Product Attributes overview

Edit on GitHub
You are browsing a previous version of the document. The latest version is 202212.0.

Creating and managing product attributes is the first step in enriching your product data. It is an essential part of maintaining and improving the online shop for your customers. Attributes can serve different purposes and are therefore stored in different ways. They may be used not only for product management, but also to improve search and filter functionality in the shop.

You can easily build a list of characteristic attributes for your products, such as brand or special features. For these you can define specific values to help you and your customers distinguish between products. All products can get assigned a multitude of attributes to simplify the filter and category functions. The Spryker Commerce OS offers a set of pre-defined Product Attribute values or offers to simply import your own sets.

  • Any attribute types and values can be imported (no pre-defined schema needed)
  • Optional pre-defined values
  • Create and manage attributes with predefined values for usage in the shop

One way to describe a characteristic of a product is the attribute mechanism. It allows to store a list of key value pairs. There are 4 possible locations where attributes can be stored.

Common attributes to all variant are stored on the level of the abstract product.

However if they contain language or locale specific data, the localized abstract product attributes are the right location.

If attributes like the size of the t-shirt distinguish variants, they are stored on product level.

Again if the attributes are language or locale specific they are stored on localized product attributes.

The attribute mechanism is only recommended if the information stored is for display purpose only. As soon as you want to program against it, we recommend extending one of the four tables.

Attribute Key Metadata

Redis can work with key value pairs as defined in the product attributes tables. However elastic search needs more information about the attributes semantic. Is the content of an attribute part of the fulltext search or is it used to sort the attribute. Therefore needed metadata is stored in the attribute meta data tables. These are used by the front end exporter to control which attributes of the product is stored where in the elastic search structure.

Product Attribute Management

In Zed, product management attributes support creating and editing products. You can add keys for your attribute name and values, enabling the translation to different languages (localized). Multiple ways of handling translations are offered. The following examples highlight the use cases.

Example 1 for localized attributes and non-localized values

In a shop with two languages Product Y exists with an attribute “weight” and an attribute values is “63.5 g”. Translations are only necessary for the attribute name and not the attribute value in this case.

In the DB a key “attribute.weight” is saved. The key can now be translated into different languages via the glossary:

Key = attribute.weight Value (en_EN) = Weight Value (de_DE) = Gewicht

The attribute value “63.5 g” exists as a value-key and serves as a key and a value at the same time in the DB:

Value (non-localized)=63.5 g What you see on the English version of the shop: Weight = 63.5 g What you see on the German version of the shop: Gewicht = 63.5 g

Example 2 for localized attributes and values

In a shop with two languages Product X exists with an attribute “Protection Feature” and an attribute value “waterproof”. Both, attribute name and value need translations.

In the DB a key “attribute.protection_feature” is saved. The key can be translated into different languages via the glossary:

Key = attribute.protection_feature Value (en_EN) = Protection Feature Value (de_DE) = Schutzfunktion

The attribute value “waterproof” is now translated, a non-localized value (as in Example 1) is not sufficient:

Value (en_EN) = Waterproof Value (de_DE) = Wasserdicht

What you see on the English version of the shop: Protection Feature = Waterproof

What you see on the German version of the shop: Schutzfunktion = Wasserdicht

Example 3 for different localized attributes and values for different languages

Product Z has the attribute “length” and is sold in 3 countries, Germany, Ireland, and the United States. Due to the fact that the United States does not use the metric system, the attribute values should be displayed in feet in the US and in meters in Germany and Ireland. For this scenario there are two ways of doing it. One way 1 is to work with a non-localized attribute key for Germany and Ireland, and a localized key for the US, because (like in Example 1) two languages could share the same translation for the attribute. The other way 2 is to add localized attribute keys and values for each language.

(1)Non-Localized key and localized key for the US In the DB a key “attribute.length” is saved for non-localized values and a localized key “” for the translations in the metric system.

Key = attribute.length Value (en_IE) = Length Value (de_DE) = Länge

Key = Value (en_US) = Length

Now for each of the attribute keys we add the attribute values, which are 1.5 meter for the Ireland and Germany, and 4.92 feet for the US.

For the key “attribute.length” the value is: Value = 1.5 meter

For the key “” the value is: Value = 4.92 feet

(2)Localized keys for US, DE and IE In the DB three localized keys, namely “”, “”, and “” are saved.

Key = Value (de_DE) = Länge

Key = Value (en_IE) = Length

Key = Value (en_US) = Length

Now for each of the attribute keys we add the attribute values.

For the key “” the value is: Value = 1.5 meter

For the key “” the value is: Value = 1.5 meter

For the key “” the value is: Value = 4.92 feet

What you see on the Irish and German versions of the shop: Length = 1.5 meter Länge = 1.5 meter

What you see on the US version of the shop: Length = 4.92 feet

Attributes in the PIM

Translations for the attribute values in the PIM are optional and users may add new values in the process of creating a new product in the PIM. The input type ranging from text input to numerical values can be defined upfront. Adding new attribute values in the product creation process can be disabled, thereby restricting the user to the predefined value. All attributes that appear in the attribute management are called declared attributes.

What are “declared” attributes?

Declared attributes are the attributes that can be used in the PIM. There are two ways of declaring an attribute. On the one hand, when you create a new attribute it is automatically declared and can then be used for creating products in the PIM. On the other hand, you may import products with existing attributes in JSON format. The imported attributes are displayed in the PIM, but cannot be edited. Only after adding the exact same attribute key to the list of attributes in the attribute management, it becomes available in the PIM.

What are super attributes?

The Spryker Commerce OS product data model allows for the creation of multiple variants of one product. For the shop to be able to differentiate between these variants some characteristic needs to differ between them. A super attribute must be added to every product variant in order to display multiple variants of an abstract product in the shop frontend (Unless there is only one variant, which makes the use of super attributes for differentiation purposes obsolete). Attributes become super attributes by marking them as such in the attribute management interface. Important to note is that the variants do not need to share the same super attributes in order to belong to the same abstract product. Here is are two examples:

(1) Abstract product X has concrete product A and B: A: Super attribute = (color:blue) B: Super attribute = (color:green)

(2) Abstract product Y has concrete products C and D: C: Super attribute = (color:red) D: Super attribute = (size: 45)

Super attributes in the Spryker Commerce OS Demoshop

In this example on the product detail page there will be the possibility to select either the attribute color or the attribute size. For a variant to be selectable, it should have a unique combination of values for its super attributes compared to the other variants from the same abstract product.

Product attribute management

Current Constraints

Currently, the feature has the following functional constraints which are going to be resolved in the future.

When creating an abstract product, you can define as many super attributes as you want and this will create the same number of variants. You can also add a new variant to an existing product that already has other variants. However:

  • you can only use the existing super attribute in this case

  • if you want to add another super attribute to variants, you need to delete the abstract product with its concrete products and recreate it from scratch

You can define attributes by importing them or inside the Backoffice:

It is possible to import a product without having defined previously its attributes. However:

  • attribute values are then non-editable for the product since the Backoffice does not know how to manage them.
It is recommended to always define the attributes that your products will use.