Skip to content

Product feed format: CSV, XML or Google Sheets?

A beginner-friendly guide to choosing the right catalog file, avoiding broken spreadsheets and preparing channel-specific feeds without rebuilding your product data.

J

Julia Villegas

· 9 min read · 318 views
Product feed format with items on CSV, XML and Google Sheets papers

Choosing the right product feed format helps you avoid problems before your catalog reaches Meta, Google, Pinterest or any other commerce channel. If your prices change often, your products have variants, or your team relies on shared spreadsheets, the format matters: it can make imports easier, break data quietly or leave products out.

What a product feed format is

A product feed is a file or URL that contains the data behind your catalog: product name, product page link, image URL, price, availability, brand, category, variants and any other fields a channel may require.

The product feed format is the way that data is written and structured. It does not change what you sell, but it changes how an advertising platform, marketplace or feed management tool reads your catalog.

The three common formats most teams consider first are:

  • CSV: a plain-text table where columns are separated by commas, semicolons or another delimiter.
  • XML: a structured file using tags, often better suited to technical catalogs and nested data.
  • Google Sheets: a shared spreadsheet that is easy for non-technical teams to view and edit.

You may also see TSV, JSON, ecommerce platform exports or URLs generated by an external tool. But if you are starting out, the practical question is usually simpler: should you maintain a spreadsheet, export a CSV or use XML?

There is one important caveat. A good format does not replace good product data. Pinterest’s catalog guidance, for example, points to requirements around business accounts, websites, product data and commercial availability before you get started with catalogs. In other words, your product, link, price, stock and business conditions still need to make sense for the channel.

Product feed format: CSV, XML and Google Sheets compared

There is no universal best format. A small ecommerce team may work perfectly well from a spreadsheet at the beginning. A store with thousands of SKUs, daily stock changes and many variants will usually need something more robust.

Format Best fit Strengths Common risks
CSV Mid-sized catalogs, ecommerce exports, recurring imports Lightweight, easy to generate, widely supported Wrong delimiters, shifted columns, character encoding issues
XML Large catalogs, variant structures, technical feeds Highly structured, good for complex or nested data Harder to edit manually, tag errors, more technical control needed
Google Sheets Small catalogs, tests, manual campaign selections Easy to read and edit, no file export needed each time Accidental changes, broken formulas, permissions, slow large sheets

The question is not only “which format does the channel accept?” It is also “which format can you maintain without breaking it?” A feed that is valid today can become invalid tomorrow if someone renames a column, enters a price as text or deletes a variant row.

CSV: simple, fast and easy to automate

A CSV looks like a spreadsheet, but it is really a text file. Each row usually represents a product or product variant, and each column represents an attribute.

Simplified example:

id title link price availability
SHIRT-001 Blue Oxford shirt https://store.example/blue-shirt 39.90 USD in stock
SHIRT-002 White Oxford shirt https://store.example/white-shirt 39.90 USD out of stock

CSV works well when your store can export it consistently or when a feed tool can generate an updated URL. It is lightweight and easy for systems to process, which is why many platforms support it.

Most CSV issues come from details that look small but affect the whole file:

  • The price is written as 39,90 when the channel expects 39.90 USD.
  • The delimiter changes from comma to semicolon.
  • A product title contains quotes or line breaks and shifts columns.
  • Character encoding breaks accents, symbols or special characters.
  • A variant loses its stable ID and the channel reads it as a new product.

If you use CSV, avoid editing the final file by hand except for controlled tests. It is safer to generate it from the same source each time and review errors after major catalog changes.

XML: more structure for complex catalogs

XML uses tags to describe each piece of data. It is not as easy to scan as a table, but it can represent more structured information. For example, one product can contain blocks for attributes, images, category paths or related variants.

XML often makes sense when your catalog already comes from a technical system, ERP, custom integration or ecommerce setup with many attributes. It can also help when you need a clear relationship between parent products and variants.

Its advantage is consistency. If the XML is well built, every field has a defined place. Its downside is that it is not friendly to manual editing. A missing closing tag, an unescaped character or an incomplete block can make the file unreadable.

