pbPassingBI
/
quicksight/exams/quicksight-exam-full

QuickSight full mock exam

11 questions sampled across all 4 modules. Timed like the real thing: answer everything, then review the explanations.

  1. You edit an analysis but consumers see the old figures. Why?

    1. SPICE has not refreshed
    2. The dashboard has not been republished
    3. Permissions are wrong
    4. The dataset is in direct query

    Answer: The dashboard has not been republished — A dashboard is a published snapshot; analysis edits appear only after republishing.

  2. Which feature requires Enterprise edition?

    1. SPICE
    2. Row-level security
    3. Calculated fields
    4. Dashboards

    Answer: Row-level security — RLS, VPC connections and directory integration are Enterprise features.

  3. A SPICE refresh suddenly fails across several datasets. First thing to check?

    1. Chart configuration
    2. SPICE capacity in the admin console
    3. Dashboard permissions
    4. Parameter defaults

    Answer: SPICE capacity in the admin console — Exhausted account capacity is the usual cause, often from abandoned datasets.

  4. Where is complex join logic best maintained?

    1. Custom SQL inside the dataset
    2. A view in the warehouse
    3. An analysis-level calculated field
    4. A dashboard filter

    Answer: A view in the warehouse — A warehouse view is version-controlled, reusable and easier to debug than embedded SQL.

  5. A calculation two reports both need belongs in:

    1. Each analysis
    2. The dataset
    3. The dashboard
    4. A parameter

    Answer: The dataset — Shared definitions prevent reports disagreeing.

  6. Which computes revenue per customer regardless of visual grouping?

    1. sum(revenue)
    2. sumOver(sum({revenue}), [{customer_id}], PRE_AGG)
    3. runningSum(sum(revenue))
    4. percentOfTotal(sum(revenue))

    Answer: sumOver(sum({revenue}), [{customer_id}], PRE_AGG) — A level-aware aggregation pins the computation to customer_id independent of the visual.

  7. Numbers in one visual do not match the rest of the dashboard. Likely cause?

    1. SPICE lag
    2. Filter applied at single-visual scope
    3. Wrong visual type
    4. Missing geographic role

    Answer: Filter applied at single-visual scope — Scope determines which visuals a filter touches; a single-visual filter is easy to forget.

  8. Which usually reads better than five overlapping lines?

    1. A pie chart
    2. Small multiples
    3. A KPI visual
    4. A table

    Answer: Small multiples — One small chart per category on a shared scale.

  9. In an RLS rules dataset, a blank value in a restriction column means:

    1. No access
    2. Access to all values of that field
    3. The row is ignored
    4. An error

    Answer: Access to all values of that field — Blank grants universal access for that field — useful for admins, dangerous by accident.

  10. In a multi-tenant embed, what actually prevents tenant A seeing tenant B data?

    1. A URL parameter
    2. Row-level security or session tags
    3. The iframe sandbox
    4. Dashboard permissions

    Answer: Row-level security or session tags — Parameters control display only; RLS or session tags restrict what can be queried.

  11. Publishing an analysis as a dashboard means:

    1. They stay linked live
    2. Readers see a published snapshot while you keep editing
    3. The analysis is deleted
    4. Readers can edit it

    Answer: Readers see a published snapshot while you keep editing — You republish when ready.