Data Visualisation and Dashboard Design
Learn to choose and build charts that answer the question honestly and show what is uncertain. No coding: you work on the data of Raket Parts, a fictional maker of rocket parts.
You will be able to
- choose a chart that fits the question
- set honest scales, sorting and labels
- show a range beside a figure
- set small groups apart
The question picks the chart
Read a chart’s scope and choose a comparison, share, spread, relationship or time view without confusing the question.
Honest charts
Choose honest bar and line scales, sort and label values clearly, and give charts precise titles with sensible units.
Show what you are not sure of
Show an observed range beside a figure, set small groups apart and avoid treating historical variation as a prediction.
Diagnose misleading visuals
Measure how far a chart distorts its data, judge whether that changes the conclusion and redesign it with honest marks and words.
Charts everyone can read
Check a chart or dashboard for colour-only meaning, contrast, keyboard order and text alternatives, and decide which fixes must ship before release.
Specifying a decision-useful report
Write down what a report must show before anyone builds it.
Building a decision-focused dashboard
Build a dashboard around the decisions it supports.
Verifying a dashboard before release
Check filters, refresh and figures before people rely on it.
Is a released dashboard still useful?
Judge whether a live dashboard still supports its decision.
Integration clinic: the quarterly leadership pack
Bring every method together on one leadership pack.
Final assessment
A full report design on an unseen case.
Data Visualisation / Module 1
The question picks the chart
Mira and Bram have different questions for the same data. Read what a chart really shows, then choose a form that helps the reader decide. One request is better answered with a table.
Sections
Data Visualisation / Module 1 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
Data Visualisation / Module 2
Honest charts
An operations pack makes delivery performance look as though it has collapsed. Jelmer checks the recorded share, the axis and the labels before recommending a new view. Correct figures need an honest shape.
This module builds on module 1. If you are new to this, start with module 1.
Sections
Data Visualisation / Module 2 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
Data Visualisation / Module 3
Show what you are not sure of
Bram wants one ranked country figure and a single headline number. Some country groups contain few customers, and a single observed value hides variation between months. Jelmer keeps the range and group size visible.
This module builds on module 2. If you are new to this, start with module 2.
Sections
Data Visualisation / Module 3 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
Data Visualisation / Module 4
Diagnose misleading visuals
Bram's team has drafted the charts for the quarterly review pack, with data to October 2025. Every number is right, yet Mira reads several of them the wrong way. Jelmer measures each distortion, decides which ones change the conclusion, and redesigns those.
This module builds on module 3. If you are new to this, start with module 3.
Sections
Data Visualisation / Module 4 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
Data Visualisation / Module 5
Charts everyone can read
Mira's delivery dashboard is due for release. It shows the five delivery outcomes in red, green, orange and two greys, has light grey labels, a value you can only see by hovering, and a filter row the keyboard cannot reach. Jelmer checks it so every reader can use it, then a screen-reader user joins the audience a day before release.
This module builds on module 4. If you are new to this, start with module 4.
Sections
Data Visualisation / Module 5 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
Data Visualisation / Module 1 / 1.1
1.1 Read a chart before trusting it
Section 1.1 complete
Next: 1.2 Five kinds of question
Data Visualisation / Module 1 / 1.2
1.2 Five kinds of question
Section 1.2 complete
Next: 1.3 Many slices, no pie
Data Visualisation / Module 1 / 1.4
1.4 When a table wins
Section 1.4 complete
Next: Module 1 check
Continue to the module check Recap this section Back to module 1
Data Visualisation / Module 1 / Check
Module 1 check
Module 1 complete
Next: Module 2, Honest charts
Continue to module 2 Try it on your own question Recap module 1 Back to module 1
Data Visualisation / Module 2 / 2.2
2.2 Order and labels
Section 2.2 complete
Next: 2.3 Titles that say the finding
Data Visualisation / Module 2 / 2.3
2.3 Titles that say the finding
Section 2.3 complete
Next: 2.4 Precision and units
Data Visualisation / Module 2 / 2.4
2.4 Precision and units
Section 2.4 complete
Next: Module 2 check
Continue to the module check Recap this section Back to module 2
Data Visualisation / Module 2 / Check
Module 2 check
Module 2 complete
Next: Module 3, Show what you are not sure of
Continue to module 3 Try it on your own question Recap module 2 Back to module 2
Data Visualisation / Module 3 / 3.1
3.1 Hiding the range hides the risk
Section 3.1 complete
Next: 3.2 Small groups set apart
Data Visualisation / Module 3 / 3.3
3.3 Ranges on bars
Section 3.3 complete
Next: 3.4 One number, with its range
Data Visualisation / Module 3 / 3.4
3.4 One number, with its range
Section 3.4 complete
Next: Module 3 check
Continue to the module check Recap this section Back to module 3
Data Visualisation / Module 3 / Check
Module 3 check
Module 3 complete
Next: Module 4, Diagnose misleading visuals
Continue to module 4 Try it on your own question Recap module 3 Back to module 3
Data Visualisation / Module 4 / 4.1
4.1 Correct numbers, wrong picture
Section 4.1 complete
Next: 4.2 Measure the distortion
Data Visualisation / Module 4 / 4.2
4.2 Measure the distortion
Section 4.2 complete
Next: 4.3 Does it change the conclusion?
