for analysis
To ensure smooth data analysis and avoid potential issues, follow these guidelines when preparing your data:
Use CSV or Excel files: The Data Analyst assistant works with data in CSV (comma-separated values) or Excel (XLSX) format only.
Organize data in a tabular structure: Your data should be arranged in a table format, with rows representing individual records and columns representing variables or attributes.
Include descriptive column headers: The first row of your spreadsheet should contain descriptive headers for each column. Avoid using empty rows or titles above the header row.
Ensure data consistency: Make sure your data is consistent and follows a uniform structure. Avoid leaving empty rows between data entries or separating data into multiple sections or tables within the same file. In addition, avoid using multiple tabs/worksheets.
By adhering to these formatting guidelines, you can ensure that the Data Analyst assistant can process and analyze your data effectively, helping you uncover valuable insights and make data-driven decisions within your Cassidy chats.