Đại học Kinh tế Quốc dân – National Economics UniversityMICRO4ASIA
🏆 Arena
Sign inSign up
MICRO4ASIAThe six micro-credentials

Courses are delivered on the Limio learning platform. Limio

← Data Visualization in Business

You are viewing a public preview

Sign in to save your progress, do assignments, join discussions and get feedback.

Sign in
Module 4 — Building Advanced Charts and Analyzing Data in Tableau

4.1 · Advanced Charts in Tableau

Print / Save as PDF

In this lesson

Combination charts, dual-axis and synchronized dual-axis charts; histogram, box plot and density plot; waterfall, funnel, Pareto and bullet charts; Gantt and candlestick charts; density maps and heatmaps.

Learning outcomes

  • LO.5: Use Tableau Desktop and Tableau Public to connect data, organize fields and build visualizations.
  • LO.6: Select, build and refine charts that fit the analytical objective and the data.
  • LO.7: Apply calculations and analytical techniques to explore business data and interpret findings.

The lecture slides for advanced chart types will be added here; the Module 4 deck covers calculated fields (lesson 4.2) and sets, groups, bins and clusters (lesson 4.3).

Study notes

By the end of this lesson you can:

  • Build combination, dual-axis and synchronized dual-axis charts and decide when each is appropriate.
  • Explain what histograms, box plots and density plots reveal about a distribution.
  • Describe how waterfall, funnel, Pareto and bullet charts support business questions.
  • Describe Gantt and candlestick charts and the difference between a density map and a heatmap.
  • Judge when an advanced chart adds insight and when a basic chart is better.

Key idea: Advanced charts are specialised tools. Each answers one type of question very well (a cumulative effect, a distribution, progress against a target), so use one only when that question is the one you are asking.

Combination and dual-axis charts

A combination chart shows two or more measures with different mark types in one view, for example bars for Sales and a line for Profit Ratio.

  • Dual axis: put two measures on Rows (or Columns), right-click the second and choose Dual Axis. The two measures now share the view but each has its own axis and its own Marks card, so you can set bar for one and line for the other.
  • Synchronized dual axis: right-click the second axis and choose Synchronize Axis. Use it when both measures have the same unit (for example Sales and Sales Target), so the same height means the same value. If the units differ (money and a percentage), do not synchronize; label both axes clearly instead.
  • Shared (blended) axis: drag a second measure onto an existing axis. Tableau uses Measure Names and Measure Values so several measures share one scale.

Module 1 warned about arbitrary dual axes: two independent scales can make unrelated series look correlated. Always label both axes and explain why the two measures belong together.

Distribution charts

  • Histogram: counts how many records fall into each bin of a measure. Show Me builds one from a single measure and creates a bin field automatically; you can change the bin size.
  • Box plot (box-and-whisker): the box spans the 25th to 75th percentile (the interquartile range, IQR), the line inside is the median, and by default the whiskers extend to the furthest data points within 1.5 × IQR. Points beyond are potential outliers. Box plots need individual marks (for example one per order or customer) so the dimension that defines those marks goes on Detail.
  • Density plot: a smoothed histogram showing where values concentrate. Show Me has no one-click density curve; a line over small bins gives a similar picture.

Business process and performance charts

  • Waterfall chart: shows how a starting value is built up or reduced by positive and negative contributions to a total. In Tableau: a running total of the measure, mark type Gantt Bar, and size set to the negative of the measure so each bar hangs from the running total.
  • Funnel chart: shows how a quantity shrinks through ordered stages (leads, quotes, orders, repeat orders). Bars are sorted by stage and often centred; a sorted bar chart is a simple alternative.
  • Pareto chart: bars sorted from largest to smallest plus a line of cumulative percentage on a second axis. It shows whether a few items account for most of the total (the 80/20 idea). The line uses a running-total table calculation with a secondary percent-of-total calculation.
  • Bullet chart: a compact bar of actual performance against a target marker and qualitative ranges (poor, satisfactory, good). Show Me builds it from two measures; if the bar and target are the wrong way round, use Swap Reference Line Fields.

Time and map-based charts

  • Gantt chart: horizontal bars that start at a date and whose length is a duration. Mark type Gantt Bar, a start date on Columns, the duration on Size. Used for project schedules or shipping times.
  • Candlestick chart: for prices over time, each mark shows the high–low range as a thin line and the open–close range as a thicker body, coloured by rise or fall. You build it with Gantt bars or a dual axis.
  • Density map: points drawn with the Density mark type, so colour intensity shows where many locations or events concentrate.
  • Heatmap: a grid of two dimensions with colour (and sometimes size) showing a measure. In Tableau, a highlight table uses square marks coloured by the measure; Show Me's heat map also varies the square size.

