If your company doesn’t already create briefing data for campaigns, you might not know where to start and how to extract the right products from your product feed.
Similarly, if you want to run more campaigns than what is available in your company's existing briefing data, you would need a way to select relevant products directly from your feed.
Why Horizon needs a briefing file
A briefing file defines which products should be in your Horizon catalog and is a subset of the products in your product feed.
You can learn more about briefing data here:
How to easily create briefing data from your feed
To help you create specific briefing files quickly and easily, we have created the Google Feed Filter tool, which helps you quickly extract and prepare briefing data to create digital catalogs with Horizon.
The output is "just" a briefing file with a list of product SKUs, but the Google Feed Filter is a powerful and easy way to get that list. And then, you simply upload that file to Horizon.
🚨Important: Google Feed Filter support
The Google Feed Filter is made available to support your workflow, but it's not an officially supported iPaper tool.
This means that, while the Google Feed Filter can be very useful, it is provided as-is and iPaper does not guarantee its performance or accuracy. We do not take responsibility for any bugs, incorrect results or issues that may arise from using it.
📖 This article contains...
What is the Google Feed Filter?
Google Feed Filter is a browser-based tool that lets you extract a targeted subset of products from a Google Shopping product feed and export them into a pre-defined Excel file (Briefing File).
Instead of manually searching through thousands of rows, you define one or more filter combinations (brand, product type, and sale status), and the Google Feed Filter will do the heavy lifting and creates a briefing file.
Key capabilities of the Google Feed Filter:
Accepts CSV, XML, and TSV product feed formats
Supports multiple simultaneous filter groups (OR logic between groups)
Smart brand ↔ product type cross-filtering so dropdowns only show relevant options
Sale / regular price filtering based on feed price fields
Live product count preview before you export
Column mapping lets you populate additional Briefing File columns from your feed
Exports directly to a structured .xlsx file with a date-stamped filename
Common use-cases for the Google Feed Filter
There are many ways you can use the Google Feed Filter to create specific briefing data for Horizon catalogs. For example...
On-sale products (weekly or monthly sale)
Specific brand campaigns (Nike shoes, Samsung TVs etc.)
Product type (groceries, coffee machines, etc.)
Getting started
What you need
A product feed file in one of the supported formats: .csv, .xml, or .tsv
A modern web browser (Chrome recommended)
💡 Note: The tool runs entirely in your browser. No data is sent to any server. Your feed file stays on your machine.
Step-by-step workflow
There are just 5 steps from feed to finished briefing file.
Step 1: Upload your feed file
1. Click the "Choose CSV, XML, or TSV file" button under the Upload Feed File section and select your product feed from your computer.
The Google Feed Filter automatically detects the file format and parses it.
Once the file is loaded, a confirmation message appears showing the file name and number of products detected.
Step 2: Configure Filter Group 1
A 'Filter Group' lets you target a specific combination of products based on either Brand, Product Type or Sales. It's how you select the group if products that should be exported from your product feed.
The Brand and Product Type dropdowns are populated automatically from the feed data.
Tip: You can leave either Brand or Product Type empty. Leaving both empty in a group means all products pass through that group.
Per default, Filter Group 1 is shown and contains three filter controls:
Brand Filter
Click the Brand Filter field and start typing to search, or click to open the full dropdown list. Select the brand you want to filter by. If you have already selected a Product Type, the brand list will only show brands available within that product type.
Product Type Filter
Works the same way as the Brand Filter. The dropdown is populated from the product_type or category field in your feed. Hierarchical product types (e.g. Clothing > Shoes > Trainers) are broken into individual levels so you can filter at any level of the hierarchy.
💡 Note: The "section" column in the exported file is always filled with the final segment of the product type path. For example, "Clothing > Shoes > Trainers" becomes "Trainers".
Sales Filter
Choose one of three options:
All Products — no price filtering applied (default)
On Sale — includes only products where a sale_price field exists and is lower than the regular price
Regular Priced — includes only products where no valid sale price is found
💡 Note: Sale detection works by comparing price against sale_price in the feed. If either field is missing or empty, a product is treated as regular priced.
Step 3: Add more filter groups (Optional)
To capture products from multiple brand/product type combinations in a single export, click the "Add Filter Group" button. Each new group gets its own Brand, Product Type, and Sales Filter controls.
Filter groups are combined using OR logic: a product is included in the export if it matches any one of your filter groups. This means you can, for example, capture "Nike Shoes" in Group 1 and "Adidas Trainers" in Group 2 in a single export.
To remove a filter group, click the × button in the top-right corner of that group card.
Step 4: Review the number of products to be exported
As soon as you select any filter, a live preview bar appears above the column selector showing:
A count badge with the total number of matching products
This updates in real time whenever you change a filter, so you can confirm your selection makes sense before exporting.
Step 5: Export
Click the "Process & Export" button at the bottom of the form. The tool filters the feed data according to all active filter groups, builds the Briefing File structure, and downloads the file automatically.
The downloaded file is named:
Filtered_Products_YYYY-MM-DD.xlsx
A results summary appears on screen showing the original product count, the filtered product count, and which filter groups were active.
💡 Note: product_id and offer_id columns are exported as text (not numbers) to prevent Excel from stripping leading zeros from long numeric IDs.
Tips & troubleshooting
Dropdowns show no options after upload
This usually means the tool could not find a recognised brand or product_type column in your feed. Check that your file has a column named brand (or similar) and one named product_type, category, or a Spanish equivalent. If you are using a custom field name, the tool may not pick it up automatically.
The Sales Filter shows no "On Sale" products
The tool compares a sale_price field against the price field. If your feed uses different column names, the sale detection will not work and all products will appear as regular priced. Ensure both columns exist in your feed with those exact names.
Exported file has empty columns
Only product_id and section are populated automatically. All other Briefing File columns (header, image_url, offer_url, etc.) remain empty unless you map them in the "Select Additional Columns" section before exporting.
The file fails to parse
If the tool shows an error after upload, verify that:
The file extension matches the actual format (.csv, .xml, or .tsv)
The file is not password-protected or encoded in a non-UTF-8 character set
The XML file does not have a parse error (the tool will display the XML error message)
The file contains at least two rows (a header row and one data row)
Larger feeds are slower to load
The tool parses the entire feed in the browser. Very large files (hundreds of thousands of rows) may take a few seconds.
Unpopulated columns may contain invisible data
If your exported briefing file does not work correctly in Horizon it may be due to invisible data in the file. Mark every empty cell in the briefing file and press delete before uploading again.