Understand data transformation
Profile properties are flexible, super customizable profile attributes that can be used to create dynamic, personalized experiences. As you integrate more data sources with Klaviyo, it becomes more challenging to keep these data fields uniform in terms of format, capitalization, and abbreviations. Learn how to use the data transformation tool to maintain consistent, clean profile data which will help you build segments and personalize messages more efficiently.
Note: You will find data transformation in Advanced KDP in your own account under Data management > Transformation. This demo video shows the feature located under Klaviyo CDP. The location in the side panel is the only difference; all feature details and capabilities outlined in the video should match what you see in your account.
Set up 3 key transformations
Profile properties differ from account to account. The data in your Klaviyo instance will look entirely different from the data in another account. However, there are likely a few basic profile properties that you rely on very regularly for segmentation, targeting, and dynamic personalization. We recommend getting started with transformation rules for the key profile properties that you use most often:
Format First Name
Apply a formatting rule to the $first_name property that removes quotes, spaces at the front and end of the value string, removes special characters, and capitalizes the first letter of the name.
Format Last Name
Apply a formatting rule to the $last_name property that removes quotes, spaces at the front and end of the value string, removes special characters, and capitalizes the first letter of the name.
Standardize a custom property
Select a custom property that you use regularly for segmentation or dynamic content personalization. If you collect profile property information through sign-up forms or surveys, it’s highly likely that the values in those fields are pretty inconsistent. Here are a few examples:
- Shoe size: could be written “7” or “7.0”
- Pet type: could be written “Dog”, “dog”, or “puppy”
- Do you live in an apartment? (or any other true/false question): answer could be “True”, “true”, “yes”, “Yes”, or “1”
Set a standardization rule that looks for any applicable value, and replaces it with the standardized version code. The example above shows a standardization rule for the $gender property.
Adjust segments and templates that utilize those properties
Once you have transformed your frequently-used custom properties, revise any segment definitions that included the original version of the property and switch the criteria to include the transformed version instead. You should see an increase in segment size once you start using the transformed property that more accurately aggregates profiles.
Do the same for any saved email templates in your account that may include properties like First Name, or dynamic display logic based on transformed properties. Use only the transformed versions of these properties moving forward.