Data Visualisation / Module 4 / 4.3
4.3 Does it change the conclusion?
Section 4.3 complete
Next: 4.4 Honest words, kept caveats
Data Visualisation / Module 4 / 4.4
4.4 Honest words, kept caveats
Section 4.4 complete
Next: Module 4 check
Continue to the module check Recap this section Back to module 4
Data Visualisation / Module 4 / Check
Module 4 check
Module 4 complete
Next: Module 5, Charts everyone can read
Continue to module 5 Try it on your own question Recap module 4 Back to module 4
Data Visualisation / Module 5 / 5.1
5.1 Colour alone cannot carry status
Section 5.1 complete
Next: 5.2 Contrast you can measure
Data Visualisation / Module 5 / 5.2
5.2 Contrast you can measure
Section 5.2 complete
Next: 5.3 Keyboard, reading order and hover
Data Visualisation / Module 5 / 5.3
5.3 Keyboard, reading order and hover
Section 5.3 complete
Next: 5.4 Text alternatives, a table route and the release deadline
Data Visualisation / Module 5 / 5.4
5.4 Text alternatives, a table route and the release deadline
Section 5.4 complete
Next: Module 5 check
Continue to the module check Recap this section Back to module 5
Data Visualisation / Module 5 / Check
Module 5 check
Module 5 complete
Next: Module 6, coming later
Back to the Path Try it on your own question Recap module 5 Back to module 5
Data Visualisation / Module 1 / Recap
Recap card
1.1 Read a chart before trusting it
Read a chart before trusting it
- Read the caption first. Name the measure and unit.
- Check scope. Name the filter and invoice period.
- Check scale. Compare bar lengths only with a stated baseline.
Common mistake
The trap. A large invoiced-sales bar cannot prove deliveries were on time.
Data Visualisation / Module 1 / Recap
Recap card
1.2 Five kinds of question
Five kinds of question
- Comparison: sorted bars for named categories.
- Share or spread: a divided bar for parts, a distribution for variation.
- Change or relationship: an ordered line for time, paired dots for numeric relationships.
Common mistake
The trap. A column containing dates does not force a line; the question does.
Data Visualisation / Module 1 / Recap
Recap card
1.3 Many slices, no pie
Many slices, no pie
- Sort categories. Put larger measured amounts first.
- Label directly. Keep euros and the invoice period visible.
- Preserve meaning. Changing the form need not change the figures.
Common mistake
The trap. A pie with many slices hides rank even when the numbers are correct.
Data Visualisation / Module 1 / Recap
Recap card
1.4 When a table wins
When a table wins
- Need the exact amount? Prefer a table.
- Need the pattern? Use a suitable labelled chart.
- Check absolutes. “All” can fail if another reason exists.
Common mistake
The trap. A chart can look polished and still make a lookup needlessly difficult.
Data Visualisation / Module 1 / Recap
Recap card
Module 1: The question picks the chart
Read a chart before trusting it
- Read the caption first. Name the measure and unit.
- Check scope. Name the filter and invoice period.
- Check scale. Compare bar lengths only with a stated baseline.
Common mistake
The trap. A large invoiced-sales bar cannot prove deliveries were on time.
Five kinds of question
- Comparison: sorted bars for named categories.
- Share or spread: a divided bar for parts, a distribution for variation.
- Change or relationship: an ordered line for time, paired dots for numeric relationships.
Common mistake
The trap. A column containing dates does not force a line; the question does.
Many slices, no pie
- Sort categories. Put larger measured amounts first.
- Label directly. Keep euros and the invoice period visible.
- Preserve meaning. Changing the form need not change the figures.
Common mistake
The trap. A pie with many slices hides rank even when the numbers are correct.
When a table wins
- Need the exact amount? Prefer a table.
- Need the pattern? Use a suitable labelled chart.
- Check absolutes. “All” can fail if another reason exists.
Common mistake
The trap. A chart can look polished and still make a lookup needlessly difficult.
Data Visualisation / Module 2 / Recap
Recap card
2.1 Where the axis starts
Where the axis starts
- Bar lengths: start at zero.
