Salesforce Certified Tableau Data Analyst — Practice Set 1

Question 1

An analyst is connecting Tableau Desktop to a large on-premises relational database and wants the workbook to always reflect the very latest transactional data, even if that means slower query performance during peak hours. Which connection type should the analyst choose?

Accepted Answer

A live connection queries the source database directly every time the view is interacted with, guaranteeing the most current data at the cost of depending on network/database performance. An extract copies data into Tableau's in-memory engine as of the last refresh, so it would not reflect real-time changes. A published snapshot and an exported .hyper file are both extract-based artifacts, not live connections.

Question 2

A data analyst is auditing a newly connected customer table and notices some rows have missing postal codes, inconsistent capitalization in the country field, and a handful of duplicate customer IDs. Which activity is the analyst performing?

Accepted Answer

Identifying missing values, inconsistent formatting, and duplicate records is exactly what assessing data quality for completeness, consistency, and accuracy means during data preparation. LOD calculations, publishing, and building stories are unrelated later-stage activities in the analytics workflow.

Question 3

In Tableau Prep, which two techniques can be used to combine multiple input tables that have different but related structures? (Choose two.)

Accepted Answer

Tableau Prep flows combine data primarily through unions (stacking rows from tables with matching structures) and joins (combining columns from related tables based on matching keys). Trend lines and reference bands are Analytics pane features used in Tableau Desktop for visual analysis, not data combination techniques in Prep.

Question 4

A survey export contains one column per question (Q1, Q2, Q3...) but the analyst needs each response in its own row to build a single visualization across all questions. Which Tableau Prep technique should be used to reshape the data?

Accepted Answer

Pivoting reshapes wide data (one column per question) into a long format (one row per question/response pair), which is exactly the transformation needed here. Aggregation summarizes values rather than reshaping columns to rows, a union stacks similarly structured tables, and a filter only removes rows rather than reshaping the table.

Question 5

A field called 'Order Priority' contains the numeric codes 1, 2, and 3, representing discrete priority tiers rather than a quantity to be summed. What should the analyst do to ensure Tableau treats this field appropriately in visualizations?

Accepted Answer

Even though 'Order Priority' is numeric, it represents categorical tiers, not a quantity to aggregate — converting it to discrete tells Tableau to treat each value as a distinct category (useful for coloring, headers, etc.) rather than a continuous, summable axis value. Converting it to a measure would push it in the wrong direction, and creating an extract or filtering the field does not address how the field's role is interpreted.

Question 6

A workbook currently connects to a legacy 'Sales_2023' extract. The organization has migrated to a new published data source, 'Sales_Unified', that contains the same field names and structure plus additional years of history. What is the most efficient way to update every worksheet and dashboard in the workbook to use the new source, without manually rebuilding each view?

Accepted Answer

Replacing the data source lets an author swap the underlying connection for every sheet and dashboard at once, as long as field names align between the old and new sources — no rebuilding required. Deleting and rebuilding worksheets, recreating a workbook from scratch, or unioning the old and new sources are all far more time-consuming and, in the union's case, would duplicate rather than replace the data.

Question 7

An analyst needs a calculated field that returns the number of days between an Order Date and a Ship Date. Which function category should be used?

Accepted Answer

DATEDIFF is a date calculation function purpose-built to compute the interval between two dates, which is exactly what's needed here. String functions manipulate text, aggregate functions summarize numeric values across rows, and type conversion functions change a field's data type rather than compute a duration.

Question 8

A dashboard shows quarterly revenue and needs a new column showing each quarter's revenue as a percentage of the year's total revenue, computed within the table itself rather than in the underlying data source. Which feature should the analyst use?

Accepted Answer

Table calculations, such as the built-in 'Percent of Total' quick table calculation, compute values based on what's already in the visualization (e.g., each quarter's share of the yearly total) without altering the underlying data source. A data source filter only removes rows, a geographic role affects mapping, and a Prep union combines tables — none compute a percentage-of-total within a view.

Question 9

A regional sales manager wants to see only the top 10 performing sales reps in a bar chart, excluding everyone else. Which filter configuration achieves this?

Accepted Answer

A Top N filter configuration is designed precisely to limit a view to the highest (or lowest) N members ranked by a chosen measure, such as showing only the top 10 sales reps. A context filter with no conditions filters nothing, a wildcard filter matches text patterns rather than rankings, and a parameter without an attached filter has no filtering effect on its own.

