When you use a REST API as a data source, you can apply an automatic data cleaning step before the response reaches your destination spreadsheet. Use Sheetgo's AI Processor to write plain-English instructions telling the model exactly how to clean, normalize, or transform your API data — configured once, applied on every run.
Requirement: Your connection must use an HTTP Request (JSON) source.
How to set it up
Step 1 — Configure your REST API connection
In app.sheetgo.com, open an existing workflow or create a new one. Click New automation, then under Select source choose HTTP Request (JSON) and enter your API endpoint URL and any required authentication headers.
Step 2 — Select "Process with AI" as the data processor
After configuring the source, click the arrow to proceed to the Select a data processor step. Under Transform data, select Process with AI.
Step 3 — Write your cleaning instructions
In the Instructions field, describe exactly how the AI should clean or format your data. You can type freely, or click one of the Most Popular suggestion chips to use a pre-built starting point:
Clean up & normalize — strip unwanted characters, fix inconsistent casing, standardize date formats, and remove extra whitespace across text fields.
Rewrite — rephrase or simplify text field values before they land in your sheet.
Categorize — classify each row into a predefined category based on field values.
Summarize — condense long text fields into a single sentence per row.
The more specific your instruction, the better the results. For example: "Convert all date fields from ISO 8601 format to DD/MM/YYYY. Strip special characters from the name field. Capitalize the first letter of each word in the category field."
Step 4 — Save and run
Click Done editing, then Save changes. On every run, the AI processes your API data after the fetch and before it loads into your destination — only clean, ready-to-use data reaches your spreadsheet.
What you can clean with instructions
Goal | Example instruction |
Standardize date formats | "Convert all date fields from ISO 8601 to DD/MM/YYYY" |
Convert Unix timestamps | "Convert the created_at field from Unix timestamp to a readable date" |
Remove unwanted characters | "Strip HTML tags and special characters from the description field" |
Normalize text casing | "Capitalize the first letter of each word in the name field" |
Fill missing values | "If the status field is empty, fill it with 'Unknown'" |