Categories: Power BI
Tags:

HR — Employees Performance Analysis Dashboard (Power BI Tutorial)

By Ankit Srivastava — Head of IT Training, Slidescope Institute

In this tutorial I’ll walk you step-by-step to recreate the HR — Employees Performance Analysis dashboard (the image you shared). We’ll start from a CSV, prepare the data, create a few calculated columns and measures (including AVERAGE(hrdata[Performance Rating]) and the Satisfaction Category using SWITCH), then design the report page with KPIs, charts, and slicers — just like a polished Power BI report.

1) Files & tools you need

  • Power BI Desktop (latest)
  • Your CSV file (example first row included in your message). Name it hrdata.csv.
  • Optional: a company color theme JSON or font preferences.
  • Here is the dataset : Human-resources-hr-employee-performance-dataset

CSV columns (from your example)

Employee ID, Employee Name, Department, Job Role, Joining Date, Experience (Years),
Age, Gender, Work Location, Working Hours (Per Week), Performance Rating,
Salary ($), Bonus ($), Training Hours (Per Year), Job Satisfaction Score,
Leave Days Taken, Remote Work (%), Promotion Status

2) Import & initial data prep (Power Query)

  1. Get Data → Text/CSV → hrdata.csv. Click Transform Data.
  2. In Power Query:
    • Ensure column data types: Employee ID → Text, Joining Date → Date/Time, numeric columns (Experience, Age, Working Hours, Performance Rating, Salary ($), Bonus ($), Training Hours, Job Satisfaction Score, Leave Days Taken, Remote Work (%)) → Decimal/Whole Number as appropriate.
    • Trim whitespace on text columns (Employee Name, Department, Job Role, Work Location, Gender, Promotion Status).
    • Check for nulls and duplicates: decide how to handle duplicates (keep last record / flag).
    • If Joining Date is 2018-11-21 00.00.00 style, change locale or use Split Column -> By Delimiter to isolate date and convert to proper Date type.
  3. Create simple transformations here if required (e.g., fill down/up, remove rows with missing Employee ID).

When done, click Close & Apply.

3) Create helper columns (Power Query or DAX)

You can create bins and categories either in Power Query or as DAX calculated columns. I usually do stable categorical transforms in Power Query and dynamic measures in DAX. Example helpful columns:

  • Age Bin (Power Query or DAX): 0-25, 26-35, 36-45, 46-55, 56+
  • Remote Work Bin: create intervals (0-25, 26-50, etc.)
  • Satisfaction Category: user requested SWITCH logic — we’ll implement in DAX (next step).

4) Calculated column: Satisfaction Category (DAX)

You said: Low less than 3; Medium 4-7; above 7 high. There is a small gap for exact value 3 and values between 3 and 4 — I’ll make a clear rule:

  • Low = Job Satisfaction Score < 3
  • Medium = Job Satisfaction Score >= 4 and <= 7
  • High = Job Satisfaction Score > 7
  • For scores >=3 and <4 (e.g., 3.0–3.99) we’ll classify as Low for conservative grouping (you can change easily).

Create this calculated column in Data view → New column:

Satisfaction Category = 
SWITCH(
    TRUE(),
    ISBLANK( hrdata[Job Satisfaction Score] ), "Unknown",
    hrdata[Job Satisfaction Score] < 3, "Low",
    hrdata[Job Satisfaction Score] >= 4 && hrdata[Job Satisfaction Score] <= 7, "Medium",
    hrdata[Job Satisfaction Score] > 7, "High",
    "Low"  -- fallback for 3 to <4, adjust if you prefer Medium
)

Explanation: SWITCH(TRUE(), ...) is a flexible pattern to evaluate ranges. Adjust the fallback to suit your policy.

5) Core DAX measures (KPIs)

Create measures in Modeling → New measure. Important base measures:

-- Count of Employees
CountEmployees = DISTINCTCOUNT( hrdata[Employee ID] )

-- Average Experience
AvgExperience = AVERAGE( hrdata[Experience (Years)] )

-- AVERAGE Performance Rating (as you requested)
AvgPerformanceRating = AVERAGE( hrdata[Performance Rating] )

-- Average Salary
AvgSalary = AVERAGE( hrdata[Salary ($)] )

-- Total Training Hours
TotalTrainingHours = SUM( hrdata[Training Hours (Per Year)] )

Additional measures you’ll find useful:

-- Avg Recovery by Department (example)
AvgPerf_ByDept = CALCULATE( [AvgPerformanceRating], ALLEXCEPT(hrdata, hrdata[Department]) )

Format these measures (Modeling pane) — set decimal places, currency for salary, display units (K) for training hours if large.

