3.8 — Identify Patterns and Trends: Analyze, Grouping, AI Visuals, Forecasting and Anomalies
PL-300 Domain 3 · Skill 3.8
What You Need to Know
The Analyze feature
Right-click a data point in a bar, column or line chart and choose Analyze:
- Explain the increase / Explain the decrease: compares the point with the previous one and shows which categories contributed most to the change, as visuals you can add to the report.
- Find where this distribution is different: looks for fields where the selected point's distribution differs most from the others.
It works with simple, additive measures and doesn't support every measure type (for example, some non-additive measures or DirectQuery sources).
Grouping, binning and clustering
| Feature | Works on | What it does |
|---|---|---|
| Groups | Categorical values (Ctrl-select points in a visual, or New group on a field) | Combine items into a group, with an optional Other group |
| Bins | Numeric or date columns | Split into equal ranges, by bin size (every 10 years of age) or number of bins |
| Clusters | Scatter charts (or tables) | Automatically find clusters of similar items; creates a new field you can use elsewhere |
Groups and bins create new fields in the model (shown with a group icon). Clustering is a quick segmentation, not a statistical model you can tune.
AI visuals
| Visual | Answers |
|---|---|
| Key influencers | What drives a metric up or down, or a category (churned = Yes), and which segments stand out |
| Decomposition tree | Breaks a measure down across dimensions in any order; AI splits pick the next highest or lowest value automatically |
| Q&A | Answers natural-language questions with a visual (uses names and synonyms) |
| Narrative | Text summary (lesson 3.3) |
The Analytics pane
Select a visual and open the Analytics pane (magnifying glass):
| Line | Available on |
|---|---|
| Constant line, min, max, average, median, percentile lines | Many cartesian charts |
| Trend line | Line, scatter and some column charts with a continuous axis |
| Forecast | Line charts with a continuous date axis and one measure (no legend). You set forecast length, confidence interval, ignore last N points, and seasonality |
| Error bars | Line, bar and column charts; upper and lower bounds from fields or percentages |
| Symmetry shading, ratio line | Scatter charts |
Detect outliers and anomalies
- Find anomalies (Analytics pane on a line chart with a time axis) marks points outside the expected range, with a sensitivity slider. Selecting an anomaly opens an explanation of possible causes.
- For outliers between two measures, use a scatter chart with clustering, reference lines, or conditional formatting on a calculated threshold (for example, values more than 2 standard deviations from the mean).
Key Concept: Analyze explains a change, groups and bins make new categories, clusters segment points, AI visuals find drivers and breakdowns, and the Analytics pane adds reference lines, error bars, forecasts and anomaly detection.
Exam Traps
- Trying to forecast on a chart with a categorical axis or a legend
- Using a decomposition tree when the question is "what influences churn" (key influencers)
- Creating bins on a text column (bins need numbers or dates; use groups)
- Expecting Analyze > Explain the increase to work on every measure and source
- Detecting anomalies on a bar chart (they need a line chart over time)
Check your understanding
Practice Scenario
🔒 This practice scenario is part of Skillmarked Pro
The lessons stay free. Pro unlocks the other 22 PL-300 practice scenarios and the 50-question timed mock exam, plus everything for every other exam.
Launch Deal: first month €9.95, then €19.90/month. Enter code LAUNCH50 at checkout. Ends 1 December.
Cancel anytime. Or €199 a year, which saves 17%.
Not sure yet? Try 10 free mock exam questions.
Already Pro? Log in.
Sources
- Apply insights in Power BI to explain fluctuations in visuals, Microsoft Learn
- Use grouping and binning in Power BI Desktop, Microsoft Learn
- Key influencers visualizations, Microsoft Learn
- Use the Analytics pane in Power BI Desktop, Microsoft Learn
- Anomaly detection, Microsoft Learn