Configure the Identity System
Learning Objectives
After completing this unit, you’ll be able to:
- Discuss identity system configurations.
- Explore the basic identity types within the identity system.
- Explain how to set up Marketing Cloud Personalization.
The Identity System Configuration
Suppose a user visits your website to look at a product they saw in an email but enters your website’s address into their browser instead of clicking the link in the email. They look at a certain product but don’t enter any information into your website to identify their email address. Before leaving the site, a popup window appears asking them for their information, so they enter a first name, last name, and a phone number to be called back at a later time. How does Marketing Cloud Personalization stitch these profiles together?
Marketing Cloud Personalization uses an identity system, which is its method for defining a unique user profile across multiple datasets. This profile consists of a collection of selected attributes, any of which can identify an individual. Examples of selected attributes include Email Address, Salesforce Marketing Cloud contact key, and Salesforce CRM Contact ID. But this is not an exhaustive list. As events and interactions are received, identity attributes determine which new or existing user profile the event is recorded under and which profiles are merged, if any.
Who Configures Identity Implementation?
A tenant administrator is the only one allowed to access and configure identities in Identity Studio. The identity system is also composed of two feature flags, which are necessary for configuration. Some accounts might be missing a few of the options you see in this module. In that event, reach out to Marketing Cloud Personalization support to resolve the issue.
You should configure the identity system before collecting any data, including data from extract, transfer, and load (ETL) or your sitemap. If you’re already collecting data or are using the previous identity system, there are additional considerations you need to think through before configuring your system. To learn more about those configurations, visit the Create an Identity Type and User Identity Mapping Salesforce Help articles.
Planning for Identity
In your discovery, it’s important to plan through the identity. The first thing you need to think about is what activation channels you’re looking to enable. For example, do you want to collect and activate through the web or email, or do you want to bring in offline interactions? Make sure you choose identities that cover all identity channels but also make sense. And understand the relationship between identities. Do you have a customer ID or an email address available across all systems? Could there also be multiple email addresses tied to that customer ID? You should also review merge logic in the documentation to understand the implications of enabling multiple identity values.
There are three components that make up an identity attribute—uniqueness, case sensitivity, and attribute data type.
Uniqueness
This is an important configuration that you set for the identity attribute. There can be two states for uniqueness. First, you can choose “not unique.” This means that two user profiles within the dataset can have the same value.
If you choose to have a “not unique” attribute, these values aren’t available when the identity system seeks to merge profiles together. For example, if you’re getting an event from the web and it has a “not unique” identity attribute, it’s not considered in determining whether the profile is new or existing. The profiles remain separate.
This uniqueness allows you to search within the platform to find a specific profile. If you’re looking to support lookup and merge, you should choose the identity namespace value in uniqueness. This means that only one user profile within that dataset can have that value. So if multiple people have the same phone number, then you shouldn’t configure this as the identity namespace value. You should choose “not unique,” but remember that this makes the value not available for profile merging.
A configured identity namespace value supports your lookup and merge, and the ability to search within the Marketing Cloud Personalization platform.
Case Sensitivity
This can have two separate values.
-
False, which means that you don’t need to have a match on the casing in order to have a match.
-
True, which means that all casing must match to lookup and merge.
Both values still support being able to lookup and merge and search within the UI.
Attribute Data Type
You configure this on the attributes screen. The only supported type for an identity value is “string,” meaning that a user profile has at least one value in that attribute. Selecting “string” allows you to lookup and merge and search within the UI.
Prebuilt Identity Types
The identity system already comes with prebuilt identity types. These attributes are frequently used and configured with best-practice settings. These identity types come out of the box.
- Email address
- Salesforce Marketing Cloud contact key
- Salesforce CRM Lead ID
- Salesforce CRM Contact ID
- Customer ID
To use these identity types successfully, you must tie them to the attribute screen. If you’re missing these definitions in your dataset, reach out to the Marketing Cloud Personalization support team for assistance. And it’s highly encouraged and best practice that if you use any prebuilt identity types, you don’t create your own.
Setting Up Identity
Once you log in to MCP, there are a few screens and settings you need to configure to complete your identity setup. The first is configuring your Identity Type page. This is available under settings, scrolling down to account setup, and selecting Identity Types. On this screen, you see the five built-in identities you learned about: Customer ID, Email Address, Salesforce CRM Contact ID, Salesforce CRM Lead ID, and Salesforce Marketing Cloud Contact Key. Again, these are already preconfigured with the uniqueness to support lookup and merge with identity namespace and the best practice case sensitivity already configured. If you need additional identity types, visit the Create an Identity Type Salesforce help article to see the steps.
Keep in mind that once you save, you cannot modify that row on the Identity Type screen. It’s best practice to define your entire identity solution before configuring the screen.
