Overview
This section covers the step that turns scattered source records into a single customer view. It has two objectives: define the purpose, process and supported use cases of unification, and model data in Data 360.
Modeling means mapping data lake objects to data model objects in the Customer 360 Data Model, such as Individual, Contact Point Email, Contact Point Phone, Party Identification, Account and Sales Order, and creating custom objects, fields and relationships when the standard model does not fit. Unification uses identity resolution rulesets. Match rules decide which records belong to the same person using exact, normalized or fuzzy criteria on names, contact points and party identifiers. Reconciliation rules decide which value wins for each unified attribute, using options such as last updated, most frequent or source priority. The result is the unified individual or unified account that segments, insights and AI use.
How it's tested
Questions show source fields or sample customer records and a desired outcome. You choose the data model object to map to, the relationship to create, the match rule that links the right records, or the reconciliation rule that picks the correct value for the unified profile.
-
Map it right
Source fields are listed, and you choose the data model object and field, such as Contact Point Email for an email address.
-
Predict the merge
Several records and a match rule are shown, and you decide which records unify into one profile.
-
Choose the winning value
Conflicting attribute values come from several sources, and you pick the reconciliation rule that yields the required value.
Two records share the email [email protected], but one says Maria Lopez and the other M. Lopez-Diaz. A match rule using exact normalized email plus fuzzy first name would link them, while an exact full name rule would not. If the business trusts the loyalty system for names, a source priority reconciliation rule placing loyalty first sets the unified name.
Objectives 2
Official objectives from the Salesforce outline, in outline order.
4.1 Define the purpose, process, and supported use cases of unification.Identity resolution rulesets, match rules and reconciliation rules.
- What it means
- Unification links source records that represent the same person or account into a unified profile. An identity resolution ruleset contains match rules, which decide which records belong together, and reconciliation rules, which decide which attribute values the unified profile shows. Rulesets can run for individuals and for accounts in B2B scenarios.
- What to understand
- Match rules combine criteria such as exact or fuzzy name, normalized email or phone, address, and party identifiers like loyalty numbers. More rules or looser criteria merge more records, which raises the risk of over-merging. Reconciliation options include last updated, most frequent and source priority, set per attribute. Unified profiles keep links to source records, so changes in sources update the profile on the next run. Use cases include a single customer view, deduplicated segment counts, cross-channel personalization and AI grounding.
- How it shows up
- Sample records and a ruleset are described, and you predict the unified result or choose the rule that achieves the goal.
- Common confusion
- Confusing match rules, which group records, with reconciliation rules, which choose values after grouping.
4.2 Model data in Data 360.Mapping data lake objects to the Customer 360 Data Model.
- What it means
- Modeling maps each data lake object's fields to data model objects in the Customer 360 Data Model, which provides standard subject areas such as Party, Engagement, Sales Order, Case and Product. Custom data model objects and fields extend the model, and relationships between objects connect individuals to their contact points, orders and events.
- What to understand
- Map personal details to Individual and contact details to Contact Point Email, Contact Point Phone and Contact Point Address, each related back to the Individual. Map external identifiers to Party Identification for use in matching. Engagement data maps to engagement objects with event dates. Define relationships with correct cardinality so insights and segments can traverse them. Prefer standard objects for compatibility with apps and activation, and add custom objects only when nothing fits.
- How it shows up
- A source dataset is described, and you choose the data model object, field mapping or relationship needed.
- Common confusion
- Mapping email and phone onto the Individual object instead of contact point objects, which identity resolution needs.
High-yield concepts
The terms, rules and formulas this domain turns into questions most often.
-
Customer 360 Data Model
Salesforce's standard canonical model for Data 360, organized into subject areas.
- Remember
- Standard objects work best with apps and activation.
- Tested as
- Mapping questions.
- Trap
- Creating custom objects when standard ones fit.
-
Individual DMO
The data model object representing a person.
- Remember
- Identity resolution for people starts from Individual records.
- Tested as
- Mapping and unification questions.
- Trap
- Putting contact details directly on it.
-
Contact point objects
DMOs for email, phone, address and app or social contact points.
