Salesforce Certified Data 360 Consultant Practice Exam

What the Salesforce Certified Data 360 Consultant exam covers

• Solution Positioning93 questions
• Data 360 Setup and Administration85 questions
• Data Source Connection and Ingestion119 questions
• Harmonization and Unification113 questions
• Data Enhancements, Sharing, and Analysis119 questions
• Data Activations and Utilization131 questions

Free Salesforce Certified Data 360 Consultant practice test questions

A sample of 10 questions with answers and explanations. Sign up free to practice all 660.

1. Question 1Solution Positioning

   A retailer already runs a data warehouse and asks why they would add Data 360. What is the strongest positioning answer?

   • AIt resolves records into unified profiles and acts on them inside Salesforce in near real timeCorrect
   • BIt stores more data at lower cost than a warehouse of comparable size
   • CIt replaces the warehouse, since both serve the same analytical purpose
   • DIt runs SQL faster than a warehouse because it is purpose-built for queries
   ✓ Correct answer: A

   A warehouse is excellent at storing and analysing history, and it does not resolve a person across systems into one profile, nor push that profile into a flow, an agent or a marketing audience. Data 360 does both: it harmonizes sources into a shared model, unifies identities, and makes the result usable by Salesforce clouds without a separate integration layer. That is why it complements a warehouse rather than competing with it, and Zero-Copy makes coexistence practical.

   Why the other options are wrong

   • BStorage cost is not where the value sits, and consumption is metered by credits rather than by cheap bulk storage. A warehouse is often cheaper per terabyte.
   • CPositioning it as a warehouse replacement invites a comparison it will lose on reporting depth and SQL maturity. The two are complementary.
   • DQuery speed against a warehouse is not the claim, and analytical SQL is what warehouses are optimised for. Activation is the differentiator.
2. Question 2Solution Positioning

   A client running a children's product line asks about handling data on under-16 customers. What is the sound guidance?

   • ATreat it as heightened risk requiring a specific lawful basis and tighter limitsCorrect
   • BHandle it identically to adult data, since the platform applies the same controls
   • CExclude minors from ingestion, which removes every obligation
   • DRely on parental consent captured by the website, with no further design change
   ✓ Correct answer: A

   Most regimes give children's data heightened protection, with a specific basis such as verified parental consent, tighter limits on profiling and advertising, and often restrictions on the retention permitted. The design consequences are real: knowing which profiles are minors, excluding them from certain activations, and being able to evidence the basis. Treating it as ordinary personal data because the platform does not distinguish is the mistake to avoid.

   Why the other options are wrong

   • BThe platform applies the controls it is configured with and does not know a profile is a minor, so identical handling is a design decision rather than a safeguard. The obligations differ.
   • CExcluding minors may be a legitimate choice and requires identifying them, which itself needs data, and it forgoes any lawful use. It is not obligation-free.
   • DCaptured consent is necessary and not sufficient, since profiling and advertising limits and retention rules still apply. Design changes follow.
3. Question 3Data 360 Setup and Administration

   An administrator must give an external agency access to build segments without seeing a sensitive attribute. What configuration achieves this?

   • AA permission set granting segmentation while withholding access to that fieldCorrect
   • BA data space containing only the fields the agency may see
   • CA calculated insight that hides the field's values from segment authors
   • DAn activation that excludes the field when publishing to the agency
   ✓ Correct answer: A

   The requirement is about what a user may read, so the answer is a permission set granting the segmentation capability without access to the sensitive field, which keeps the field out of their picker and out of any query they write. That is one control rather than a structural workaround. Duplicating data into a restricted space would work and creates a second copy to keep current and to include in every privacy exercise.

   Why the other options are wrong

   • BA space containing a subset of fields means duplicating ingestion and maintaining two copies, which is a heavy answer to a permissions question. Spaces partition brands rather than field visibility.
   • CCalculated insights compute metrics and are readable like other data, so they conceal nothing from an author. Computation is not concealment.
   • DActivation governs what leaves the platform, and the agency is being given access inside it. The exposure is internal rather than at publication.
4. Question 4Data Source Connection and IngestionSelect all that apply

   A client's warehouse team objects that federation will add query load to their platform. Which two responses are honest?

