Microsoft Power BI Quiz: Data Lineage And Endorsement
10 questions · exam conditions
0:00
Data Lineage And EndorsementQuestion 1 of 10

A finance team publishes several semantic models with similar names. One model is maintained centrally and has passed the organization's reconciliation and governance reviews. Business users must be able to distinguish this model from models that authors merely consider useful.

Which endorsement approach should the organization use?

Promote the centrally maintained semantic model because promotion confirms that formal governance checks were completed.
Certify the centrally maintained semantic model through a user authorized by the tenant's certification policy.
Add users to the semantic model's permissions because access rights identify the officially approved version.
Display the model in lineage view because lineage status confirms that its calculations were independently validated.
← Back to quizzes

Microsoft Power BI Quiz

Microsoft Power BI Quiz: Data Lineage And Endorsement

Practice Data Lineage And Endorsement in Microsoft Power BI with focused quiz questions that help you check what you know, review explanations, and build confidence with test-style prompts.

What this quiz covers

This quiz focuses on Data Lineage And Endorsement, giving you a quick way to practice the rules, question types, and explanations that matter most for Microsoft Power BI.

How to use this quiz

Try each quiz question before looking at the correct answer. Use the explanations to review missed ideas, then come back to similar questions until the pattern feels familiar.

All questions

Question 1

A finance team publishes several semantic models with similar names. One model is maintained centrally and has passed the organization's reconciliation and governance reviews. Business users must be able to distinguish this model from models that authors merely consider useful.

Which endorsement approach should the organization use?

  1. Promote the centrally maintained semantic model because promotion confirms that formal governance checks were completed.
  2. Certify the centrally maintained semantic model through a user authorized by the tenant's certification policy. (correct answer)
  3. Add users to the semantic model's permissions because access rights identify the officially approved version.
  4. Display the model in lineage view because lineage status confirms that its calculations were independently validated.
Explanation: When a question mentions governance reviews, reconciliation, and the need to formally distinguish a trusted dataset from others, you're being tested on Power BI's endorsement feature — specifically the difference between promotion and certification. Power BI offers two endorsement levels. Promotion is a lightweight signal that any content owner can apply themselves, indicating they find an item useful. Certification, by contrast, is a formal designation that only specific, tenant-authorized users can grant — and it's typically reserved for content that has passed official governance, quality, and reconciliation checks. Because the scenario describes a model that has completed the organization's formal review process and needs to be distinguishable from models that authors merely "consider useful," certification (answer B) is the correct approach. It conveys organizational authority, not just personal preference. Answer A is wrong because promotion does not confirm formal governance checks. Any content owner can promote their own model — it carries no institutional weight and is exactly the kind of endorsement that authors apply to content they merely find useful, which the question is trying to move away from. Answer C is wrong because permissions control who can access a model, not whether it has been officially approved. Granting access says nothing about quality or governance status. Answer D is wrong because lineage view shows data flow and dependencies — it doesn't validate calculations or signal any endorsement status whatsoever. Study tip: Remember the hierarchy — Certified > Promoted > None. On the exam, if a question mentions formal governance or organizational authority, the answer is almost always certification, not promotion.

Question 2

A semantic model is certified and protected by row-level security (RLS). Two users open reports based on the model. One user is assigned to the North region role, and the other is assigned to the South region role.

How does certification affect the data displayed to these users?

