Consistency between different systems occurs in two categories:
Identifier Level - product identifier and variant identifier.
Differences in identification levels between systems make it difficult to connect data because it is impossible to attribute product sales to variant sales.
Example:
Google Analytics 4 - in the system configuration, we use
item_group_id. So we know that the Classic T-shirt sold 3 times and generated PLN 450 in revenue, but we do not know exactly which variants were sold, because our identifier in the system isitem_group_id: 1.
Google Ads - in the system configuration, we use
id, thanks to which we assign, for example, costs to a specific variant. Thanks to this, we know that:Classic T-shirt, white, size S >>
item_group_id: 1id: a SPENT: PLN 15Classic T-shirt, white, size M >>
item_group_id: 1id: b SPENT: PLN 2Classic T-shirt, black, size S >>
item_group_id: 1id: c SPENT: PLN 5
We have the most accurate information about the results at the last level of identification, i.e. the variant. If we want to combine data from Google Analytics 4 and Google Ads, it cannot be done in a simple way, we need to prepare this data and bring it to a common identifier, i.e. item_group_id, because we do not know how to break down PLN 450 of revenue into individual variants. Unfortunately, due to this situation, we lose data quality, because after transformations we will know that: Classic T-shirt, white, which has the product identifier item_group_id: 1 earned PLN 450 and spent PLN 22, but we will not know which of the variants is the most popular. If we had the same level, i.e. the variant level in each system, in our example Google Analytics 4 and Google Ads, our data would look like this, for example:
Classic T-shirt, white, size S >> item_group_id: 1 id: a SPENT: PLN 15, REVENUE: PLN 300
Classic T-shirt, white, size M >> item_group_id: 1 id: b SPENT: PLN 2 REVENUE: PLN 0
Classic T-shirt, black, size S >> item_group_id: 1 id: c SPENT: PLN 5 REVENUE: PLN 150
Thanks to which we know exactly which product is the most popular.
2. Identifier Type - SKU, id, EAN, MPN, etc.
The second important consistency of identifiers is their type. Different identifiers often occur:
product card number - often assigned automatically by the online store panel
warehouse number
accounting number
SKU
MPN
GTIN
The most important thing is to use one identifier within all systems. A bad situation is when:
Google Analytics 4 - uses the product card number
Google Ads - uses SKU
MetaAds - uses the warehouse number
In this case, it is not possible to combine data from these systems in a simple way. To do this, you need to perform additional mappings of identifiers from many systems. Identifiers in systems should be of the same type, for example:
Google Analytics - uses SKU
Google Ads - uses SKU
Meta Ads - uses SKU
A good practice is to assign an identifier to a product and use the same ID in every system, regardless of whether it is a store panel, accounting system, warehouse, marketplace or advertising system.