   • AThe load is real, and predicate pushdown limits how much data is scannedCorrect
   • BIngesting selected tables instead moves that load off their platformCorrect
   • CFederation adds no load, since Data 360 caches results after the first query
   • DThe warehouse team can throttle Data 360's queries at the source without effect
   • EQuery load only occurs during identity resolution, which runs rarely
   ✓ Correct answer: A, B

   The objection is valid: federated queries execute on their platform and consume its compute, and their cost model may bill for it. The honest answer accepts that and offers two mitigations, namely that pushdown keeps the scanned volume proportionate to the predicate, and that any table where the load is unacceptable can be ingested instead. Framing it per table rather than as a platform-wide decision usually resolves the conversation.

   Why the other options are wrong

   • CNothing caches federated results so that subsequent queries avoid the source, which is why the load recurs. Denying it damages credibility.
   • DThrottling at the source would slow or fail Data 360's queries, so it very much has an effect. It is a mitigation with consequences rather than a free one.
   • ESegmentation, insights and analysis all query federated tables, so load is not confined to unification. The scope is much wider.
5. Question 5Data Source Connection and Ingestion

   A client asks what happens if two streams write to the same data lake object. What should the consultant explain?

   • AEach stream populates its own object, so this arrangement does not ariseCorrect
   • BBoth write to the object, with the later refresh overwriting the earlier
   • CThe second stream is rejected, since an object accepts one stream
   • DRecords are interleaved, and the primary key decides which survives
   ✓ Correct answer: A

   A stream lands its records in its own data lake object, so several sources contributing to one concept converge later, at mapping, where each object maps to the same data model object. That is where harmonization happens. Understanding the one-to-one relationship at the lake layer explains why an implementation has many lake objects and comparatively few model objects, and why the model is where the shared shape lives.

   Why the other options are wrong

   • BStreams do not share a lake object, so there is no overwrite between them. Convergence happens at the model layer.
   • CNothing is rejected, because the situation does not occur; each stream creates its own object. Rejection implies a contention that is absent.
   • DInterleaving would require a shared object, which is not how the layer works. The primary key reconciles within one object.
6. Question 6Harmonization and UnificationSelect all that apply

   A client asks how to test a ruleset change safely. Which two approaches are practical?

   • AApply it in a sandbox and compare profile counts and samples against productionCorrect
   • BApply it to a data space holding a limited subset before the main oneCorrect
   • CApply it in production during a quiet period, reverting if counts move
   • DApply it to one segment first, so only that audience is affected
   • EApply it and monitor activation match rates at the destinations
   ✓ Correct answer: A, B

   A ruleset change redraws every profile in its space, so testing means applying it somewhere the consequences are contained. A sandbox gives that at the cost of differing data, and a data space holding a subset gives realistic data with a bounded population. Either way the evidence is the same: profile counts before and after, plus inspection of profiles the change affected, judged against a stated expectation.

   Why the other options are wrong

   • CApplying it in production affects live audiences immediately, and reverting does not recall an activation that already published. Quietness does not contain it.
   • DRulesets operate per data space rather than per segment, so scoping a change to one audience is not available. Every segment in the space is affected.
   • EDestination match rates are a lagging and noisy signal that arrives after audiences have been sent. It is monitoring rather than testing.
7. Question 7Harmonization and UnificationSelect all that apply

   A client's ruleset was tuned six months ago and a new source has since been added. Which two actions are appropriate now?

   • AReview whether the new source's attributes justify an additional match ruleCorrect
   • BRe-measure match quality, since the population being matched has changedCorrect
   • CReset the ruleset to its defaults and tune it again from scratch
   • DIncrease the unification frequency to accommodate the extra source
   • EMove the new source into its own data space to isolate its effect
   ✓ Correct answer: A, B

   A new source brings records with their own attribute mix, so two things follow. The existing rules may not reach them, which is what an additional rule tuned to the new source's populated attributes addresses. And the overall match quality has changed, both because there are more records to link and because the new source may introduce values that widen or narrow existing rules, so the measurement should be repeated.

   Why the other options are wrong

   • CResetting to defaults discards tuning that reflects real understanding of the data. Incremental review is the proportionate response.
   • DIncreasing unification frequency affects how promptly links appear rather than how well records match. Cadence is not a quality lever.
   • EIsolating the new source in its own space would prevent it unifying with the existing profiles, which defeats the reason for adding it. Isolation is not a test here.
8. Question 8Data Enhancements, Sharing, and Analysis

   A client asks what a prompt template's grounding placeholders can reference. What is accurate?

