Data presentation that readers can check and explore

Suppose you have a spreadsheet of service requests by district. Readers need to find their district, compare response performance, and check the numbers behind your summary. A screenshot of the workbook is not enough: they need readable values, clear definitions, and a table they can explore.

Data presentation is the selection, arrangement, and explanation of data for a reader’s task. For a public dataset, the useful result is a page with a clear question, an appropriate table or chart, and enough context to interpret the figures without opening the original workbook.

1. Start with the question readers need answered

Write the question before choosing a display. “Which district received the most requests?” and “Which district resolved the highest proportion within seven days?” require different measures.

Here is a fictional teaching dataset, not observed service performance. Each row represents one district’s requests received during September 2026; every request has had its full seven-day follow-up period.

District	Requests received	Resolved within 7 days	Resolved within 7 days (%)
North	120	96	80.0
South	60	54	90.0
West	200	170	85.0

The percentage is resolved within 7 days ÷ requests received × 100.

West received the most requests, while South had the highest within-seven-day resolution rate. Neither statement establishes why the districts differ. Keep the counts beside the rates so readers can see the denominator—the number of requests each percentage is based on.

2. Choose a table, chart, or both

Use the reader’s task to choose the format:

Reader task	Useful presentation
Find a district’s exact count or rate	Table
Compare district resolution rates	Bar chart, with counts available in a table
Follow the same measure across months	Line chart, with dates in chronological order
Examine the relationship between two numeric measures	Scatter plot

The UK Government Analysis Function’s guidance on demonstration tables recommends tables for comparing values and charts for patterns, trends, and relationships. Its chart guidance connects chart choice to the statistical relationship being shown.

For this example, make the table the primary view. Add a bar chart only if comparing rates is an important task. Start its numerical axis at zero: truncating a bar chart’s axis distorts the relative bar lengths, as the chart guidance explains. Three districts in one month do not support a time-trend chart.

3. Make labels and ordering carry meaning

Replace headings such as Total, Done, and % with labels that name the measure and its unit. Use consistent displayed precision: here, percentages have one decimal place.

Choose an intentional order. Alphabetical order helps readers locate a district; descending rate order emphasizes ranking. Keep dates and ordered categories in their natural sequence. These choices follow the Analysis Function’s guidance on precision, labels, and category ordering.

Do not mix missing values with confirmed zeros. If a district did not report, leave the numeric value missing under a documented convention rather than assigning zero. See empty fields, zeros, and missing values.

If you publish an overall rate, calculate it from the underlying counts. In this example, 320 ÷ 380 × 100 = 84.2%; the simple average of the three district rates is 85.0% and gives each district equal weight despite their different request volumes.

4. Put context next to the data

Prepare a short description to accompany the public table:

Service requests resolved within seven days, by district, September 2026 Each row covers requests received in one district during the month. The rate is the percentage resolved within seven days of receipt, with complete follow-up for every request. Figures are fictional and supplied for demonstration only.

For a real dataset, add the original source link, coverage, update date, missing-value rules, and any exclusions or definition changes. Link to the actual source dataset or release, not just the organization’s homepage.

A caption identifies a table; a summary explains its organization when that needs clarification. W3C’s table guidance makes this distinction and advises against duplicating the caption in the summary. For wording examples, see accessible table captions and summaries.

If you include a chart, retain the underlying table and describe the important comparison in text. Do not make color the only way to identify a district. The Analysis Function’s chart accessibility guidance explains how tables, text descriptions, and labels help convey chart content without relying on sight or color alone.

5. Publish the reviewed file and check the public result

Create a release copy with one header row, consistent columns, and no merged headings or decorative blank rows. Include only data intended for public release; check all sheets and columns for sensitive information before upload. A public dataset page is not a private workspace.

As of October 2026, TablePage accepts CSV, TSV, XLSX, and XLS files and generates a public, shareable dataset page with a filterable table and charts. Upload the reviewed file, then inspect the resulting page rather than assuming the import or generated visuals are correct.

Before sharing the link, check:

• Values: Row count, counts, percentages, and missing values match the release file.
• Interaction: Find one district and test the available filters; confirm numeric comparisons behave as expected.
• Context: Definitions, period, source, and limitations remain available wherever you share or embed the data. Supply accompanying text if the shared view omits them.
• Readability: Check a narrow screen and keyboard navigation; verify table headers and chart text alternatives.

The page is ready when a reader can find a value, understand what it measures, and trace it back to its source—not merely when the upload succeeds.