> For the complete documentation index, see [llms.txt](https://presight.gitbook.io/presight-v1.0/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://presight.gitbook.io/presight-v1.0/introduction/core-concepts/segments/source-columns.md).
# Source Columns
## **Columns and Values**
Columns originate from the attributes available in your raw data, usually in the form of categorical data fields such as **Region**, **Segment**, or **Age Group**. Each column serves as a category by which you can segment and analyze your data. The unique entries within a column are known as **values**.
For example, consider the column **City**. The values within this column could be **New York**, **London**, **Tokyo**, etc. These values represent the specific instances or categories within the **City** column.
### **Columns**
Columns are essential for organizing data and providing structure to your datasets. They define the variables or attributes that describe each data entry, from which you can perform detailed analyses, identify patterns, and uncover relationships between different data attributes.
Tables and their columns on Presight
For example, in a sales dataset, columns might include:
* **Product Category**: Electronics, Clothing, Home Goods
* **Region**: North America, Europe, Asia
* **Sales Representative**: Names of sales staff
* **Date of Sale**: Specific dates of transactions
Each of these columns allows you to slice your data in different ways to gain various insights.
### **Values Within Columns**
Values are the actual data entries within each column. They represent the specific details or categories under a column. In the **Region** column, values might be **East Coast**, **West Coast**, or **Midwest**. Analyzing metrics based on these values enables you to compare different segments of your data.
For instance, you might want to compare revenue across different regions or see how sales performance varies over time. By filtering and grouping data based on column values, you can generate reports and visualizations that highlight these differences.
***
## **Combining Columns and Values with Metrics**
#### **Making Metrics Meaningful**
Metrics such as revenue, units sold, or customer satisfaction scores provide quantitative information. However, without context, these numbers lack actionable insights. By associating metrics with specific columns and their values, you can contextualize the data to make informed decisions.
For example:
* **Revenue by Product Category**: Understand which categories generate the most revenue.
* **Units Sold by Region**: Identify which regions have the highest sales volume.
* **Customer Satisfaction by Service Representative**: Assess performance at an individual level.
#### **Practical Example**
When you query **Revenue by City New York this year**, you are essentially asking:
> "Give me the **sum** of all values in the **rev\_amount** column where the **City** column has the value **'New York'** and the transaction date falls within this year."
In this case:
* **City** is your **column**.
* **'New York'** is the specific **value** within that column.
By organizing data into columns and analyzing metrics within the context of specific values, you can derive more insightful and granular information. This approach allows for precise data analysis, enabling better decision-making based on detailed breakdowns of the data.