  1. Certification overrides RLS because certified semantic models must present one authoritative result to every user.
  2. Certification validates the RLS assignments and prevents model owners from changing the role definitions.
  3. Certification applies only to administrators, so ordinary report users cannot query the semantic model.
  4. Certification communicates organizational trust, while RLS continues to filter each user's results independently. (correct answer)
Explanation: When working with Power BI governance features, it helps to keep two concepts clearly separated: endorsement (which includes certification) and security. Certification is a trust signal — it tells users and report authors that a semantic model has been reviewed, approved, and is considered the authoritative source of data for the organization. It says nothing about who can see what data. Row-level security, on the other hand, is a data access control that filters query results based on the user's assigned role. In your scenario, the North region user sees only North data, and the South region user sees only South data — and certification has absolutely no bearing on that filtering. Both users interact with the same certified, trustworthy model, but each receives a personalized, role-filtered view. That's exactly what D describes. Choice A is built on a misconception: certification is about data quality and trust, not about collapsing filtered views into a single universal result. Overriding RLS would be a serious security breach, not a feature. Choice B invents a validation relationship between certification and RLS that simply doesn't exist — certification doesn't inspect, lock, or interact with role definitions in any way. Choice C confuses certification with some kind of access restriction, when in reality certification broadens discoverability; it makes a model easier to find and trust, not harder to access. A useful study habit: whenever a Power BI question mentions both endorsement/certification and a security feature like RLS, ask yourself whether these two systems interact at all. They operate on completely separate layers — endorsement governs trust, RLS governs data visibility.

Question 3

A lineage view shows that an imported semantic model obtains data from an on-premises SQL database and that three reports depend on the semantic model. A database administrator updates rows in SQL at noon. The semantic model's next scheduled refresh is at 6:00 PM.

What can be concluded from the lineage information?

  1. The reports depend on the SQL-to-semantic-model path, but imported data will not necessarily reflect the noon changes until refresh occurs. (correct answer)
  2. The reports must reflect the noon changes immediately because a lineage connection represents a live query relationship.
  3. The semantic model is automatically certified because scheduled refresh establishes a governed connection to SQL.
  4. The SQL update cannot affect the reports because lineage records relationships without identifying downstream dependencies.
Explanation: When you see a question involving Power BI lineage views, focus on two distinct concepts: what lineage shows (data flow relationships and dependencies) versus what it guarantees about data freshness. These are completely separate things. In an import mode semantic model, Power BI pulls a snapshot of the source data and stores it internally. Once that snapshot is taken, the model is frozen until the next refresh — no matter what happens to the source. So when the database administrator updates rows at noon, the semantic model doesn't "see" those changes. The three dependent reports read from the cached snapshot, not from SQL directly. Answer A correctly captures this: the lineage path exists, but imported data won't reflect the noon changes until the 6:00 PM scheduled refresh runs. Answer B confuses import mode with DirectQuery mode. A live query relationship (DirectQuery) would pass queries through to SQL in real time. Import mode does not — the lineage connection simply documents the data flow, not the query mechanism. Answer C introduces a completely unrelated concept. Certification in Power BI is a manual endorsement process managed by authorized users or admins; scheduled refresh has nothing to do with automatically certifying a semantic model. Answer D misrepresents what lineage does. Lineage absolutely does identify downstream dependencies — that's one of its primary purposes. The three reports are clearly visible as downstream consumers in the lineage view. Study tip: On the Power BI exam, always ask yourself whether a semantic model uses import or DirectQuery mode before reasoning about data freshness — the behavior is fundamentally different between the two.

Question 4

A sales report is exported to PDF and emailed to several executives. The report depends on a semantic model that is visible in lineage view. The data team later changes the model and wants to identify every executive who received the exported PDF.

Can lineage view satisfy this requirement?

  1. Yes, because lineage records every recipient who consumes content derived from a semantic model.
  2. Yes, but only when both the report and the semantic model have been certified by an authorized user.
  3. No, because lineage tracks managed item dependencies, not distribution of static files outside the service. (correct answer)
  4. No, because lineage can track interactive reports but cannot include any semantic model dependencies.
Explanation: Whenever you see a question about Power BI's lineage view, anchor your thinking to what it was actually designed to do: map dependency relationships between managed artifacts inside the Power BI service — things like dataflows, semantic models, reports, and dashboards. It is a governance and discovery tool for living, connected content, not a distribution tracking system. Lineage view cannot satisfy this requirement because once a report is exported to a static PDF and emailed, it leaves the Power BI service entirely. At that point, the file exists as a standalone attachment on someone's email client — there is no service-side mechanism that logs who received it, opened it, or forwarded it. Lineage view has no visibility into that distribution chain, making C the correct answer. A is wrong because it describes a capability lineage view simply does not have. It never records recipients or consumers of exported content — it only shows how in-service items depend on one another. B introduces certification as a condition, which is a red herring; certification affects trustworthiness and discoverability of datasets, but it has nothing to do with tracking PDF recipients regardless of certification status. D is wrong in a different direction — it falsely claims lineage cannot include semantic model dependencies, when in fact tracking those dependencies is lineage view's primary purpose. The error in D is that it undersells lineage while C correctly identifies its actual limitation. As a study tip, remember the boundary: lineage view = in-service dependency mapping. Any requirement involving external distribution, email recipients, or file downloads will fall outside its scope and typically requires usage metrics or audit logs instead.