For beginners, XML usually makes sense when your system already generates it. If you are trying to build XML manually from a spreadsheet, you may be adding complexity before you need it.

Google Sheets: convenient to start, fragile at scale

Google Sheets is attractive because almost anyone can open it, inspect products and correct a cell. For an initial test, a small catalog or a one-off campaign, that can be enough.

It is also useful when marketing needs to prepare a specific product set without asking a technical team for a new export. For example, a sheet with 80 products selected for a holiday gift campaign, with reviewed titles and internal labels.

But the convenience has limits:

  • Someone can rename a column without realizing the feed depends on it.
  • A formula can return empty cells.
  • Sharing permissions can prevent a platform or tool from reading the sheet.
  • The sheet can grow too large and become slow.
  • Automatic formatting can change product IDs, prices or dates.

Google Sheets is not a bad option. It simply needs discipline. If you use it, protect critical columns, document who can edit it and avoid making it the only source of truth for a large or fast-changing catalog.

Fictional example: one store, three feed needs

Imagine an online home decor store with 4,000 products. It sells lamps, textiles and small furniture. Stock changes every day, the team runs seasonal campaigns, and some products have color variants.

A practical setup could look like this:

  1. Main catalog from the ecommerce platform: imported from the store or from a file generated automatically.
  2. Feed for Google Shopping: complete product fields, consistent prices, availability and channel-specific required attributes.
  3. Feed for social campaigns: products with strong images, available stock and creative-friendly fields.
  4. Campaign spreadsheet: a small Google Sheets selection for “gifts under $50” reviewed by the marketing team.

In this case, using one manual spreadsheet for everything would be fragile. A stable CSV or XML can feed the base catalog, while a spreadsheet can support a specific campaign selection. Then each channel should receive its own output feed with the right fields, rules and product set.

This is an important principle: one catalog can power several feeds, but those feeds do not have to use the same format, contain the same products or expose the same fields. In most mature setups, you create channel-specific outputs from a reliable source.

How to choose a product feed format without breaking the catalog

Before deciding, answer these questions:

  1. How many products and variants do you have?
    If you have a small catalog and occasional changes, a spreadsheet may be enough to start. If you have thousands of SKUs or many variants, use a more stable source.

  2. How often do price and stock change?
    The more frequently they change, the less you should rely on manual edits. You need a source that updates in a controlled, repeatable way.

  3. Who maintains the catalog?
    If marketing owns the workflow, Google Sheets is easy but risky. If the store platform or a technical team generates the feed, CSV or XML will usually be more consistent.

  4. What does each channel require?
    Channels do not all accept the same fields, values or validation rules. Check their specifications and ingestion errors before assuming one file will work everywhere.

  5. Do you need different feeds by country, campaign or channel?
    If you sell in several markets or run separate campaigns, create sets and outputs instead of duplicating spreadsheets.

A simple rule: choose the format you can generate repeatedly and validate before sending. Ease of editing should not matter more than reliability.

How to do it with AdMotion

AdMotion lets you start from different catalog sources without deciding everything manually from day one. You can connect a store such as Shopify, BigCommerce or Tiendanube, or use a file or existing feed, then turn it into a catalog base you can clean, segment and publish.

A typical workflow with Feed Manager looks like this:

  1. Connect the source
    You can start from a store, a file, a hosted feed or a manual upload if that fits your operation. The source should contain coherent product data: ID, title, link, price, availability, images and any additional fields you need.

  2. Map the fields
    Mapping means telling the system which source column or attribute should become each catalog field. For example, name may become title, sale_price may become promotional price, and inventory may become availability.

  3. Normalize with rules
    Rules help you clean and transform data. You might exclude products without images, convert an internal stock value into in stock, add a label to products in a specific category or fix price formatting. In AdMotion, rules can also be written in natural language and reviewed in the editor before they are applied.

  4. Create product sets
    A set is a subset of the catalog. You can separate products by channel, campaign, country, category or stock status without duplicating files.

