Many stores keep every product fact in one place, the description box. Size, ingredients, care steps, warranty and country of origin all sit in a block of text. That works until you want to show the same detail in a table, filter products by it, reuse it on another page or give it to Google. Then someone has to read hundreds of descriptions and copy the facts out by hand.
Shopify has two tools for this: metafields and metaobjects. This guide explains both in plain words and how to plan your data. I read Shopify's custom data overview and options pages on 11 November 2026. They show no dates, so check the limits in your own admin.
Metafields and metaobjects, in one minute
Shopify's help page groups them as custom data.
- A metafield adds a piece of information to something Shopify already has, such as a product. Think of an extra labelled box on every product: material, fit, care steps.
- A metaobject is a standalone, structured entry that belongs to nothing in particular until you link it. The page's example is author profiles. You create one entry for each author, with a name, a photo and a short bio, and then reuse it wherever it is needed.
A simple rule: if the information describes one product, order or customer, use a metafield. If it is a thing in its own right that several products or pages share, such as a size chart, a brand story or a care guide, use a metaobject, and point a metafield at it.
Definitions: the part that keeps data tidy
Metafields can have a definition. Shopify describes a definition as a name, a namespace and key, an optional description, a type and optional validations. The type might be text, a file or a URL. Validations are rules such as a character limit or a minimum and maximum value. The definition also has options for where the data is available.
The help page names several kinds of metafield:
- Standard metafields are preconfigured by Shopify. Standard product metafields come in categories.
- Custom metafields are ones you define yourself.
- Unstructured metafields have no definition. They can be migrated to a definition later.
- App-owned metafields belong to an app.
Use a definition for anything you will rely on. A text box with a validation stops one person entering "100% cotton" and another "cotton 100 percent", and it gives the next developer a clear name to work with.
Metaobjects work the same way. You create a definition in Settings, then Custom data. It has a name, a type identifier and fields, each with a name, a description, a type and validations. One field is the display name. Field types include text, files, URLs, images and colours. You then create entries under Content, then Metaobjects.
What you can do with the data
The options page lists what each definition can be switched on for. For metafields it says:
- All types can be reached from your storefront, through Liquid or the Storefront API. Without that, a customer never sees the value.
- Some types can be used to filter and group data in Analytics.
- You can filter the product index by up to 50 metafield definitions, and the orders index by up to 5. Company and company location indexes also take up to 50.
- You can use them as conditions in collections, up to 128 product and 128 variant definitions.
- They can be used in the Customer Account API.
For metaobjects, the page lists storefront access, use as dimensions or filters in reports when a metafield on a product, variant, customer or order refers to them, an active or draft status, translation of entries and filtering in the metaobject index, up to 50 definitions. You can even publish entries as web pages, which needs storefront access. If you use Hydrogen, Shopify's page says to turn off Enable in Online Store.
You change these options under Settings, then Metafields and metaobjects.
Three small plans that pay off
These are examples I made up to show the thinking. Replace them with your own products.
- Care instructions. A clothing store creates a metaobject for each care guide, such as wool wash or silk wash, and a metafield on each product pointing to one. When a care rule changes, they edit one entry, not 300 descriptions.
- Specification table. An electronics store defines metafields for battery life, weight and warranty length, each with a validation. The theme shows them in a table. The same fields can be used as collection conditions.
- Brand or maker pages. A gift shop creates a metaobject for each maker with a name, photo and story, and publishes them as web pages. Products link to their maker.
How to plan your data
- List the facts customers ask about before they buy. Those are your first definitions.
- Mark the ones used on many products. They may deserve a metaobject.
- Decide the type and the rule for each: a number with a minimum, text with a length limit, a choice from a list.
- Choose clear names and write them down. A short document that lists each definition and where it shows on the site saves hours later.
- Decide who edits the data and who checks it, and agree a way to correct mistakes in bulk.
- Switch on only the options you need, such as storefront access or collection filters.
A good first afternoon: pick one product type, define three metafields with validations, fill them in for ten products, ask your developer to show them on the product page and check the result on a phone. If it works and the team finds the data easy to enter, extend it to the rest of the catalogue. If it does not, you have found the problem on ten products, not on three thousand.
Common mistakes to avoid
- Creating a field for everything. Every definition is something to fill in, check and translate. Start with the five facts customers use most.
- Free text where a choice would do. If a value should be one of five options, use a type that allows only those, so reports and filters group properly.
- Forgetting other languages. The options page says metaobject entries can be translated. If you sell in more than one language, plan the translation of this data from the start.
- Leaving unstructured metafields for ever. They work, but they have no definition, so nothing checks them. Shopify's page says they can be migrated to a definition later, which is a good clean-up job.
- Treating app-owned fields as yours. Metafields owned by an app belong to that app. Ask the vendor what happens to the data if you remove it.
What the help pages do not tell you
A metafield stores a value. It does not show it. The pages I read do not describe how a theme displays the data, nor any plan limit or price. In practice a theme developer needs to add the value to your product template, or you need a theme that already supports it. Ask whoever builds your theme how it will display the data, and test the result on mobile.
Structured data also helps outside your own site. Product feeds such as Google Merchant Center need consistent fields like colour and material. See our guide to fixing product data for Google Merchant Center.
ESTASOFT's web development page describes the theme and store work we do. To plan a data model for your catalogue, send a short brief on the contact page.
Sources
- Custom data overview (Shopify Help Center): metafields and metaobjects, definitions, standard and unstructured metafields, example uses; page shows no date (checked 11 November 2026)
- Custom data options (Shopify Help Center): storefront access, filtering, collection conditions, limits of 50, 5 and 128 definitions, web pages from metaobjects, Hydrogen note; page shows no date (checked 11 November 2026)