How to Set Up Sourcetable for AI Data Analysis

What Sourcetable Does and Why It Matters

Sourcetable is an AI-powered spreadsheet that combines data connectivity, natural language analysis, and Python-level analytical power in a single browser-based interface. Unlike traditional spreadsheets that top out at about one million rows and require manual formula writing, Sourcetable processes datasets exceeding that limit server-side and lets you ask questions about your data in plain English.

The platform connects directly to over 900 data sources, including PostgreSQL, MySQL, MongoDB, Stripe, HubSpot, Google Ads, and QuickBooks. You can also upload CSV, Excel, JSON, TSV, and PDF files up to 10GB. This means analysts, ops managers, and founders can pull live data into a familiar spreadsheet interface without writing SQL or setting up ETL pipelines.

Step 1: Create Your Account

Head to sourcetable.com and click the sign-up button. Account creation takes under 30 seconds. Every new user gets a free tier with basic spreadsheet functionality and limited AI credits. If you want full access to AI features and unlimited data connectors, activate the 7-day Pro trial, which does not require a credit card.

Once inside, you will see a clean spreadsheet interface with an AI chat panel on the right side. This chat panel is where most of the analysis work happens.

Step 2: Connect Your First Data Source

Click the data connector icon in the left sidebar. Sourcetable offers two paths for getting data in.

For file uploads, drag and drop a CSV, Excel, or JSON file directly into the workbook. The platform parses the file, detects column types, and makes the data immediately queryable. Files up to 10GB are supported, with large files processed server-side so your browser stays responsive.

For live database connections, select your database type (PostgreSQL, MySQL, BigQuery, Supabase, Clickhouse, or MongoDB), enter your connection credentials, and authenticate. Sourcetable syncs the data and keeps it updated. You can also connect SaaS applications like Stripe, Salesforce, Xero, and GA4 using OAuth authentication, which typically takes two to three clicks.

Once connected, your data appears in a spreadsheet tab. You can have multiple data sources across different tabs in the same workbook, which is useful for cross-referencing datasets.

Step 3: Run Your First AI Analysis

With data loaded, open the AI chat panel and type a question in plain English. For example, if you uploaded a sales CSV, you might type: "What were the top 10 products by revenue last quarter?" or "Show me monthly revenue trends as a line chart."

The AI assistant interprets your question, writes the necessary formulas or Python code behind the scenes, and returns the result directly in your spreadsheet. It can generate pivot tables, summary statistics, charts, and written insights. If the output is not quite right, refine your prompt. The assistant maintains context within a session, so you can say "break that down by region" as a follow-up without restating the full query.

For more complex analysis, you can ask the AI to run statistical tests, build regression models, or create financial projections. The platform supports all standard Excel formulas (VLOOKUP, INDEX/MATCH, SUMIFS, array formulas) alongside AI-generated Python analysis, so you can mix manual formulas with AI outputs in the same sheet.

Step 4: Create Charts and Visualizations

Type a visualization request in the chat panel, such as "Create a bar chart of sales by category" or "Plot a scatter chart of price vs. units sold." Sourcetable generates interactive charts that you can share with your team.

You can also use the chart generator from the toolbar for more manual control over chart type, axis labels, colors, and formatting. Charts update automatically when the underlying data changes, which is particularly useful when working with live database connections.

Step 5: Set Up AI Workflows for Recurring Tasks

Released in March 2026, AI Workflows is Sourcetable's answer to repeatable analysis. If you find yourself running the same type of analysis regularly, such as weekly sales reports, monthly churn calculations, or daily inventory checks, you can convert a successful AI chat session into a reusable workflow.

After completing an analysis in the chat panel that produces the output you want, click the option to create a workflow from it. The system saves the instruction set as an editable, code-free document. You can then deploy this workflow across different workbooks or share it with your team. Think of it as a macro that understands natural language: it handles fuzzy tasks that vary slightly each time, like cleaning data from different sources that follow similar but not identical formats.

Workflows can chain together multiple agents, tools, and data integrations. For example, a single workflow could pull fresh data from your PostgreSQL database, clean and normalize it, calculate key metrics, generate a summary chart, and format the output as a report tab.

Step 6: Collaborate and Share

Sourcetable supports real-time collaboration with multiple simultaneous users. Permission levels include Owner, Editor, and Viewer roles, with granularity down to individual sheets. All changes are automatically saved and versioned, so you can view edit history, see who made changes, and restore previous versions at any time.

To share a workbook, use the share button in the top-right corner and invite team members by email. Viewers can see the data and charts without editing access, which is useful for distributing dashboards to stakeholders.

Pricing and Plans

The free tier covers basic spreadsheet functionality with limited AI usage. The Pro plan costs $20 per month (or 20% less with annual billing) and includes unlimited connectors, full AI features, and advanced analysis capabilities. Students and faculty get 50% off Pro and Max plans with a .edu email. Teams of 10 or more qualify for volume discounts, and nonprofits receive 50% off all plans.

If you need an even simpler path to data analysis without setting up connectors or learning a new interface, tools like VSLZ AI let you upload a file and get end-to-end analysis from a single prompt, with no configuration required.

What to Do Next

Start with a dataset you already know well, like a sales export or a financial report. Upload it to Sourcetable, ask the AI three or four questions you would normally answer manually in Excel, and compare the speed and accuracy. Once comfortable, connect a live data source and experiment with workflows to automate your most repetitive reporting tasks.