Question 10

A dashboard has several filters applied to a large data source. Performance is slow because Tableau reprocesses all filters against the full data set for every query. Which technique can improve performance by having Tableau apply certain filters first, reducing the data considered by subsequent filters and calculations?

Accepted Answer

Adding a filter to context forces Tableau to materialize a temporary, filtered subset of data before applying other filters and calculations, which can significantly improve performance when a high-impact filter greatly reduces the working data set. Simply making filters 'quick filters' doesn't change processing order, removing filters changes the analysis entirely, and switching to a live connection typically adds — not reduces — query overhead.

Question 11

An analyst wants end users to be able to dynamically choose which measure (Sales, Profit, or Quantity) is displayed on a chart's axis using a dropdown control. Which feature should be used?

Accepted Answer

A parameter, referenced inside a calculated field, lets users pick from a list of options (like Sales, Profit, or Quantity) and dynamically swap which measure drives the visualization. A set defines a subset of dimension members, a group combines related dimension members, and a geographic role only affects location-based mapping — none of these let users swap between measures.

Question 12

An analyst wants to group individual customers into 'High Value', 'Medium Value', and 'Low Value' segments based on manually defined lifetime spend ranges, for use consistently across multiple worksheets. Which feature best supports this?

Accepted Answer

A group combines related dimension members (or, when built on a calculated condition, arbitrary custom categories like value tiers) into a single, reusable field that can be applied across worksheets. A bin creates equal-sized numeric buckets automatically rather than custom-named tiers, a hierarchy organizes related dimensions for drill-down (e.g., Country > State > City), and a trend line is a statistical overlay, not a categorization tool.

Question 13

An analyst wants to visualize sales concentration across countries by shading each country a darker color based on total sales volume. Which map type is most appropriate?

Accepted Answer

A choropleth, or filled map, shades entire geographic areas (like countries) using color intensity to represent a measure, which is exactly the 'sales concentration by country' scenario described. A symbol map places sized/colored marks at specific points rather than shading whole regions, a density map shows the concentration of individual data points as a heat-map, and a plain point mark layer doesn't fill geographic boundaries.

Question 14

A leadership team wants to see a projected sales trend for the next four quarters based on historical patterns, without manually building a statistical model. Which Analytics pane feature should the analyst add to the view?

Accepted Answer

Tableau's forecasting feature uses default settings to automatically project future values based on historical trends and seasonality, exactly matching the leadership team's request. Reference bands and distribution bands shade a range around existing data rather than projecting forward, and totals/subtotals only summarize existing values.

Question 15

An analyst needs a calculated field that computes each customer's total lifetime sales, and that total must remain fixed at the customer level even when the view is later broken down by Region and Product Category. Which type of calculation should be used?

Accepted Answer

A FIXED LOD calculation computes an aggregation at an explicitly specified level of detail (here, per customer) and holds that value constant regardless of what other dimensions are later added to the view, exactly matching the requirement. A quick table calculation depends on the current view's structure and would change as dimensions are added, a plain SUM without LOD would recompute at whatever level the view currently shows, and a parameter is a user input control, not an aggregation mechanism.

Question 16

An analyst needs a calculation that shows each product's sales relative to the total sales of its category, but the category boundary must be ignored when the view is filtered down to a single region so that the comparison still spans all regions. Which two LOD expression types could plausibly be involved in solving this correctly, depending on how filters interact with the calculation? (Choose two.)

Accepted Answer

A FIXED LOD calculation can compute the category total independent of the view's dimensions (and, depending on configuration, independent of certain filters), while an EXCLUDE LOD calculation can be used to remove a specific dimension (like Region) from an otherwise view-level aggregation so the comparison spans all regions. INCLUDE adds a dimension to a coarser aggregation rather than removing a filtering dimension, and a quick table calculation is view-dependent and would not reliably ignore a region filter's effect on the category total.

Question 17

An analyst wants to compare the trend of sales over time for five different regions, all on the same set of axes so overlapping trends are easy to spot. Which basic chart type is best suited for this?