6) Page layout & KPIs (visual design)

  • Set page size (16:9 or custom).
  • Add a top title textbox: HR – Employees Performance Analysis Dashboard.
  • Top row: create 5 KPI cards using Card or Multi-row card visuals and place them horizontally:
    • CountEmployees
    • AvgExperience
    • AvgPerformanceRating
    • AvgSalary (format currency)
    • TotalTrainingHours (display as 15K or raw)
  • Style: remove card backgrounds, set border/shadow, use company colors.

7) Visuals — replicate the dashboard

Below are the visuals you see in the image and what to drag/drop:

  1. Avg Performance Rating by Department
    • Visual: Clustered Bar Chart
    • Axis: Department | Value: AvgPerformanceRating
    • Sort by value descending; show data labels.
  2. Top 10 Roles by Performance Rating
    • Visual: Bar Chart (horizontal) with Job Role on Y and AvgPerformanceRating on X; apply Top N filter (Top 10) on Job Role by measure AvgPerformanceRating.
  3. Performance Rating by Gender
    • Visual: Donut Chart
    • Legend: Gender | Values: AvgPerformanceRating (or count/average depending display preference).
  4. Performance Rating by Age Group
    • Visual: Combo (Clustered Column for avg rating by age bin; line for Salary sum if desired)
    • Axis: Age Bin | Values: AvgPerformanceRating & SUM(Salary ($)) (secondary axis for salary).
  5. Performance Rating by Work Location (Map)
    • Visual: Map / Filled Map or Bubble Map
    • Location: Work Location | Size: AvgPerformanceRating or CountEmployees as bubble size. Use categorical tooltips.
  6. Performance Rating by Satisfaction Category
    • Visual: Donut / Pie showing average rating per Satisfaction Category. Place percentages and labels.
  7. Performance Rating by Remote Work % (bins)
    • Visual: Clustered Column — create RemoteWorkBin column using Grouping or DAX BIN and plot AvgPerformanceRating.
  8. Performance Rating by Gender and Department
    • Visual: Stacked Area / Stream Graph / Stacked Bar with Department on X, AvgPerformanceRating as Y, Gender as Legend — to create the banded look.
  9. Promotion Status Table with Avg Performance
    • Visual: Table or Matrix with rows Promotion Status and column showing AvgPerformanceRating. Add KPI markers via conditional formatting (e.g., color icons).

8) Filters and Interactivity

On the right pane add slicers (drop-down style) and sync them if you have multiple pages:

  • Department
  • Job Role
  • Gender
  • Work Location
  • Promotion Status

Enable Edit interactions so slicers filter the visuals as required. Turn on cross-report tooltips or custom tooltip pages for richer hover content.

9) Formatting & polish

  • Use a consistent theme (corporate colors). Import a JSON theme if you have one.
  • Add shadows to card visuals (Visual effects), rounded shapes with images, and ensure text sizes are legible.
  • For data labels: show value and maybe 1 decimal for ratings.
  • Use Bookmarks and Buttons for storytelling (e.g., “View Top Performers”, “View Low Satisfaction”).

10) Validate & Publish

  • Test slicer combinations and cross-filtering.
  • Check performance: if dataset is large, consider aggregating in Power Query or a data model.
  • Publish to Power BI Service and configure row-level security if necessary. Share with stakeholders via apps or embed in internal portals.

Closing notes (practical tips from my training experience)

  • Keep your measures simple and well-named — this helps non-technical stakeholders understand the model.
  • Document how Satisfaction Category is created (note the convention for edge values like 3.0). If stakeholders prefer different boundaries, change the DAX accordingly.
  • Use tooltips to show supporting metrics (count of employees, median rating) so the chart tells a story without clutter.
  • For live HR systems, schedule dataset refreshes and keep performance in mind — incremental refresh helps for large tables.