> For the complete documentation index, see [llms.txt](https://docs.darcyiq.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.darcyiq.com/organize/lists/columns.md).

# Columns & Configuration

Define, configure, and manage columns for your lists

Columns define the structure of your list. Each column represents a data field — from simple text and numbers to AI-enabled fields that your agent fills automatically during processing.

{% hint style="success" %}
**Manual + AI Columns**: Combine columns you fill yourself (company name, website URL) with AI-enabled columns the agent researches (company overview, tech stack, decision makers) for a powerful hybrid workflow.
{% endhint %}

## Column Types

Every column has a type, and the type is real — the database stores the column as that type rather than keeping everything as text. That's what makes sorting, filtering, and arithmetic behave correctly: a **Number** column sorts 9 before 10, and a **Date and Time** column filters by actual chronology.

There are four types, kept deliberately few because the choice is permanent:

| Type              | Use it for                                            | Notes                                    |
| ----------------- | ----------------------------------------------------- | ---------------------------------------- |
| **Text**          | Anything. The default, and what every older column is | No restrictions                          |
| **Number**        | Counts, money, measurements                           | Decimals welcome, with no rounding drift |
| **Boolean**       | A yes-or-no flag                                      | Cells accept **True** or **False**       |
| **Date and Time** | A moment in time                                      | Stored with its timezone                 |

{% hint style="warning" %}
**A column's type is fixed when it's created.** There's no way to change it afterward, because converting a populated column could invalidate data that's already there. If you picked wrong, add a new column with the right type and move the data across.
{% endhint %}

Typed cells tell you what they expect when you edit them: a Number column hints *Numbers only*, a Date and Time column *Dates and times only*, and a Boolean *Allowed values: True, False*.

{% hint style="info" %}
**AI columns are always Text.** A model needs to be able to answer "unable to determine" for any question, and that has to be storable. When you toggle **AI Column** on, the type picker locks to Text.
{% endhint %}

## AI-Enabled Columns

AI-enabled columns are the core of what makes Lists powerful. When a row is processed, the assigned AI agent uses its tools — web search, webpage visits, tech stack inspection, CRM lookups — to research and fill each AI-enabled column.

### How AI Columns Work

1. You create a column and toggle **AI Enabled** on
2. You write a **Prompt Template** that tells the agent what to research for that column
3. When rows are processed, the agent executes the prompt for each row and writes the result

### Writing Effective Prompt Templates

The prompt template is the instruction your AI agent follows when filling the column. Be specific about what you want and the expected format.

{% hint style="info" %}
**Pro Tip**: Reference other columns in your prompt to give context. For example: *"Research the company at {website} and provide a 2-3 sentence overview of their core business, target market, and key differentiators."*
{% endhint %}

**Good prompt examples:**

| Column Name         | Prompt Template                                                                                     |
| ------------------- | --------------------------------------------------------------------------------------------------- |
| Company Overview    | Research the company and provide a concise 2-3 sentence overview of their business model and market |
| Decision Makers     | Find the top 3 decision makers (C-level or VP) at this company with their name, title, and LinkedIn |
| Tech Stack          | Inspect the company's website and identify key technologies, frameworks, and cloud providers used   |
| Qualification Score | Based on all available data, rate this lead 1-10 for fit with our ideal customer profile            |

**Avoid vague prompts:**

| Bad Prompt             | Better Alternative                                                                           |
| ---------------------- | -------------------------------------------------------------------------------------------- |
| "Research the company" | "Provide a 2-3 sentence overview covering business model, founding year, and employee count" |
| "Find contacts"        | "List the top 3 C-level executives with name, title, and LinkedIn URL"                       |
| "Is this a good lead?" | "Score 1-10 based on: company size 50-500, B2B SaaS, and US-based. Explain your reasoning."  |

### Knowledge Base Integration

Some AI columns can be configured to reference your organization's knowledge base during processing. Toggle **Requires Knowledge Base** when the agent needs internal context — such as your product catalog, pricing tiers, or qualification criteria — to fill the column accurately.