  5. Generate the output feed
    With Export Channels, you can prepare feeds for Meta, Google Shopping, Google Dynamic Feed, TikTok, Pinterest, Snapchat, Criteo, Microsoft Advertising, Thribee, ChatGPT or a custom channel. Each output can have its own format and fields based on the channel specification.

  6. Review errors and warnings
    The analyzer checks each feed against the channel specification. This helps you catch issues before adding the feed URL to the ad platform, but it does not guarantee final approval. Each platform applies its own policies, account requirements, territory availability and ingestion process.

  7. Copy the feed URL into the channel
    AdMotion generates a feed URL on each download. You register that URL in the relevant channel. Keep one detail in mind: the frequency at which data is imported into your feed workflow and the frequency at which the channel reads the feed are different. You should not assume instant synchronization inside the advertising platform.

If you also work on visual catalog ads, the Templates Editor can generate one image per product using catalog fields such as image, title, price and brand. But for feed format decisions, the first priority is making sure clean data goes in and channel-specific data comes out.

Common mistakes when choosing a format

Treating Google Sheets like a product database

A shared spreadsheet is useful, but it is not a stable system when you have many SKUs, variants or frequent changes. If several people edit it without controls, you can end up sending products with missing prices, broken links or empty columns.

Assuming one feed works for every channel

One catalog can feed multiple channels, but each channel may request different fields, names, formats or accepted values. Trying to maintain one “universal” file often creates exceptions and hard-to-debug errors.

Confusing import with publication

Importing a catalog into a tool does not mean the channel has approved it or refreshed it already. You still need to check ingestion errors, policies, territory availability and commercial requirements.

Editing the final file by hand

If you manually fix the final CSV or XML, the next export can overwrite your changes. It is better to fix the source, the mapping or the rule that generates the feed.

What to do this week

  1. Find your main product source
    Decide which source is reliable: your ecommerce platform, ERP, recurring CSV, XML file or controlled spreadsheet. Avoid maintaining contradictory versions of the same catalog.

  2. List your critical fields
    At minimum, review ID, title, link, image, price, availability, brand, category and variants. Add any channel-specific fields required by your target platforms.

  3. Choose by maintenance, not convenience
    If the catalog changes often, prioritize automatically generated CSV or XML. If it is a small test, Google Sheets can work, but protect permissions and critical columns.

  4. Separate feeds by channel or campaign
    Do not force one file to do everything. Create different product sets if you need to exclude out-of-stock products, split countries or prepare campaign-specific selections.

  5. Validate before uploading or refreshing
    Review errors and warnings before copying the feed URL into a channel. A feed manager can help you map, normalize, segment and check the feed before activation.

Choosing between CSV, XML and Google Sheets is not an isolated technical decision. It affects how you maintain prices, stock, variants and campaigns. Start from a reliable source, then generate the outputs each channel actually needs.

Frequently asked questions

What is the best product feed format for beginners?

If you have a small catalog and only need a controlled test, Google Sheets can be the easiest place to start. If your catalog changes often or includes many variants, a CSV or XML generated from a stable source is usually safer. The best format is the one you can maintain, validate and update without editing the final file manually.

Can I use the same product feed for every channel?

You can start from the same catalog, but sending the same feed everywhere is not always the best approach. Each channel may require different fields, formats, values and validation rules. A safer workflow is to keep one main catalog source and generate channel-specific feeds by market, campaign or product set.

Does AdMotion automatically send my feed to every channel?

AdMotion generates feed URLs for different channels and lets you review feeds against channel specifications, but you still register the URL in each platform. The catalog import frequency and the channel’s own feed reading frequency are different, so you should not assume instant synchronization or automatic approval by the channel.

Sources

  1. Feed Manager de AdMotion — AdMotion
  2. Export Channels de AdMotion — AdMotion
  3. Primeros pasos con catálogos de Pinterest — Pinterest
Share:
J

Written by

Julia Villegas

Performance Marketing

Julia Villegas works in performance marketing at AdMotion.

Related posts