> ## Documentation Index
> Fetch the complete documentation index at: https://docs.retailgrid.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Create grid

> Create a new grid by selecting products from your catalog to build a focused pricing and analysis workspace.

A Grid in Retailgrid is your focused pricing workspace — a selected set of products where you analyze, optimize, and execute pricing decisions.

## What is a grid?

A Grid is:

* A filtered subset of your product catalog
* A working environment for pricing and analysis
* A place to run agents, apply rules, and export results

You can create multiple grids for:

* A specific brand (e.g., *Arla*)
* A category (e.g., *Dairy*)
* A campaign
* A pricing experiment
* A country or channel

## Step 1. Go to grid

From the main navigation:

* Click **Grids**
* Click **+ New** (top right)

You'll enter the **Manual grid** creation screen.

## Grids view

<Frame>
  <img src="https://mintcdn.com/retailgrid/yby33A9XHZVL6uVo/images/grids/grids_view.png?fit=max&auto=format&n=yby33A9XHZVL6uVo&q=85&s=30b3738460399b7b30fc4377538f1ef0" alt="Grids dashboard with existing grids and the New button in the top-right" width="2880" height="704" data-path="images/grids/grids_view.png" />
</Frame>

## Step 2. Name your grid

At the top:

* Enter a clear, descriptive name\
  Example:
  * `Arla`
  * `Dairy – Margin Optimization`
  * `March Price Test`

You'll see a live counter showing how many products are selected.

## Name your grid

<Frame>
  <img src="https://mintcdn.com/retailgrid/b03Y19EUSCCXDNFG/images/name-grid-and-select-products.png?fit=max&auto=format&n=b03Y19EUSCCXDNFG&q=85&s=7621f4596781d0ebf3a635130d5407b2" alt="New-grid creation screen with the grid name field" width="2738" height="1122" data-path="images/name-grid-and-select-products.png" />
</Frame>

## Step 3. Search and filter products

Use filters to define your working scope:

* **Search products** (by name or ID)
* **Brand**
* **Product type**
* **Status**

Example:

* Filter by **Brand = Arla**
* Select all filtered products

The product counter updates in real time.

## Filter products

<Frame>
  <img src="https://mintlify.s3.us-west-1.amazonaws.com/retailgrid/images/image-11.png" alt="Filter products step in the new-grid flow" />
</Frame>

## Step 4. Select products

You can:

* Select individual SKUs
* Select all filtered results
* Clear selection

This defines exactly which products will be included in your grid.

## Step 5. Click "Proceed"

After selecting products:

* Click **Proceed**
* Retailgrid creates your Grid
* You are redirected to the grid view

## Step 6. Start working in your grid

Your new grid opens in spreadsheet-style grid format.

From here you can:

### Analyze

* Sort and filter
* Compare price vs cost
* Review margin gaps

### Modify

* Edit prices directly
* Add calculated columns
* Apply bulk updates

### Execute

* Use **Agents**
* Run pricing rules
* Export data
* Prepare price updates

## Grid is ready

<Frame>
  <img src="https://mintlify.s3.us-west-1.amazonaws.com/retailgrid/images/grids/grid-spreadsheet.png" alt="Grid open in spreadsheet view with toolbar, columns, and product rows" />
</Frame>

## Change which products are in the grid

The selection you made in Step 3 isn't permanent. To add or remove products from an existing grid, click **Edit scope** in the top-right corner of the grid view and adjust your filters or selection.

## How your grid stays fresh

Every grid is refreshed automatically overnight with the latest data from your datasets. You don't need to trigger a manual refresh after replacing a file — the next night's run picks it up.

* **Nightly refresh** — runs once per day during an off-hours window and covers every active grid.
* **On-demand updates** — replace a dataset on the [Datasets page](/data-requirements/datasets) at any time; the change becomes visible in your grids on the next nightly refresh, or sooner if you manually run an agent against the grid.
* **Per-grid status** — failed refreshes are visible in the [Runs](/runs/overview) section, where `system`-triggered nightly jobs appear alongside user-triggered runs.