Question 5

A certified semantic model is modified to use a new source and substantially revised business rules. The certification badge remains visible after the change. The organization's certification process requires validation of sources and calculations.

What should the content owner conclude?

  1. The visible badge proves the revised source and calculations were automatically revalidated against governance requirements during publication.
  2. The model remains trustworthy because certification technically prevents material changes to endorsed item definitions and sources.
  3. The badge replaces the need for another review because certification, once granted, applies permanently to all future versions of the item.
  4. The changed model should be reviewed again, and certification should be retained only if it still meets the organization's policy. (correct answer)
Explanation: When you see a question about Power BI endorsement and certification, keep this principle in mind: certification is a point-in-time validation, not a permanent guarantee. It reflects the state of a semantic model when it was reviewed — not what it becomes afterward. Power BI allows certified items to be modified after certification is granted. The badge persists visually, but it does not automatically re-trigger governance reviews or revalidate sources and calculations. The certification system has no built-in mechanism that freezes content or forces re-review upon change. That responsibility falls entirely on the content owner and governance team. So when a model undergoes substantial revision — new data sources, revised business rules — the ethical and policy-correct action is to treat it as a new candidate for certification review. D is correct because it aligns with how responsible data governance actually works: the badge should only be retained if the revised model still satisfies the organization's documented certification criteria. A is wrong because Power BI does not automatically revalidate content during publication. Republishing a dataset does not trigger any certification logic. B is wrong because certification does not technically prevent changes — owners can modify certified items freely, and the badge simply remains until manually revoked. This makes B factually false. C is wrong because certification is explicitly not permanent; Microsoft's own documentation recommends re-evaluating certified content when significant changes occur. As a study tip: on Power BI governance questions, always remember that endorsement features (promoted, certified) are manual, human-driven processes — Power BI automates nothing about governance compliance.

Question 6

An organization plans to retire a legacy semantic model. A replacement model has been validated and certified. Several reports still depend on the legacy model, and users are expected to prefer the replacement when creating new content.

Which plan best uses both lineage and endorsement?

  1. Use lineage to identify reports requiring migration, and use certification to signal that the replacement is the approved model. (correct answer)
  2. Certify the replacement to redirect all existing report connections, and use lineage to grant users access automatically.
  3. Use lineage to certify the replacement based on its dependencies, and promote the legacy model until connections change.
  4. Promote both models to preserve existing access, and remove lineage records for the model being retired.
Explanation: When managing a semantic model transition in Power BI, you need to keep two tools distinct in your mind: lineage shows you what connects to what, while endorsement (promotion or certification) signals how trustworthy or approved a dataset is. These are complementary tools — one is diagnostic, the other is communicative. Option A correctly leverages both for their intended purposes. You use lineage view in Power BI to trace which reports are downstream of the legacy model — this gives you a concrete migration checklist. Then you apply certification to the replacement model to signal to the organization that it's the officially approved option for new content. That's exactly what each tool was built for. Option B misrepresents what certification does. Certifying a dataset doesn't redirect existing report connections — it simply adds a badge indicating trustworthiness. Lineage also has nothing to do with granting user access; that's handled through workspace roles and permissions. Option C gets the logic backwards. Certification isn't derived from a dataset's dependencies in lineage — it's granted manually by authorized users based on data quality and governance standards. Promoting the legacy model while migrating away from it also sends a contradictory signal to users. Option D confuses endorsement with access control and misuses both tools. Promoting both models doesn't preserve access (again, that's a permissions function), and lineage records aren't something you manually remove — they reflect live relationships in the tenant. Study tip: On Power BI governance questions, always ask yourself: "Is this tool about discovery/visibility or about signaling quality?" Lineage = discovery; endorsement = quality signal. Keep those roles clean in your mind.

