Data Cloud - Identity Resolution

What is an anonymous profile?

A profile created from sources lacking specific identification. The determination of anonymity status during data ingestion is based on the Is Anonymous field in the Individual DMO.

What is a unified link?

A unified link is an object that allows users to view the source data for each unified profile.

In Identity Resolution, what does Processing History tell you?

Click Processing History to view the processing history of your Resolution Rules including last processing data, individual sources, matching, and consolidation rates. Match rules can be edited, so be sure to monitor regularly. To increase the consolidation rate, try adding more match rules. To reduce the consolidation rate, try removing some of the match rules.

What are the unification timings?

Create or Update Rules - Refresh on-demand and maximum of four times in 24 hours Scheduled - Refresh on batch schedule every 24 hours

After configuration, how can you monitor and edit your rulesets?

Identity Resolutions Tab From the individual ruleset page, you can view the ruleset properties, details, and processing history.

Give an example of Consolidation Rate?

If you have 10 source records and after unification you get 7 unified profiles you have a consolidation rate of 30%.

Will adding more rules to a ruleset increase or decrease the match rate?

Increase The more rules, the more opportunities to match. (Using OR operator)

Which object is used for unique external identifiers?

Party Identification

What are Reconciliation Rules?

Reconciliation rules determine the logic for data selection. Reconciliation is the process of summarizing key attributes that have been unified. For example, if the same email address is available from two data sources, a reconciliation rule helps the unified profile know which one to display.

After the Identity Resolution process completes, what should you examine to validate the Identity Resolution ruleset outcome?

Resolution Summary

Which reconciliation rule should be used if you trust one data source over another?

Source Sequence

What is the Individual ID?

The individual ID is the most important attribute for identity resolution rulesets. It serves as the primary key for the individual object and is required for mapping and identity resolution. In your datastream, whatever field is the unique identifier for that customer should be mapped to this ID. Here are the specific mapping requirements and options for the individual object.

What is consolidation rate? (know this!)

The percentage of profiles that have been consolidated. (If your system has no matches, the consolidation rate is zero.)

What is Reconciliation?

Your goal in reconciliation is to determine which system gets priority for which field at what time. Reconciliation is the process of picking attributes to describe a unified profile.

What type of Identity Resolution match rules are available?

1) Fuzzy name and normalize email 2) Fuzzy name and normalized phone 3) Fuzzy name and normalized address 4) Fuzzy name and normalized phone and Normalized email 5) Custom Rule

What are the steps to create a unified profile?

1) Ingest raw data from data sources 2) Map and model data 3) Create identity resolution rulesets 4) Create unified profiles

Name two characteristics of anonymous profiles?

1) Once at least one known individual profile is matched with an anonymous record, that record will be marked as known going forward. 2) They are excluded from the known profile utilization consumption.

What is required to create a Match Rule?

1) Select an object from either: individual, contact points (email, app, phone, and address), device, or party identification. 2) Select your field. Select the attributes available based on the object selected. 3) Pick a Match Method, depending on the object and the field type selected. (Exact, Fuzzy or Normalized)

What factors should you consider when creating data sets?

1) Take inventory on all the data sources you might want to incorporate 2) Identify all data sets required for each data source, such as ecommerce data with data sets for sales order details and sales order header data.

Successful Identity Resolutions results in what?

1) Unified Profile 2) Unified Link

What are the steps for Identity Resolution Implementation?

1. Profile Data across data sources - Understand whether or not data from various sources matches expectations. 2. Configure Match rules - Make appropriate tradeoffs between over and under grouping of Individual records. 3. Configure Reconciliation Rules - Specify how matched records should be reconciled if there are attribute conflicts. 4. Validate Results - Impact resulting unified records and plan for next steps.

What is a ruleset?

A grouping of criteria. Adding more rules increases the match rate.

What is a Unified Profile?

A unified profile is composed of data from multiple sources linked together using identity resolution match and reconciliation rules. If the same data exists in multiple places, profiles are linked together based on established rules.

What does "Match on Blank" mean when creating Identity Resolution Rulesets?

Allows to match even if the field is blank (match even if Birth Date field is blank)

In the App builder, what dos the Data Cloud Profile Related Record do in a page layout?

Builds out the Profile Explorer page Shows unified links and the Data Source Object values

What is the default reconciliation rule?

By default, this field inherits the Last Updated reconciliation rule set as the default rule for this object.

What are best practices for using Data Cloud and Identity Resolution?

