> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nrev.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Custom Code
> Run simple Python on your data—great for quick transformations, filters, or cleanups that aren’t yet available via built-in nodes.
## What It Does
* **Applies Python logic** to reshape, filter, or enrich your data
* **Great for edge cases** where no standard node (yet!) exists
* **Executes safely** and returns updated data just like any other node
* **Ideal for RevOps, GTM, and CS teams** who need a little extra power
***
## 🏁 Getting Started
Drop **Custom Code** anywhere you need more control.
Choose any upstream table you want to transform.
Your data is already loaded as `df`. Edit and manipulate it as needed.
Hit **Run** — your logic is applied row-by-row.
***
## Inputs
### 🛠️ Required
* **Python Code (✅)**\
Simple Python edits applied to your `df`.\
*Why it matters:* This is where your logic lives — whether filtering rows, scoring leads, or cleaning fields.
⚠️ Invalid code will throw an error, and the node will not run until fixed.
***
## Output
* Your updated data, transformed exactly how your logic says
* All original columns are preserved unless you overwrite them
* You can keep building off this output downstream
***
## 🚀 Example Use Cases
| Use Case | Code Snippet |
| ----------------- | ---------------------------------------------------- |
| Add a column | `df['is_target'] = df['industry'] == 'SaaS'` |
| Clean up names | `df['name'] = df['name'].str.title().str.strip()` |
| Filter rows | `df = df[df['region'] == 'North America']` |
| Create lead score | `df['score'] = df['revenue'] / df['employee_count']` |
***
## ✨ Pro Tips
Use this node **only when a built-in node doesn’t meet your need** — it keeps flows simpler and more scalable.
Leverage our AI configurer to generate code. You don't have to do it all yourself!
Start small — even a one-liner can save hours of spreadsheet work later.
Use the toggle button to switch between test mode (**purple**) and full mode (**green**).\
Test mode runs your code on just the first 5 rows — great for testing your logic before applying it across the full dataset.
***
## ⚠️ Considerations
Broken code = node won’t run. The UI will surface any syntax issues immediately.
Only use this node for scenarios we don’t already support — we’re constantly adding more built-in nodes to make your life easier.
Be mindful when renaming or dropping fields — changes persist downstream.
***
## 📝 FAQ
It’s your input data — a DataFrame you can write Python against.
Nope. If you can describe the logic, you can usually get AI tools to write it for you.
Built-in nodes are faster, safer, and easier to maintain. Use code only when necessary.
***
## 💰 Pricing
> This node is **free** — no credits required.
***
Use code only when you need it. Custom Code gives you flexibility — but we’re always building smarter nodes to save you the trouble.