## Creating Columns

{% stepper %}
{% step %}
**Open Your List** Navigate to the list where you want to add columns.
{% endstep %}

{% step %}
**Add a Column** Click the **+** button in the table header or use the column management panel. Provide a name and select a type.
{% endstep %}

{% step %}
**Configure AI (Optional)** Toggle **AI Enabled** and write a prompt template. Optionally enable **Requires Knowledge Base** if the agent needs internal data.
{% endstep %}

{% step %}
**Save** The column is added to your list immediately and will be included in future processing runs.
{% endstep %}
{% endstepper %}

## Column Library

DarcyIQ provides a reusable column library so you don't have to recreate common columns for every list.

### Library Levels

| Level              | Scope                                   | Who Can Create      |
| ------------------ | --------------------------------------- | ------------------- |
| **System Columns** | Pre-built columns available to everyone | DarcyIQ (read-only) |
| **Organization**   | Shared across your entire organization  | Admins and Owners   |
| **User**           | Personal columns only you can see       | Any user            |

### Using the Column Library

1. Open the **Column Selector** when editing your list
2. Browse available system, organization, and user columns
3. Select columns to add them to your list
4. Customize the prompt template or configuration for your specific use case

{% hint style="info" %}
**Pro Tip**: Create organization-level columns for standardized fields like "Company Overview" or "Qualification Score" so every team member uses the same enrichment prompts.
{% endhint %}

## Managing Columns

### Column Operations

| Action        | Description                                                                                      | Permission Required |
| ------------- | ------------------------------------------------------------------------------------------------ | ------------------- |
| **Add**       | Add a new column or select from the library                                                      | Editor              |
| **Edit**      | Update the name, the AI toggle, and the prompt. Type is shown but can't be changed               | Editor              |
| **Reorder**   | Drag columns to change their display order                                                       | Editor              |
| **Hide/Show** | Toggle column visibility without removing data                                                   | Editor              |
| **Remove**    | Remove the column and its data from this list. It stays in your column library for use elsewhere | Editor              |

### Column Visibility

Use the **Column Visibility Selector** to show or hide columns in your table view. Hidden columns retain their data and will still be processed by AI — they're just not displayed in the table.

### Pinning Columns

Wide lists scroll horizontally, which is a problem when the column that tells you *which row you're looking at* scrolls off the left edge. Pin it instead.

Each column header has a **pin** control — the icon toggles between **Pin** and **Unpin**. A pinned column stays in place while the rest of the table scrolls, so the company name stays beside the enrichment you're reading.

Pins are remembered **per person, per list**: they survive sorting, filtering, and reloading the page, and yours don't change anyone else's view. Your first column starts out pinned, and you can unpin it like any other.

## Column Configuration Fields

| Field               | Description                                                                        | Required |
| ------------------- | ---------------------------------------------------------------------------------- | -------- |
| **Name**            | Display name shown in the table header                                             | Yes      |
| **Type**            | Text, Number, Boolean, or Date and Time. Fixed once created                        | Yes      |
| **AI Enabled**      | Whether the agent should fill this column during processing                        | No       |
| **Prompt Template** | Instructions for the AI agent when populating the column                           | If AI    |
| **Requires KB**     | Whether the agent should reference the knowledge base                              | No       |
| **Description**     | Help text or notes about the column's purpose                                      | No       |
| **Mandatory**       | Whether the column must have a value. Set when a list or template is first created | No       |
| **Visible**         | Whether the column is shown in the table by default                                | No       |

## Best Practices

1. **Get types right the first time**: Type can't be changed later, so decide up front whether a column is a Number, a Date and Time, or just Text
2. **Name columns descriptively**: "Primary Decision Maker" is better than "Contact"
3. **Limit AI columns to 3–5**: More AI columns means longer processing time per row
4. **Order matters**: Place identifying columns (name, website) first, AI columns after
5. **Use the library**: Save time by reusing organization-level columns across lists
6. **Test prompts early**: Process a few rows to validate your prompt templates before running the full list
7. **Combine manual and AI**: Let AI handle research while you provide the seed data


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.darcyiq.com/organize/lists/columns.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.