   • AProfile attributes, insight values and content a retriever returnsCorrect
   • BOnly fields on the CRM record the template is invoked from
   • COnly content from a search index, not structured attributes
   • DAny object in the org, resolved automatically at run time
   ✓ Correct answer: A

   A template can pull in the customer's attributes and computed insights, which are the structured facts, and passages a retriever finds in indexed content, which are the unstructured ones. Combining both is what lets an assistant answer using this customer's situation and the relevant policy text together. Which of the two a given template needs is a design decision, and many need both.

   Why the other options are wrong

   • BRestricting to fields on the invoking CRM record would exclude unified data and insights, which is most of what makes grounding valuable. The scope is broader.
   • CStructured attributes are very much available alongside retrieved content, so excluding them misdescribes the capability. Both paths are supported.
   • DReferences are declared in the template rather than any object resolving automatically, which keeps what leaves the platform deliberate. Nothing is implicit.
9. Question 9Data Enhancements, Sharing, and Analysis

   A client asks how they would know a data graph is meeting its latency requirement. What should be monitored?

   • ARetrieval response times against the stated budgetCorrect
   • BThe number of objects the graph includes
   • CHow often the graph is referenced by segments
   • DThe storage the graph occupies in the data space
   ✓ Correct answer: A

   A graph exists to meet a latency requirement, so the monitoring is response time measured against that budget, including the slower tail rather than only the average, since a page or conversation is judged by its worst experiences. Establishing the budget when the consumer is designed is what makes the monitoring meaningful, because otherwise there is nothing to compare the measurement against.

   Why the other options are wrong

   • BThe number of objects the graph includes affects payload size and maintenance rather than telling you whether the latency requirement is met. Breadth is a design input.
   • CHow often the graph is referenced by segments describes batch usage, and graphs serve real-time retrieval rather than segmentation. The consumer set is different.
   • DThe storage the graph occupies is a cost figure rather than a latency measurement. Size does not indicate response time.
10. Question 10Data Activations and Utilization

    A consultant is asked what the risk is in building an activation without agreeing the destination's expectations first. What follows?

    • AThe payload may not match what the destination accepts or usesCorrect
    • BThe segment's criteria will need rewriting
    • CThe platform will refuse to create the activation
    • DThe consumption will exceed the forecast
    ✓ Correct answer: A

    An activation is an interface with a receiving system, so building it without agreeing the identifiers, the required fields, the format and the cadence means discovering the mismatch at the first publication, sometimes after a failed campaign. A short conversation with the receiving team, ideally with a sample payload they parse, prevents almost all of it. The segment and the platform are rarely the problem.

    Why the other options are wrong

    • BSegment criteria define the audience and are unaffected by the destination's payload expectations. The two are independent.
    • CThe platform creates the activation regardless of whether the destination's expectations were agreed, which is why the mismatch surfaces later. Nothing is refused.
    • DConsumption follows volume and cadence rather than whether the interface was agreed. Cost is a separate matter.

Who this Salesforce Certified Data 360 Consultant practice exam is for

This practice set is for anyone preparing for the Salesforce Certified Data 360 Consultant exam - from first-time candidates building a foundation to experienced Salesforce practitioners doing a final review before test day. If you learn best by working through realistic questions and reading why each answer is right or wrong, it is built for you.

How to use this Salesforce Certified Data 360 Consultant practice exam

1. Start with the free sample questions above to gauge your current baseline.
2. Read the full explanation on every question, including why each wrong option is wrong.
3. Track your weak domains and focus your study where you are losing the most marks.
4. Once you are scoring consistently well, take a timed, full-length mock exam.
5. Use your readiness score to decide when you are ready to book the real Salesforce Certified Data 360 Consultant exam.

Related Salesforce resources

• Salesforce Certified Data 360 Consultant study guideKey concepts
• Salesforce practice examsAll Salesforce
• Certification pathWhere this fits
• Certification exam guides & tipsBlog
• Plans & pricingFree & paid
• How these questions are written and reviewedMethodology
• Report a problem with a questionCorrections
• Salesforce Certified Platform Administrator (ADM-201) practice examRelated
• Salesforce Certified Platform App Builder practice examRelated
• Salesforce Certified Platform Developer I practice examRelated

Salesforce Certified Data 360 Consultant practice exam FAQ