7.3 Chart Visualization Reports
Key Takeaways
- Chart visualization reports use types such as column, bar, line, area, and pie to communicate trends and comparisons from statistical results.
- The Visualization Picker lets Users select a chart type after a search returns chart-ready series data.
- Chart-ready data usually comes from transforming commands that produce numeric series over categories or time—not from an untransformed Events list alone.
- chart and timechart are common producers of chart-friendly tables; User exam depth stops at recognizing that role, not Power User eval gymnastics.
- Match chart type to the question: line/area for trends over time, column/bar for category comparisons, pie for simple part-to-whole with few slices.
7.3 Chart Visualization Reports
Objective 6.4 covers reports that display visualizations (charts). Where 6.3 emphasized tables, this objective asks you to turn statistical results into column, line, pie, and related chart types using the Search Visualization tools—and to save those choices as reports.
From Statistics to a Chart Report
Charts need series-shaped data: numeric values plotted against categories or time. In Splunk Search:
- Run a search that ends with a transforming command.
- Confirm Statistics shows a chartable table (categories + numbers, or
_time+ numbers). - Open the Visualization tab.
- Use the Visualization Picker to choose Column, Line, Area, Bar, Pie, and similar types.
- Adjust basic format options (legend, axis titles) if needed.
- Save As Report so the chart type travels with the saved search.
| Visualization type | Best for | Weak fit |
|---|---|---|
| Column / Bar | Comparing counts across categories (hosts, status codes) | Dozens of nearly equal categories (hard to read) |
| Line / Area | Trends over time | Single static category with no time axis |
| Pie | Part-to-whole with few slices | Many categories or tiny percentages |
| Table | Exact multi-column values (objective 6.3) | At-a-glance trend storytelling |
Chart-Ready Data: What the Exam Expects
An Events-only result set (no transforming command) is usually not chart-ready. You need aggregation or time-series shaping first.
| Data shape | Example SPL (illustrative) | Likely chart |
|---|---|---|
| Category + count | ... | stats count by host | Column/Bar |
| Category + count (ranked) | ... | top status | Column or Pie (if few values) |
| Time + metric | ... | timechart span=1h count | Line/Area/Column over time |
| Category matrix | ... | chart count over host by method | Clustered/stacked column |
chart and timechart as series producers (User-level mention)
You may see chart and timechart called out as commands that return tables designed for charting:
timechart— buckets_timeand computes stats per time bucket (great for line/area trend reports).chart— builds a chart-oriented table over a field (often with a split-by series).
For the User exam, know that these commands (along with stats/top) create chart-friendly statistical tables. You do not need deep Power User eval chains, complex overlay formatting, or custom visualization apps to answer 6.4 questions.
Worked example: column chart report
index=web sourcetype=access_combined status=5*
| stats count by host
| sort - count
Visualization → Column Chart → Save As Report 5xx Errors by Host. Readers instantly see which host dominates failures.
Worked example: line chart over time
index=web sourcetype=access_combined status=500
| timechart span=1h count as errors
Visualization → Line Chart → Save As Report HTTP 500 Errors Over Time. This is the classic “is it getting worse?” report.
Worked example: pie chart (use carefully)
index=web sourcetype=access_combined
| top limit=5 method
With only a few methods, Pie can show part-to-whole share. If top returns many thin slices, switch to Column or stay on Table.
Using the Visualization Picker (User Skill)
The Visualization Picker is the UI control that lists available visualization types for the current statistical results. User exam behaviors:
- Pick a type that matches the data shape (time → line; categories → column; few parts → pie).
- If a chart looks empty or wrong, return to Statistics and verify the transforming command produced numeric series.
- Save the report after the picker selection so the report retains that visualization.
Formatting options (stack mode, null handling, axis scale) may appear in the Format menu. Recognize that formatting exists; do not memorize every Format panel for User.
Matching Questions to Chart Types
| Stakeholder question | Chart-oriented answer |
|---|---|
| “Which host has the most 500s right now?” | Column/Bar of stats count by host |
| “Are failures rising this afternoon?” | Line/Area from timechart count |
| “What share of requests are GET vs POST?” | Pie or column from top method / stats count by method |
| “Show me exact counts and percents for ten users” | Prefer Table (6.3), not a crowded pie |
This matching skill is frequently how scenario items disguise objective 6.4.
Saving Chart Reports vs Dashboard Panels
Saving as a report stores the search + chart choice. You can later:
- Open the report from Reports and see the same chart
- Add the report to a dashboard as a panel (7.4)
- Edit the report later to change chart type (7.1)
Do not confuse Save As Report with Save As Dashboard Panel—both can start from Visualization, but reports are the reusable named objects Domain 6 emphasizes before dashboard assembly.
Exam Traps for 6.4
| Trap | Reality |
|---|---|
| Charting raw Events with no transform | Transform first to get series data |
| Using pie for 40 categories | Column/table is clearer |
Assuming only timechart can make charts | stats/top/chart results chart too |
| Studying Studio drilldown tokens for chart reports | User skill is picker + save report |
Writing complex eval to “unlock” visualizations | Not required for Core User 6.4 |
Why Charts Matter on This Exam
Domain 6.0 is not only about storing SPL—it is about communicating results. Chart reports prove you can choose a visualization that matches the analytical question and preserve it for others. Pair that with table reports (6.3) and you can pick the right display mode under exam pressure.
What must a search usually produce before the Visualization Picker can show a useful column or line chart?
Which visualization type is usually best for showing how error counts change hour by hour?
At User level, what is the role of the chart and timechart commands when creating visualization reports?