- Rate lines: a labelled focused scale may be useful.
- Define the share: confirmed on-time divided by confirmed on-time and late.
Common mistake
The trap. A non-zero line axis is not automatically dishonest, but a truncated amount bar is misleading.
Data Visualisation / Module 2 / Recap
Recap card
2.2 Order and labels
Order and labels
- Order by value when the question is a category ranking.
- Label values with units.
- Keep time ordered by date rather than value.
Common mistake
The trap. An attractive but unsorted chart makes the reader work harder to see the answer.
Data Visualisation / Module 2 / Recap
Recap card
2.3 Titles that say the finding
Titles that say the finding
- State the comparison in the title.
- Name the denominator when a rate excludes unknown outcomes.
- Avoid causal language without supporting evidence.
Common mistake
The trap. A rate difference is real, but a title that says what caused it can be made up.
Data Visualisation / Module 2 / Recap
Recap card
2.4 Precision and units
Precision and units
- Exact lookup: keep exact euro amounts.
- Executive overview: use clearly labelled €m figures.
- Rates: use one decimal point and a stated group.
Common mistake
The trap. More decimals can make an honest chart look precise beyond its purpose.
Data Visualisation / Module 2 / Recap
Recap card
Module 2: Honest charts
Where the axis starts
- Bar lengths: start at zero.
- Rate lines: a labelled focused scale may be useful.
- Define the share: confirmed on-time divided by confirmed on-time and late.
Common mistake
The trap. A non-zero line axis is not automatically dishonest, but a truncated amount bar is misleading.
Order and labels
- Order by value when the question is a category ranking.
- Label values with units.
- Keep time ordered by date rather than value.
Common mistake
The trap. An attractive but unsorted chart makes the reader work harder to see the answer.
Titles that say the finding
- State the comparison in the title.
- Name the denominator when a rate excludes unknown outcomes.
- Avoid causal language without supporting evidence.
Common mistake
The trap. A rate difference is real, but a title that says what caused it can be made up.
Precision and units
- Exact lookup: keep exact euro amounts.
- Executive overview: use clearly labelled €m figures.
- Rates: use one decimal point and a stated group.
Common mistake
The trap. More decimals can make an honest chart look precise beyond its purpose.
Data Visualisation / Module 3 / Recap
Recap card
3.1 Hiding the range hides the risk
Hiding the range hides the risk
- Name your uncertainty. Recorded historical variation is not future prediction.
- Give low and high. Tell readers the period and source.
- Show one figure with context rather than a false guarantee.
Common mistake
The trap. A tidy point estimate can look certain when the recorded months varied widely.
Data Visualisation / Module 3 / Recap
Recap card
3.2 Small groups set apart
Small groups set apart
- Count distinct customers rather than rows.
- Mark small groups using a stated practice rule.
- Keep flagged groups visible without treating their rank as settled.
Common mistake
The trap. Many invoices can still come from a small set of customers.
Data Visualisation / Module 3 / Recap
Recap card
3.3 Ranges on bars
Ranges on bars
- Show the group size beside the rate.
- Use an explicit denominator: only confirmed on-time and late lines here.
- Label the observed range without claiming a probability interval.
Common mistake
The trap. A percentage alone can hide very different group sizes and unknown outcomes.
Data Visualisation / Module 3 / Recap
Recap card
3.4 One number, with its range
One number, with its range
- Central figure: state what it summarises.
- Range: keep period, measure and meaning visible.
- Use consistent rounding across the whole card.
Common mistake
The trap. A precise-looking single figure can imply certainty that the data do not justify.
Data Visualisation / Module 3 / Recap
Recap card
Module 3: Show what you are not sure of
Hiding the range hides the risk
- Name your uncertainty. Recorded historical variation is not future prediction.
- Give low and high. Tell readers the period and source.
- Show one figure with context rather than a false guarantee.
Common mistake
The trap. A tidy point estimate can look certain when the recorded months varied widely.
Small groups set apart
- Count distinct customers rather than rows.
- Mark small groups using a stated practice rule.
- Keep flagged groups visible without treating their rank as settled.
Common mistake
The trap. Many invoices can still come from a small set of customers.
Ranges on bars
- Show the group size beside the rate.
- Use an explicit denominator: only confirmed on-time and late lines here.
- Label the observed range without claiming a probability interval.
Common mistake
The trap. A percentage alone can hide very different group sizes and unknown outcomes.
One number, with its range
- Central figure: state what it summarises.
- Range: keep period, measure and meaning visible.
- Use consistent rounding across the whole card.
Common mistake
The trap. A precise-looking single figure can imply certainty that the data do not justify.