DRUMS 1) Consider Data accuracy and cleanliness. Can you clean any data before importing your raw data into Data Cloud? 2) Use Rulesets to compare and test match and reconciliation rules. Once you have your first ruleset configured, you can create a second to conduct A/B tests. 3) Review your Unified profiles regularly from Profile Explorer to see if tweaks need to be made to your ruleset. 4) Determine Matching rules and requirements before you begin mapping your data. 5) Unified profiles are only as trustworthy as the Source system data. What data source has the most up-to-date information? Use this as a guide for your match and reconciliation rules.

Will adding more criteria increase or decrease the likelihood of matching profile?

Decrease (Criteria) AND (criteria) AND (criteria) Adding more criteria will make it more difficult to match.

When working with datasets, what data sources might you want to incorporate?

Digital engagement data (including web and mobile) Ecommerce External databases Marketing and email databases Data lakes Analytics CRM Customer service Traditional software

Describe the three types of Match Methods.

Exact: Matching based on an exact match. No typos or alternative formats. Fuzzy: Matching based on a similar match. Typos and slightly different spelling are OK. This is only available for first name. Normalized: Matching based on the same exact info, regardless of formatting. This is available for email, phone, and address.

True or False: Keep identity resolution rules relatively loose when contacting individuals.

False Use identity resolution rules that are fairly strict when contacting individuals. Don't use reconciled data to contact or present back to customers. (Because the system may inaccurately consolidate records that are truely different people. A married woman and her sister-in-law may have the same name and could live together. You wouldn't want to contact one of them for medical information or an unpaid bill.)

True or False - You can have many rules in a ruleset, but only one ruleset.

False You can name multiple rulesets.

True or False: Identity Resolution can span across Data Spaces.

False Identity Resolution is isolated to a Data Space

True or False - Only Individuals can be unified.

False - Accounts and Individuals can be unififed

What attributes are available for fuzzy or normalized matching?

Fuzzy: Only first name Normalized: Phone, Email, Name & Address

What are the options for reconciliation rules?

Last Updated - This rule specifies that the most recently updated value must be selected for inclusion in the unified profile. It's worth considering what data gets updated most regularly—would it be customer service data or perhaps Marketing Cloud Engagement preference data? Most Frequent - This rule specifies that the most frequently occurring value must be selected for inclusion in the unified profile. Source Sequence - This rule allows you to sort your data sources in order of most to least preferred for inclusion. Basically it allows you to select based on your confidence in the data source. As an example, you can specify that the system use Commerce Cloud data first and S3 data last.

What are the types of reconciliation? (know this!)

Last modified Most frequently occurring Source Priority

What are the three levels of fuzzy matching

Low Precision Medium Precision High Precision

What is the cardinality relationship of the party identification object?

Many To One - you can have multiple party fields mapped to one object. In this example, there are 3 party identification types: one from LinkedIn, another called Contact ID, and a Marketing Cloud Subscriber Key. All three are possible and, therefore, the reason for the cardinality of Many to One.

What are Match Rules?

Match rules are used to link together data into a unified customer profile. Matching is the process of grouping profiles together through commonly shared criteria.

What is overgrouping? (know this!)

Match rules that are too loose and will incorrectly match prospects. Excessive merging of individuals in a single unified profile. This happens when the resolution process lacks specific criteria, resulting in the inclusion of individuals who should ideally be kept distinct. Email OR phone OR Party Identifier OR Address OR Device

In the Identity Resolution tab, what does the Resolution Summary tell you?

Number of Unified profiles Consolidation Rate Known Unified profiles Anonymous Unified profiles

What attributes are typically normalized in match criteria?

Phone Email Name Address

What are Identity Resolution Rulesets?

Rulesets allow you to configure match rules and reconciliation rules about a specific object, such as an individual. The system follows these rules to link together multiple sources of data into a unified profile. Match Rules: Fuzzy Name and Normalized Email -OR- Fuzzy Name and Normalized phone -OR- Advertiser ID -OR- Fuzzy Name and Normalized Address

In the Identity Resolution tab, what does processing history tell you?

Run date Total Source Profiles Unified Profiles Total Known Profiles Total Unknown Profiles Consolidation Rate Processed Records Aggregate Status

How often do resolution rulesets run?

Scheduled to run at least one time per day after publication

True or False: Data Cloud unification is non-destructive, and it allows data within source systems to change

True

True or False: Deleting a ruleset permanently removes all unified customer data, eliminate dependencies on data model objects, ends it's processing, and erases its past activity history. (know this!)

True

True or False: You can have multiple rules to make a ruleset.

True

True or false. You can create 2 rulesets to use for testing identity resolution rules.

True

True or False - Identity Resolution runs matching first and then reconciliation second.

True These processes are all user-defined and customizable. This is a scheduled process that runs on new and changed records.

When adding Party Identification as a match rule, what additional information is required?

You must add Party Identification Type and Party Identification Name