Accepted Answer

A line chart is the standard choice for showing trends over a continuous variable like time, and multiple lines (one per region) can be overlaid on the same axes for easy comparison. A pie chart shows part-to-whole composition at a single point, a box plot shows the distribution/spread of a measure, and a tree map shows hierarchical part-to-whole proportions using nested rectangles — none are designed for time-trend comparison across categories.

Question 18

A bar chart currently sorts products alphabetically, but a stakeholder wants them ordered to match a specific internal product-tier ranking that doesn't match alphabetical or value-based order. What should the analyst use?

Accepted Answer

A custom sort lets an author manually define the exact display order of dimension members, which is necessary when the desired order doesn't correspond to alphabetical order or any measure's value. A trend line is a statistical overlay, a union is a Prep data-combination technique, and a geographic role affects map interpretation — none control the display order of dimension members in a chart.

Question 19

An author is assembling a dashboard with three worksheets, a company logo image, and a text header, and wants precise control over how each element resizes as the browser window changes. Which feature should the author use to arrange these elements?

Accepted Answer

Containers (horizontal/vertical layout containers) and their layout options are the mechanism for precisely arranging and controlling how worksheets, images, and text objects resize and reflow on a dashboard. A story is a separate, sequential presentation format for a series of visualizations, a set action is a dashboard interactivity feature (not layout), and a data source filter restricts data rather than arranging dashboard objects.

Question 20

A dashboard has a summary chart and a detail chart. The author wants clicking a bar in the summary chart to filter the detail chart to only that category, without requiring a separate filter control. Which dashboard feature should the author configure?

Accepted Answer

A filter action lets a user's interaction with one worksheet (like clicking a bar) automatically filter one or more other worksheets on the dashboard, exactly the behavior described. Story point navigation moves between pre-built story panels rather than filtering dynamically, a reference line is a static visual annotation, and a geographic role only affects map interpretation.

Question 21

A dashboard has a detailed breakdown panel that should stay hidden until a user clicks a 'Show Details' button, and should hide again when a 'Hide Details' button is clicked, without navigating to a different dashboard. Which combination of features accomplishes this?

Accepted Answer

Dynamic zone visibility, combined with show/hide buttons, lets an author toggle a dashboard zone's visibility on and off in place, exactly matching the described behavior. A story with two story points would require navigating between separate story panels rather than toggling visibility in place, a data source filter affects which rows are queried (not whether a zone is shown), and a geographic role has nothing to do with visibility.

Question 22

An author wants every visualization in a workbook to use the same corporate color palette (specific hex codes for brand colors not included in Tableau's default palettes). What should the author do?

Accepted Answer

Adding a custom color palette (defined with specific hex codes) lets an author apply consistent brand colors across visualizations, which default built-in palettes may not support. A geographic role affects location interpretation, a union in Prep is a data-combination technique, and a reference band is an Analytics pane overlay — none control color palettes.

Question 23

A dashboard will be viewed on both desktop monitors and mobile phones, and the author wants the layout to automatically rearrange (e.g., stacking panels vertically) depending on the device. Which feature should be used?

Accepted Answer

Tableau lets authors create device-specific dashboard layouts (Desktop, Tablet, Phone) so the arrangement automatically adapts to the viewer's device, delivering a responsive design. A single fixed-size layout would not adapt across devices, and a data source filter or parameter control has no effect on layout responsiveness.

Question 24

An analyst wants to be automatically notified by email whenever a key KPI on a published dashboard crosses a certain threshold, without needing to check the dashboard manually every day. Which feature should be configured?

Accepted Answer

A data-driven alert monitors a specific mark or value against a defined threshold and sends an email notification when that condition is met — exactly the scenario described. A subscription emails a snapshot of a view on a regular schedule regardless of any threshold, a custom view simply saves a personalized view state, and a scheduled extract refresh only updates the underlying data, not user notifications.

Question 25

A dashboard's data source is published to Tableau Cloud and needs its underlying extract updated automatically every night at 2 AM without any manual intervention. What should be configured?

Accepted Answer

Scheduling a data extract refresh is exactly how an author configures automatic, recurring updates to a published extract's underlying data at a set time, such as nightly at 2 AM. A data-driven alert notifies users of a threshold being crossed rather than refreshing data, a custom view saves a personalized filter/sort state, and a filter action is a dashboard interactivity feature unrelated to data refresh scheduling.

Analytics-DA-201 — Salesforce Certified Tableau Data Analyst