Choose charts that reveal rather than manufacture patterns
Learn the reasoning. Try the task. Compare your answer.
Prepared for DataCareerHub · Curriculum version 2026-10-15.2
By the end of this lesson
- Match a chart to a distribution, comparison or composition question.
- Use explicit non-overlapping bins and consistent axes.
- Reconcile visual summaries to counts and threshold boundaries.
Harbor Helpdesk · UTC · Snapshot 22 June 2026 00:00 · Prepared fictional extract · No Python or SQL required.
Understand the method
A chart is an argument about what the reader should notice. Make that argument inspectable: identify the population, show units, keep comparable scales and provide the counts behind the picture. For this small extract, a well-labelled table or a dot plot can be more informative than a visually impressive dashboard. Choose the view after deciding which uncertainty it resolves.
Three questions, three views
For the spread of response durations, use individual dots by cohort or a histogram with shared bins. For the share meeting the four-hour target, use a cohort comparison with counts and denominators. For channel composition, show email/chat counts or shares by cohort. These are different views of the same tickets; do not stack a duration mean and a percentage in one bar as if their units were additive.
A useful chart title states the measured pattern, for example “Recorded four-hour response timeliness: 8/12 and 10/12 tickets.” Its subtitle gives creation windows and the snapshot. A stronger explanatory note can say that composition changes and causal attribution is unavailable. Avoid “New routing wins” because it makes an unsupported explanation the headline. The accessible text or adjacent table should convey the important counts without requiring colour recognition.
Make bins explicit
Use these common half-open bins in hours: [0,1), [1,2), [2,4), [4,8), and [8,infinity). A value on the lower boundary belongs in that bin; a value on the upper boundary belongs in the next one. Week 1 counts are 2, 2, 3, 4 and 1. Week 2 counts are 3, 2, 3, 3 and 1. Each row sums to 12. Keep the same bin definitions for both cohorts and show that wider bins cover wider duration intervals. Equal-width drawn bars with these unequal intervals are a frequency-category display, not a density histogram unless widths/heights are handled appropriately.
These bins are useful summaries but do not reproduce the inclusive four-hour service measure by simply adding the first three counts. Values exactly four hours fall in [4,8), yet the exercise policy counts them as timely. Week 1 has one exactly-four-hour ticket and Week 2 has two. Therefore timely counts are 7+1=8 and 8+2=10. A display boundary must not silently rewrite a business definition.
Keep comparisons honest
A count-bar chart should start at zero; a truncated baseline can exaggerate a small count difference. For rates, use a common, clearly labelled percentage scale and display n. Keep colour meanings consistent, distinguish cohorts with labels as well as colour, and avoid a smooth trend line across a tiny number of unevenly populated dates. A line connecting daily points suggests continuity that needs a clear time-series definition.
Your deliverable is a chart plan, not a requirement to use a particular software product. Specify audience, question, marks/axes, units, population, annotation and the fallback table. You may build a local chart in a spreadsheet after checking its data ranges. The supplied worked exploration provides static tables and frequency bars; it does not refresh from a neighbouring CSV or run a production dashboard. Verify each visual total against the source before writing the explanation.
Practical exercise
Complete chart_plan_template.txt for a duration distribution, a timeliness comparison and a channel-mix view. Use the shared bins to tally each cohort and reconcile to twelve.
- Explain where exactly 4h and 8h fall.
- Recover the inclusive timeliness counts without changing the bins.
- Replace “New routing wins” with an evidence-based title.
- State how unequal bin widths and colour-only distinctions can mislead.
Sketch or build one chart locally and include the underlying count table.
Open the worked solution
Bin counts are W1 [2,2,3,4,1], W2 [3,2,3,3,1]. Exactly 4h belongs in [4,8), exactly 8h in [8,infinity). The first three bins total 7 and 8, not the timely 8 and 10: add the exactly-4h counts of 1 and 2 for the inclusive service measure.
Suggested views: cohort dot plots on the same elapsed-hour scale with a labelled 4h reference; a zero-based count/rate comparison with n=12 per cohort; a channel-composition count table or 100% bar with both labels and fractions. Title: “Recorded timeliness rises while channel composition changes.” Describe dates, snapshot and the absence of causal attribution. If rendering bins as equal-width categories, label them as frequency categories; do not imply equal numeric intervals. Provide a table and text summary so meaning survives monochrome display and screen-reader use.
For a rate chart, retain the denominators rather than substituting bins: the visual must still communicate 8/12 and 10/12. With an open-ended final category there is no finite bin width to use for an ordinary density height; a labelled count-category display avoids that unsupported conversion.
One more question.
How these learning materials are prepared and maintained · Report a correction: support@datacareerhub.io
Read and practice without an account. Sign in with an existing account to save lesson completion.
Next lesson →Harbor Helpdesk is a fictional teaching case. Descriptive comparisons do not establish routing causality. Portfolio practice is not a claim of client work or an accredited qualification. Explore practical lessons, worked examples and projects. Follow a published path, practise the method and build work you can explain.