- Remember
- Each relates to an Individual and feeds matching and activation.
- Tested as
- Mapping and activation questions.
- Trap
- Leaving contact points unmapped.
-
Party Identification
DMO for external identifiers such as loyalty or customer numbers.
- Remember
- Allows exact matching on trusted IDs.
- Tested as
- Match rule design.
- Trap
- Matching on IDs stored in custom fields only.
-
Identity resolution ruleset
The configuration that defines match and reconciliation rules for unification.
- Remember
- Produces unified individual or unified account objects.
- Tested as
- Unification process questions.
- Trap
- Expecting unification without a ruleset.
-
Match rule
Criteria that decide which source records represent the same person.
- Remember
- Rules are OR'd together; criteria within a rule are AND'd.
- Tested as
- Predict-the-merge questions.
- Trap
- Thinking all rules must match together.
-
Fuzzy matching
Approximate matching on names to tolerate typos and variations.
- Remember
- Used with stronger exact criteria to avoid false merges.
- Tested as
- Name variation scenarios.
- Trap
- Using fuzzy name as the only criterion.
-
Normalized matching
Matching after standardizing formats such as email case or phone formatting.
- Remember
- Improves matches without loosening logic.
- Tested as
- Contact point matching.
- Trap
- Assuming formatting differences block matches.
-
Reconciliation rule
Rule that picks which source value the unified profile shows.
- Remember
- Options include last updated, most frequent and source priority.
- Tested as
- Choosing-the-value questions.
- Trap
- Confusing it with match rules.
-
Over-merging
Linking records of different people into one profile.
- Remember
- Caused by loose rules such as fuzzy name alone or shared household emails.
- Tested as
- Rule tuning scenarios.
- Trap
- Assuming more matches are always better.
-
Unified account
Account-level unification for B2B data using account rulesets.
- Remember
- Separate from individual unification.
- Tested as
- B2B use cases.
- Trap
- Using individual rules for companies.
Side by side
Concepts candidates mix up, compared on the point that decides the answer.
Match rules vs reconciliation rules
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Match rules | Decide which records are the same person | Grouping step | Using them to choose a phone number |
| Reconciliation rules | Decide which value the profile shows | Value selection after grouping | Expecting them to merge records |
Reconciliation options
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Last updated | Uses the most recently changed value | Favors fresh data | Choosing it when one source is more trusted |
| Most frequent | Uses the value that appears most often | Favors consensus across sources | Applying it to rarely repeated fields |
| Source priority | Uses the value from the highest ranked source | Favors a trusted system | Forgetting to order the sources |
Match criteria types
| Concept | Meaning | Key difference | Common trap |
|---|---|---|---|
| Exact | Values must be identical | Most precise | Missing typos and case differences |
| Normalized | Values compared after standardizing format | Precise but tolerant of format | Thinking it allows spelling errors |
| Fuzzy | Approximate similarity on names | Tolerant but less precise | Using it alone and over-merging |
Exam traps
Each trap is an answer that looks right. Learn the clue that should change your mind.
-
Contact on Individual
- What candidates pick
- Map the email field directly onto the Individual object.
- Why it looks right
- Email describes the person.
- Why it's wrong
- Identity resolution and activation read contact point objects, so email belongs in Contact Point Email related to the Individual.
Clue to watch for The question asks how to enable matching on email.
-
Fuzzy alone
- What candidates pick
- Create a match rule on fuzzy first and last name only.
- Why it looks right
- It catches spelling variations.
- Why it's wrong
- Names alone link different people with similar names; pair fuzzy names with an exact or normalized contact point.
Clue to watch for Many customers share common surnames.
-
Reconcile to merge
- What candidates pick
- Change the reconciliation rule to combine duplicate profiles.
- Why it looks right
- Reconciliation sounds like merging.
- Why it's wrong
- Reconciliation picks attribute values after grouping; only match rules change which records merge.
Clue to watch for Duplicates remain after a run.
-
All rules together
- What candidates pick
- Assume every match rule must succeed for records to merge.
- Why it looks right
- Rules look like a combined checklist.
