Adding an Input Feed
Table of Contents
To add a new source file, go to Feed Settings in the left menu, then click the Add New Source File button. A two-step wizard will open:
- Step 1 — Input Data: specify where Sembot should fetch the file from
- Step 2 — Field Mapping: indicate how the columns in the file should be interpreted by Sembot
Step 1 — Input Data
Source Name
The first field is Source Name — a label visible in the list of source files. It helps you identify files when a project has multiple ones (e.g., Main Feed PL, Promotional Feed, Supplementary Feed EN). The name is internal only and does not appear in advertising channels.
Import from URL (HTTP, FTP)
The most common method. In the File URL field, paste a direct link to the feed file — Sembot will fetch it automatically according to the configured schedule.
Supported protocols:
- HTTP / HTTPS — standard web address to the file (e.g.,
https://shop.com/feed.xml) - FTP — FTP server address (e.g.,
ftp://ftp.shop.com/feeds/products.csv)
Important — IP addresses for whitelisting: Some platforms (e.g., IdoSell) block downloads from external servers. If the file is not publicly accessible, you need to add Sembot's IP addresses to the exceptions list on your server or in your e-commerce platform's control panel:
52.28.49.250,51.77.44.22,51.77.44.187,51.38.111.87,51.38.107.36If there are any issues, the form will display this message automatically.
If access to the file is password-protected, expand the Advanced section and fill in the Login (file access) and Password (file access) fields. This data is encrypted and used solely for authorization during download.
Import from File (CSV, XML)
Sembot supports three source file formats:
| Format | When to Use |
|---|---|
| XML | The most common product feed format — exported by most e-commerce platforms (Shoper, IdoSell, WooCommerce, Magento) |
| CSV | A text file with data separated by a delimiter — common in exports from ERP systems or spreadsheets |
| Google Sheet | A public Google Sheets spreadsheet with product data — only publicly accessible sheets (no login required) |
Sembot does not support JSON as a direct source file format. If data is only available through a JSON API, use the API Connection option described below.
Selecting the File Format
The File Type field in the Advanced section (hidden by default) lets you choose the format:
- Auto-detect (default) — Sembot will automatically recognize the format based on the file structure. This is sufficient in most cases.
- XML — force parsing as XML
- CSV — force parsing as CSV
- Google Sheet — for public Google Sheets spreadsheets
CSV Format Options
When the file format is CSV (or Auto-detect for a CSV file), two additional fields are available:
CSV Column Separator — the character separating columns in the file:
| Option | Symbol | When to Use |
|---|---|---|
| Tab | TAB | Exports from Excel, ERP systems |
| Semicolon | ; | Polish and European files |
| Comma | , | English-language and US files |
| Pipe | | | Non-standard formats |
CSV Value Qualifier — the character surrounding values that contain the separator or special characters:
"(double quote) — default, most common'(single quote) — less common
API Connection
Instead of providing a URL, you can select a pre-configured Connection set up earlier in the Integrations section. Click the Select Connection Source field — a list of active connections assigned to the project will appear (e.g., Shopify, custom store API).
This option is an alternative to the URL — one field excludes the other. The separator between options in the form is labeled OR.
An API connection is useful when:
- The e-commerce platform does not generate a static feed file but provides data through its own API
- You want to use an OAuth-authenticated connection configured in Integrations (instead of a login and password in the URL)
- You are fetching data from a platform for which Sembot has a dedicated integration
If the required connection is not on the list, configure it first in Workspace Settings → Connections, then return to adding the feed.
Supplementary Feed
The Supplementary Feed checkbox indicates that the file being added is not the main data source, but extends or overrides selected attributes of the main feed. Common use cases:
- A separate file with marketing descriptions (when store descriptions are too technical)
- A file with additional product images
- A spreadsheet with promotional or seasonal prices overriding prices from the main feed
When the checkbox is unchecked, the feed is treated as the Main feed.
Proceeding to Mapping
After filling in the form, click Set Mapping — Sembot will fetch the file in the background and proceed to step two. Fetching may take a moment for large files.
Step 2 — Field Mapping
Mapping is the step where you tell Sembot how to interpret the columns in your file. Sembot has already read the file structure and displays a table with the detected columns.
How Mapping Works
Each row in the table represents one association: a column from your file → an attribute in Sembot.
The table has four columns:
| Column | What It Shows |
|---|---|
| Original Source Attributes | Column names detected in your file (e.g., product_name, ean, price) |
| → | Direction of mapping |
| Sembot Attributes | Standard Sembot attributes to which you map the data (e.g., Title, EAN, Price) |
| Sample Values | The first value from the given column in the fetched file — a preview of real data |
Data Preview Before Saving
The Sample Values column serves as a preview — before you save the mapping, you can see what is actually contained in each column of your file. This allows you to verify that the price column contains numbers, ean contains 13-digit codes, and image_link contains valid URLs.
This is the only point in the process where you can assess the quality of the source data before it is saved in Sembot.
Mapped and Unmapped Attributes
Sembot automatically attempts to match source columns to its own attributes based on column names. Results are shown in two sections:
Mapped Attributes — columns that Sembot recognized and matched automatically. You can change them by expanding the dropdown for a given attribute.
Unmapped Attributes (visible only when editing an existing feed) — columns from the file that have not yet been assigned to any Sembot attribute. A warning icon is displayed next to the name. Clicking the + icon next to an unmapped attribute moves it to the mapped section.
Message for unmapped attributes: "These attributes have not been mapped yet. Click the plus icon to add them to the mapping."
Creating Custom Attributes
If your file contains custom columns that are not on the list of Sembot attributes (e.g., collection, sole_material), you can add them as custom attributes. The Create Sembot Attribute Field option will appear in the Sembot attribute selector — selecting it will register the new attribute, making it available in product rules and elsewhere.
Similarly, if a column in the file has a custom name that is difficult to identify, you can use the Create Original Attribute Field option.
Saving the Mapping
After configuring all associations, click Save New Input Channel. Sembot will confirm the mapping and begin the first data processing run.
Confirmation Page
After saving the mapping, Sembot displays a confirmation page with the message:
"Feed added successfully. Your feed file is now being downloaded. This process may take a few minutes."
From this screen, you can proceed to the next steps:
| Button | What It Does |
|---|---|
| Back to Feed Page | Returns to the project's list of source files |
| Add Another Output Feed | Opens the output file wizard (e.g., for Google Shopping) |
| Add New Input Feed | Starts the process of adding another source file from the beginning |
| Project List | Goes to the list of all projects |
| Create Campaign | Goes to creating an advertising campaign |
| Connect Google Account | Goes to configuring the Google Ads connection |