Worked example

Question from the logistics team: "How long does each order take to ship, and is Same Day really same day?" Use the course Orders table.

  1. Duration field. Create a calculated field Days to Ship = DATEDIFF('day', [Order Date], [Ship Date]).
  2. Choose the chart. Each order has a start (Order Date) and a duration (days until Ship Date). That is the Gantt pattern.
  3. Build. Put continuous DAY(Order Date) on Columns and Order ID on Rows, set mark type to Gantt Bar and drag AVG(Days to Ship) to Size (it is the same value for every line of an order). Colour by Ship Mode.
  4. Focus. Filter to one month; a Gantt chart of every order for three years would be unreadable.
  5. Read. Same Day bars should have length 0; any Same Day order with a visible bar breaks the service promise and deserves a look. Slower modes such as Standard Class are expected to have the longest bars.
  6. Compare with a simpler chart. To compare ship modes overall, a box plot of Days to Ship by Ship Mode (Order ID on Detail) summarises the spread better than thousands of Gantt bars. The Gantt view is best for inspecting a short period order by order.

Common mistakes

  • Synchronizing axes with different units. Sales and a percentage on one scale make the percentage line flat.
  • Box plots with one mark per category. Without a dimension on Detail there is no distribution to summarise.
  • Unsorted Pareto bars. The cumulative line only makes sense when bars are sorted descending.
  • Using an advanced chart to impress. If a sorted bar chart answers the question, use it.

Summary

  • Dual-axis charts combine two measures; synchronize only when the units match.
  • Histograms, box plots and density plots show distributions at different levels of summary.
  • Waterfall, funnel, Pareto and bullet charts answer specific performance and process questions.
  • Gantt charts show durations; candlesticks show price ranges; density maps and heatmaps show concentration.
  • Choose an advanced chart only when its specific question is the question you are asking.

Flashcards · Advanced Charts in Tableau

10 words
WordMeaningExample
  • Dual-axis chart

    Two measures in one view, each with its own axis and Marks card.

    Bars for SUM(Sales), line for Profit Ratio.

  • Synchronize Axis

    Aligns the scales of a dual axis so equal heights mean equal values; use only with the same units.

    Sales and Sales Target on synchronized axes.

  • Box plot

    Shows median, interquartile range and whiskers (default 1.5 × IQR) to summarise a distribution and outliers.

    Order value by Segment, Order ID on Detail.

  • Waterfall chart

    Shows how positive and negative contributions build a running total, using Gantt bars sized by the negative value.

    Profit by Sub-Category building to total profit.

  • Pareto chart

    Sorted bars plus a cumulative percentage line, showing whether a few items make most of the total.

    Sub-Category sales with running % of total.

  • Bullet chart

    A bar of actual value against a target marker and performance ranges.

    2025 Region sales vs Sales Target.

  • Funnel chart

    Ordered stages shown as shrinking bars to reveal where volume drops.

    Leads → quotes → orders → repeat orders.

  • Gantt chart

    Bars that start at a date and whose length shows a duration.

    Order Date start, size = Days to Ship.

  • Candlestick chart

    Shows open, close, high and low values for each period, typically for prices.

    Daily share price of a listed retailer.

  • Density map

    A map using the Density mark so colour intensity shows where points concentrate.

    Concentration of store locations in a city.

Practice in Tableau — Lab 4A: Advanced charts

Use Sales_Data.xlsx (Customer, Product and Targets related to Orders).

  1. Dual axis. Monthly SUM(Sales) as bars and Profit Ratio (SUM([Profit]) / SUM([Sales])) as a line on a second axis. What happens to profitability in the months with high sales?
  2. Histogram. Create bins on Quantity (bin size 1). Check: the most common quantity is 3, the largest is 11.
  3. Box plot. Order value by Segment: use a FIXED LOD for order value (see lesson 4.2) or SUM(Sales) at Order ID level, then Show Me → box-and-whisker. Which segment has the widest spread?
  4. Pareto chart. Sub-Category sales sorted descending (bars) plus a running % of total (line, dual axis). Check: 6 of the 13 sub-categories make up 80% of sales.
  5. Waterfall. Profit by Sub-Category with a running total, using a Gantt bar sized by -[Profit]. Check: only Tables (-19,314) and Bookcases (-18,090) lose money.
  6. Bullet chart. 2025 sales by Region against Sales Target from the Targets table. Check: every region beat its 2025 target, e.g. North 408,554 against a target of 368,100.

For each chart, note one situation in which a basic chart would have been the better choice.