- Why it's wrong
- Match rules act as alternatives, so meeting any one rule links the records, while criteria inside one rule must all match.
Clue to watch for A ruleset with several rules is shown.
-
Shared household email
- What candidates pick
- Match on email alone in a market where families share an address.
- Why it looks right
- Email is a strong identifier.
- Why it's wrong
- Shared emails merge several family members; add name or a party identifier to the rule.
Clue to watch for The scenario mentions households or shared contact details.
-
Last updated everywhere
- What candidates pick
- Use last updated for every attribute.
- Why it looks right
- Newest data feels most accurate.
- Why it's wrong
- A trusted system of record may be older; source priority keeps its values when trust matters more than recency.
Clue to watch for The business names one system as authoritative for a field.
Answer framework
Use these steps when a scenario involves mapping data or designing identity resolution.
- Map each source field to the correct standard data model object, placing contact details on contact point objects and external IDs on Party Identification.
- Check that relationships link contact points, identifiers and engagement records back to the Individual or Account.
- Choose match criteria from strongest to weakest: party identifiers, normalized contact points, then fuzzy names combined with another criterion.
- Test the rules mentally against the sample records, looking for both missed matches and over-merging of different people, such as relatives who share one email address or phone number.
- Set reconciliation per attribute using source priority for trusted systems, last updated for changing values and most frequent for consensus.
Practice question
Want more like this? The Data-Con-101 practice test has questions for every domain.
Checklist
0 of 12
Tick each one you can do without notes. Your ticks stay in this browser.
Study strategy
Begin with the data model. Study the Customer 360 Data Model module and draw the Individual object with its contact point objects, Party Identification and a sales order or engagement object attached. Label the relationship keys. If you have a playground, map a sample customer file and confirm the mappings in Data Explorer.
Then move to identity resolution, the most scenario-heavy part of the section. Create a small table of six or seven fictional records with variations: a typo in a name, two formats of the same phone number, a shared household email and a loyalty ID present in only two records. Apply different rulesets on paper and write down which unified profiles result. Repeat with reconciliation rules to see which values each option would choose.
Finish with the Data and Identity in Data 360 module from the official trail and review B2B account unification. In practice questions, separate grouping problems from value problems first, because that single distinction eliminates many wrong options.
FAQ
How much of the exam is Harmonization and Unification?
Harmonization and Unification is 17% of the exam, about ten of the 60 scored questions as a planning estimate.
Why does the old outline list Identity Resolution separately?
Earlier Data Cloud Consultant guides had separate sections for data modeling and identity resolution. The Spring '26 Data 360 guide combines modeling and unification in this section.
How do multiple match rules work together?
Records unify when they satisfy any one match rule. Within a single rule, every criterion must match. Adding rules therefore usually increases merges.
Can I change which source wins for one field only?
Yes. Reconciliation rules are set per attribute, so you can use source priority for name and last updated for address, for example.
Is identity resolution only for individuals?
No. Account rulesets support unification of business accounts for B2B scenarios, separate from individual unification.
Does unification delete source records?
No. Source records stay in their data lake and model objects, and the unified profile links to them. Changes in sources flow into the profile on later runs.
Other CRM & Business Applications certifications
Where the same subject shows up in other exams. Cards marked "Same subject" open the matching domain.
- HubSpot Inbound Marketing Certification Study guide
- Salesforce Certified Platform Developer I Study guide
- Google Ads Search Certification Study guide
- ServiceNow Certified System Administrator Study guide
- Salesforce Certified Platform Administrator Study guide
- ServiceNow Certified Application Developer Study guide
Test your Domain 4 knowledge
You have covered data mapping, the Customer 360 Data Model and identity resolution rules. Test whether you can predict unified profiles from sample records under time pressure.
Keep going
- Training Training Built on your Weak areas
- Practice test Questions by domain, with explanations
- Mock exam Full length, timed like the real exam
- Readiness test Find your weakest domain in minutes
- Cheat sheet Key terms and formulas on one page
- Ask the Data-Con-101 AI tutor Get any concept explained your way