Question 7

A paginated report connects to a certified semantic model. The report author has not requested endorsement for the paginated report itself. A user can discover the report but does not have permission to query the semantic model.

Which statement best describes the user's situation?

  1. The report is automatically certified, and certification grants the user read access to its underlying semantic model.
  2. The report inherits the certified status of its source, so no additional permissions are needed to view the underlying data.
  3. The report can use a certified source, but its own endorsement and the user's data access remain separate considerations. (correct answer)
  4. The report loses its connection to the semantic model until both items receive the same endorsement status.
Explanation: When working with endorsement in Power BI, keep two things clearly separated in your mind: what endorsement means and what it controls. Endorsement (promoted or certified) is a discoverability and trust signal — it helps users find reliable content. It is entirely independent of data access permissions, which are governed by workspace roles, dataset permissions, and row-level security. That's exactly what answer C captures. A paginated report can legitimately connect to a certified semantic model without itself being endorsed, and a user discovering that report still needs explicit permission to query the underlying semantic model. Endorsement status of either item does not automatically grant data access to anyone. Answer A is wrong because certification never grants read access to a semantic model. These are separate systems — one signals quality, the other controls security. Conflating them is a classic trap on this exam. Answer B makes a similar mistake by suggesting that endorsement inheritance replaces permission requirements. Endorsement is not inherited between connected items, and even if it were, it would still not replace access control. A certified source is trustworthy, not automatically accessible. Answer D introduces a fabricated rule: there is no mechanism in Power BI that breaks a report's connection to a semantic model due to mismatched endorsement statuses. Reports and semantic models can have different endorsement levels (or none at all) and remain fully connected. Your study takeaway: on Power BI exam questions involving endorsement, always ask yourself — does this answer confuse trust/discoverability with access control? If yes, it's a distractor.

Question 8

A department creates a reusable semantic model that has been tested by its authors and is useful to several project teams. The enterprise data governance group has not yet reviewed it, and only designated members of that group are authorized to certify content.

What is the most appropriate action for the department?

  1. Promote the model now, and request certification later if it passes the organization's formal review. (correct answer)
  2. Certify the model now, because successful testing by the authors is sufficient evidence of enterprise approval.
  3. Leave the model unendorsed permanently, because promotion is available only after certification has expired.
  4. Assign consumers Build permission, because Build permission provides the same discovery signal as promotion.
Explanation: Whenever you see a Power BI question about endorsement, keep two things straight: promotion and certification are separate tiers, and each has different permission requirements. Promotion can be applied by anyone with Edit access to the content, while certification requires a specific, organization-designated group — typically a data governance body. In this scenario, the model is tested and ready for broader use, but the enterprise governance group hasn't reviewed it yet. That means certification is off the table for now — not because the model is unworthy, but because the authorization process hasn't completed. The right move is to promote the model so teams can discover and use it today, and then pursue certification once the formal review is done. That's exactly what answer A describes, making it the correct choice. Answer B is wrong because certification isn't granted based on author testing alone — it requires review by the designated governance group. Claiming certification without that approval violates Power BI's endorsement rules and organizational trust boundaries. Answer C reflects a fundamental misunderstanding: promotion is a standalone endorsement tier, not a fallback state after certification expires. There is no expiration logic that blocks promotion. Answer D confuses access control with discoverability. Build permission lets users create new content from a dataset, but it doesn't surface the model as organizationally vetted the way endorsement does — the two serve entirely different purposes. For the exam, remember this hierarchy: Promoted = ready for sharing, Certified = organizationally approved. You can promote without certifying, but you can never certify without proper authorization — regardless of how well the content performs internally.