Data Visualisation / Module 4 / Recap
Recap card
4.1 Correct numbers, wrong picture
Correct numbers, wrong picture
- Running total: it only rises, so it hides months that fell
- Monthly bars from zero show June's 27.2% fall
Common mistake
Trusting correct labels. Every label was right, and readers still saw growth every month.
Data Visualisation / Module 4 / Recap
Recap card
4.2 Measure the distortion
Measure the distortion
- Visual-to-data ratio: drawn ratio divided by data ratio
- Bubbles and dual axes: area squares; hand-set scales can reverse a comparison
Common mistake
Reading a crossing point. Where two hand-set axes cross means nothing.
Data Visualisation / Module 4 / Recap
Recap card
4.3 Does it change the conclusion?
Does it change the conclusion?
- Material: a reader would conclude, rank or size things differently
- Mix: split an average by kind before saying why it moved
Common mistake
Calling a mix shift a price rise. +47.9% overall, 7.2% to 11.3% within each line kind.
Data Visualisation / Module 4 / Recap
Recap card
4.4 Honest words, kept caveats
Honest words, kept caveats
- Honest titles: 18.5% to 14.7% is about a fifth
- Cleaner keeps caveats: September and October are not final
Common mistake
A title with a cause. Averages differ by order size; the median is 3 days for all.
Data Visualisation / Module 4 / Recap
Recap card
Module 4: Diagnose misleading visuals
Correct numbers, wrong picture
- Running total: it only rises, so it hides months that fell
- Monthly bars from zero show June's 27.2% fall
Common mistake
Trusting correct labels. Every label was right, and readers still saw growth every month.
Measure the distortion
- Visual-to-data ratio: drawn ratio divided by data ratio
- Bubbles and dual axes: area squares; hand-set scales can reverse a comparison
Common mistake
Reading a crossing point. Where two hand-set axes cross means nothing.
Does it change the conclusion?
- Material: a reader would conclude, rank or size things differently
- Mix: split an average by kind before saying why it moved
Common mistake
Calling a mix shift a price rise. +47.9% overall, 7.2% to 11.3% within each line kind.
Honest words, kept caveats
- Honest titles: 18.5% to 14.7% is about a fifth
- Cleaner keeps caveats: September and October are not final
Common mistake
A title with a cause. Averages differ by order size; the median is 3 days for all.
Data Visualisation / Module 5 / Recap
Recap card
5.1 Colour alone cannot carry status
Colour alone cannot carry status
- Second cue: a text label on every status
- Simulated view: green and red 1.09:1 apart, a teaching approximation
Common mistake
Colour-only status. LATE, the largest group at 9,563 lines, can look like ON_TIME.
Data Visualisation / Module 5 / Recap
Recap card
5.2 Contrast you can measure
Contrast you can measure
- Checklist: text 4.5:1; large text, marks and icons 3:1
- Marks and text differ: green passes as a mark, fails as text
Common mistake
Pale labels. #AAAAAA at 2.32:1 fails; #767676 at 4.54:1 passes.
Data Visualisation / Module 5 / Recap
Recap card
5.3 Keyboard, reading order and hover
Keyboard, reading order and hover
- Focus order: filters, KPI row, chart, table link, icons
- Names and values: every control named, no value on hover only
Common mistake
A hover-only value. Q2's 395 NOT_SHIPPED lines were invisible without a mouse.
Data Visualisation / Module 5 / Recap
Recap card
5.4 Text alternatives, a table route and the release deadline
Text alternatives, a table route and the release deadline
- Text alternative: state the finding with the numbers
- Must ship: every fix without which some readers cannot use the page
Common mistake
Passing one check. Colour was fixed; keyboard and hover faults were still open.
Data Visualisation / Module 5 / Recap
Recap card
Module 5: Charts everyone can read
Colour alone cannot carry status
- Second cue: a text label on every status
- Simulated view: green and red 1.09:1 apart, a teaching approximation
Common mistake
Colour-only status. LATE, the largest group at 9,563 lines, can look like ON_TIME.
Contrast you can measure
- Checklist: text 4.5:1; large text, marks and icons 3:1
- Marks and text differ: green passes as a mark, fails as text
Common mistake
Pale labels. #AAAAAA at 2.32:1 fails; #767676 at 4.54:1 passes.
Keyboard, reading order and hover
- Focus order: filters, KPI row, chart, table link, icons
- Names and values: every control named, no value on hover only
Common mistake
A hover-only value. Q2's 395 NOT_SHIPPED lines were invisible without a mouse.
Text alternatives, a table route and the release deadline
- Text alternative: state the finding with the numbers
- Must ship: every fix without which some readers cannot use the page
Common mistake
Passing one check. Colour was fixed; keyboard and hover faults were still open.