Launch Deal · 50% off 1st month ·

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.

See plans

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.