Understanding KPI Field Type: A Practical Guide for Analysts

KPI Field Type Examples and When to Use Each

KPIs (Key Performance Indicators) are only useful when their data is modeled and displayed correctly. Choosing the right KPI field type ensures accurate aggregation, correct visualizations, and clear interpretation. Below are common KPI field types, concrete examples, and guidance on when to use each.

1. Numeric (Measure)

• What it is: A continuous numerical value that supports arithmetic (sum, average, min, max).
• Examples: Sales amount, revenue, cost, profit, units sold, time in minutes.
• When to use: Any KPI that requires mathematical aggregation, trend analysis, or rate calculations (e.g., monthly revenue growth, average order value).
• Visualization fit: Line charts, bar charts, area charts, KPI cards with value and delta.

2. Integer / Count

• What it is: Whole numbers used for counting distinct events or entities.
• Examples: Number of customers, number of orders, number of support tickets, active users.
• When to use: Use when aggregation should be counted rather than summed as fractional values; use distinct counts for unique entities (unique customers) and simple counts for events.
• Visualization fit: Bar charts, stacked bars, tables, simple numeric tiles.

3. Percentage / Ratio

• What it is: A normalized number expressed as a percentage or ratio comparing two measures.
• Examples: Conversion rate (orders/visitors), margin percentage (profit/revenue), uptime (%).
• When to use: Use for KPIs that show proportions, efficiency, or relative performance. Prefer ratios when you need context (e.g., click-through rate rather than raw clicks).
• Visualization fit: Gauge charts, bullet charts, KPI indicators with target lines.

4. Boolean / Binary

• What it is: True/false or yes/no values.
• Examples: SLA met (yes/no), payment received (true/false), feature enabled.
• When to use: Use for pass/fail style KPIs or to filter datasets. Not for aggregation other than counts or shares (percent true).
• Visualization fit: Simple status indicators (green/red), pie chart of true vs false, or conditional icons.

5. Categorical / Dimension

• What it is: Text labels used to group or slice data.
• Examples: Region (North America, EMEA), Product category, Customer segment, Channel (email, organic, paid).
• When to use: Use when breaking down measures by group to compare performance across categories. Often paired with numeric measures.
• Visualization fit: Stacked bars, grouped bars, treemaps, heatmaps, pivot tables.

6. Date / Time

• What it is: Date or timestamp fields used for temporal analysis.
• Examples: Order date, signup date, last login timestamp, incident resolved time.
• When to use: Any time-series KPI or rolling-window calculation (week-over-week growth, 30-day retention, month-to-date revenue).
• Visualization fit: Line charts, area charts, time-series heatmaps.

7. Currency

• What it is: Numeric values with currency semantics and formatting.
• Examples: USD revenue, EUR expenses, budget allocations.
• When to use: Financial KPIs where currency formatting and possibly currency conversion is required.
• Visualization fit: Same as numeric but with currency labels; financial tables and waterfall charts.

8. Duration / Interval

• What it is: Time intervals measured in seconds/minutes/hours.
• Examples: Average response time, time to resolution, average session duration.
• When to use: Use for performance or experience KPIs where time-to-complete matters; choose median over mean for skewed distributions.
• Visualization fit: Box plots, histograms, bar/line charts with time units.

9. Rank / Ordinal

• What it is: Ordered categorical values indicating position or priority.
• Examples: Product rank by sales, priority level (1–5), quartile.
• When to use: When relative ordering matters but numeric distances do not (e.g., rank 1 vs 2).
• Visualization fit: Leaderboards, sorted tables, bar charts showing rank positions.

10. Calculated / Derived Field

• What it is: A field computed from other fields (formulas, ratios, window functions).
• Examples: Churn rate = churned customers / starting customers; LTV = average revenue per user × average lifespan.
• When to use: When raw data doesn’t provide the KPI directly and you need business logic, normalization, or rolling calculations.
• Visualization fit: Any, depending on the underlying measure — often used in dashboards to surface higher-level metrics.

Practical guidance for choosing field types

• Prefer numeric/measures for any metric you’ll aggregate; choose integer for counts, decimal for monetary/precision values.
• Use date/time fields for any temporal analysis; ensure consistent time zones.
• Use categorical fields strictly for grouping/slicing; avoid encoding numeric information as text.
• Create calculated fields for ratios, normalized metrics, or business rules; keep raw measures for traceability.
• For skewed distributions (response times, revenue by customer), show medians or percentiles in addition to averages.

Quick mapping (common KPIs → field type)

• Monthly revenue → Currency (numeric measure)
• Active users daily → Integer (count) with Date dimension
• Conversion rate → Percentage (calculated ratio)
• SLA compliance → Boolean or Percentage (share)
• Average handle time → Duration (median preferred)
• Revenue by product category → Numeric measure + Categorical dimension

Use the correct field type consistently in your BI tool and document definitions so users and downstream reports interpret KPIs correctly.