Top Tableau Skills Every Beginner Must Learn in 2026

Before building any sheet in Tableau:

[ ] Connect to the source (Excel, CSV, database)
[ ] Use Data Interpreter for messy Excel sheets
[ ] Pivot wide month columns into rows (Month, Value)
[ ] Check data types: dates as dates, numbers as numbers
[ ] Rename fields to business-friendly names
[ ] Hide fields you will not use
[ ] Create relationships on the correct keys
[ ] Compare a total with the source to confirm

-- Custom SQL or a database view feeding Tableau
-- Pre-aggregate large data to keep dashboards fast
SELECT
  CAST(order_date AS DATE)  AS order_date,
  region,
  category,
  SUM(sales)                AS sales,
  SUM(profit)               AS profit,
  COUNT(DISTINCT order_id)  AS orders
FROM orders
WHERE order_date >= '2025-01-01'
GROUP BY CAST(order_date AS DATE), region, category;

// Discount band
IF [Discount] = 0 THEN "No discount"
ELSEIF [Discount] <= 0.2 THEN "Up to 20%"
ELSE "Above 20%"
END

// Days to ship
DATEDIFF('day', [Order Date], [Ship Date])

// Late shipment flag
IF DATEDIFF('day', [Order Date], [Ship Date]) > 4 THEN "Late" ELSE "On time" END

// Customer count
COUNTD([Customer ID])

// Percent of total (table calculation; set 'Compute using')
SUM([Sales]) / TOTAL(SUM([Sales]))

// Moving 3-month average of sales
WINDOW_AVG(SUM([Sales]), -2, 0)

// Rank categories by profit
RANK(SUM([Profit]))

// Average order value per customer, independent of the view
{ FIXED [Customer ID] : SUM([Sales]) / COUNTD([Order ID]) }

// 1. Create a string parameter [Choose Measure]
//    with values: "Sales", "Profit", "Orders"

// 2. Calculated field: Selected Measure
CASE [Choose Measure]
  WHEN "Sales"  THEN SUM([Sales])
  WHEN "Profit" THEN SUM([Profit])
  WHEN "Orders" THEN COUNTD([Order ID])
END

// 3. Put Selected Measure on Rows and show the parameter control.
// 4. Title the sheet: <Parameters.Choose Measure> by Region

Dashboard performance checklist

[ ] Use an extract for large or slow sources
[ ] Hide unused fields before creating the extract
[ ] Aggregate data to the needed level
[ ] Limit marks (avoid tables with thousands of rows)
[ ] Fewer quick filters; use actions instead
[ ] Use context filters wisely for top N
[ ] Keep dashboards to a few focused sheets
[ ] Run Help > Settings and Performance > Start Performance Recording

Skill                       | Practised | Used in a project | Can explain
----------------------------|-----------|-------------------|------------
Relationships vs joins      |           |                   |
Unions and pivots           |           |                   |
Chart choice                |           |                   |
Calculated fields           |           |                   |
Table calculations          |           |                   |
LOD expressions             |           |                   |
Parameters                  |           |                   |
Dashboard actions           |           |                   |
Sets and groups             |           |                   |
Performance tuning          |           |                   |

Daily 30-minute Tableau drill

Day 1  Rebuild a chart type you have not used
Day 2  Write 3 calculated fields for a new question
Day 3  One table calculation, change 'Compute using'
Day 4  One LOD expression, compare with a table calc
Day 5  Add a parameter to an existing view
Day 6  Build a small dashboard with one action
Day 7  Review: what was hard? Repeat it next week

Which chart should I use?

Trend over time        -> line
Compare categories     -> sorted bar
Part of whole          -> stacked bar / treemap (few parts)
Relationship           -> scatter
Distribution           -> histogram / box plot
Geography              -> filled or symbol map
Single key number      -> big number (KPI card)
Ranking                -> sorted bar with top N