Question 9

Two semantic models have nearly identical names and descriptions. Model A connects to a governed warehouse. Model B imports data from an analyst-maintained spreadsheet. A report connects to Model B, although its title contains the phrase "Warehouse Sales."

Which interpretation is most appropriate when reviewing lineage?

  1. The report should be treated as warehouse-based because matching business names determine lineage relationships.
  2. The report's actual connection to Model B determines its lineage, regardless of the report's title or description. (correct answer)
  3. Both models are upstream dependencies because similar names cause Power BI to consolidate related semantic models.
  4. Neither model is a dependency until one model is promoted or certified by the organization's administrators.
Explanation: When reviewing lineage in Power BI, remember that lineage is determined by actual data connections, not by names, labels, or descriptions. The Lineage view in Power BI service traces the real upstream and downstream relationships between data sources, dataflows, semantic models, and reports based on how they are technically wired together. In this scenario, the report physically connects to Model B — the analyst-maintained spreadsheet import. That connection is what Power BI's lineage view will display, making Model B the upstream dependency. The report's title containing "Warehouse Sales" is just metadata; it has no bearing on which model feeds the report. Answer B correctly reflects this: lineage follows the actual connection, full stop. Answer A is tempting because business users often assume that matching names signal a relationship, but Power BI lineage is a technical graph, not a semantic one. Names never determine lineage. Answer C introduces a fictional behavior — Power BI does not consolidate or merge semantic models based on naming similarity. Each model is a distinct artifact, and lineage only reflects explicit connections. Answer D confuses endorsement (promotion or certification) with dependency. Endorsement is a trust signal that helps users identify reliable content; it has nothing to do with whether a model appears as a dependency in lineage. An uncertified model can absolutely be an upstream dependency if something connects to it. Your takeaway: on exam questions about lineage, always ask yourself, "what is actually connected to what?" Titles, descriptions, and endorsement status are all metadata — only the real data connection defines a lineage relationship.

Question 10

An administrator plans to replace a dataflow used by multiple semantic models. Those models support both interactive reports and paginated reports across several workspaces. The administrator must first identify content that could be affected.

What should the administrator use as the primary starting point?

  1. The dataflow's endorsement status, because certified items maintain a complete list of all downstream report consumers.
  2. Lineage and impact information, because it identifies connected downstream items and the workspaces that may be affected. (correct answer)
  3. Semantic model permissions, because every downstream report owner must have direct access to the original dataflow.
  4. Report usage metrics, because every technical dependency produces a recorded report view before it can refresh.
Explanation: When you need to assess the blast radius of a change to a shared Power BI artifact — like a dataflow feeding multiple semantic models — your first instinct should be to trace the dependency chain, not audit permissions or usage logs. Power BI's lineage view (available in workspace settings) does exactly this: it maps the flow of data from sources through dataflows, semantic models, and finally to reports and dashboards, even across workspaces. It also surfaces impact analysis, which lets you see every downstream item that would be affected if you modify or remove the dataflow. That makes B the correct starting point. Choice A is tempting if you've seen endorsement badges in Power BI, but certification is a trust signal, not a dependency registry. A certified dataflow has no built-in list of "who consumes me" — endorsement says nothing about downstream connections. Choice C confuses access control with dependency mapping. While report owners may need permissions to underlying data, those permissions aren't stored in a way that reveals technical lineage. A semantic model can refresh from a dataflow without the report owner having explicit dataflow access at all. Choice D sounds logical at surface level, but usage metrics capture user interactions with reports — page views, refreshes initiated by people — not the automated refresh dependencies between artifacts. A semantic model could be connected to the dataflow and refresh nightly without ever producing a user-facing "view" that appears in metrics. Study tip: On Power BI exam questions about change management or impact assessment, look for answers involving lineage, impact analysis, or the workspace lineage view — these are the native tools designed specifically for tracing data dependencies.