Once you have the corresponding identity types you need, the second step is tying your identity types to the associated user attributes. Find this screen under Settings by clicking Attributes. Here, on a new account, you can see that three of the five built-in attributes are already preconfigured and tied to their identity type: Customer ID, Email Address, and Salesforce Marketing Cloud Contact Key. If you plan to use the other default identities, you must create the attribute inside them, just like you create one of the new values.
For the default values that are selected, you can untie the identity type. Simply select it to edit, and then remove it. Keep in mind that if you alter this setting, you can no longer read it or tie it back to Customer ID. So if you plan on not using that default identity, you can either remove the identity type, or you can simply choose not to send customer ID as an attribute into the platform and not use that identity type.
To add client-specific identity types as attributes, follow these steps.
- Click Add New Attribute.
- Name the attribute.
- Add a preconfigured label.
- Select String as no other types will be supported.
- Choose your Classification Value.
- Tie it to the appropriate identity type.
Once you set up the identity type screen and tie your attributes together, you complete the identity setup. But to successfully use the identity system for a particular channel, there are a few settings you need to be aware of and process. First, you configure a few things so you can use the web for tracking and activation.
Configure the Identity System
- From main navigation, select Settings then General Setup.
- Select Advanced Options to expand it.
- Select the identity attribute from the WebSDK Identity dropdown. Note: You can use the preconfigured Customer ID option or you can modify it and select different values. This is also where you choose your most prevalent webID to merge.
- Once you select your identity attribute, click Save.
If available in a web event, the other identities are considered for lookup and merge. This should include the primary ID and the value you selected, which affect how you configure your sitemap. Once you select your identity attribute, click Save.
There are considerations for how your sitemap is selected based on what you chose as your primary webID. Based on that selection, there are other considerations. You must configure the ID that you choose as the default to send as the user ID with the name in your sitemap. Your other identity types that you captured on the web are still able to be sent and will be captured as user.attributes. So if you plan on using the identity system with the web, there are these two setup steps: First setting your default web ID, and then making appropriate updates to your sitemaps to capture these values.
Extract, Transform, and Load
If you plan to use the identity system in extract, transform, and load (ETL), you don’t need to configure additional settings within the platform that you have to process. Instead, ETL formats and ETL gears are built to support the identity system. The heading or column to configure identities should say attribute:identity name. This is also valid for any feeds that are for users.
You can also include multiple headers, and if you’re sending the email address as well as the customer ID, you can format both columns with attribute: customer ID and attribute: email address. The identity system naturally accepts and uses the columns for lookup in merging profiles to determine whether that record should be tied to an existing or new profile.
The Open Time Email Identity Attribute
To create unified customer profiles for each of your customers, use the Personalization identity system to configure the identifiers in your Open Time Email campaigns. If you plan to use Open Time Email within the identity system, you need an additional configuration.
To configure Identity Resolution for open time email, follow these steps.
- From the main navigation, select Settings then General Setup.
-
Open Time Email Campaigns - Enable Identity Resolution has “email address” selected by default. Keep this value or select another as the match key between Marketing Cloud Personalization and ESP. Note: Marketing Cloud Personalization supports one identity as the match key. For example, to determine the content within the Open Time Email block, the Open Time Email code that renders at inbox delivery looks up the email address in the ESP and within Marketing Cloud Personalization.
- Select the checkbox for Open Time Email Campaigns. This option specifies the email user ID tag, and you can also specify the user ID tag from your ESP. This setting automatically configures the token to be in the HTML block and allows it to be added to the box.
- Click Save.
Which Profile Is Assigned an Event?
The identity system finds all user profiles that match any identity attribute value. It only considers identity attributes with a uniqueness of “identity namespace.” These profiles merge, and the event is recorded to that merged profile.
Merge Logic
Receiving an event can result in multiple existing users merging into a single user. For example, when user profiles merge, older ones are merged into the most recently updated profile. All events, transactions, and interactions are therefore merged into the destination profile.
Attributes, including identity attributes, are arbitrated with this logic.
- Any attribute that exists with only one user is kept and applied to the destination user.
- When an attribute exists in more than one user, the attribute with the latest timestamp in the metadata is selected (can be specified in Gear APIs).
- If the metadata does not exist or does not have a timestamp, the attribute with the latest attribute updated timestamp is selected (controlled by the platform).
Identity Considerations
Consider these factors regarding identities:
- Only unique identities are available for lookup and merge.
- Don’t create a custom identity attribute that already exists as a default attribute.
- Identity attributes can only be string values.
- Combining multiple attributes to form a single identity attribute is unsupported, such as looking up against primary email and secondary email or sending a phone attribute and email attribute. The transformation would have to happen externally.
- Do not store two attributes that are tethered together, such as email address and hashed email as merging could cause mismatches.
Note: Remember, once an identity attribute is set up, it cannot be modified. If there’s no reason to track identities, don’t track them. Also, don’t store attributes that are tethered together, such as hash email and email addresses.
Now that you know about identity configuration, planning, and ETL, you’re ready for some segment strategizing.