This Power BI dashboard provides a comprehensive view of Company A's digital marketing performance.
This Power BI dashboard provides a comprehensive view of Company A's digital marketing performance across 4 channels (Email, SMS, Social Media, Paid Search) over 104 weeks (Jan 2023 – Dec 2024).
The dashboard translates complex Marketing Mix Model (MMM) outputs and incrementality test results into clear, actionable insights for leadership and marketing teams.
Note: Company A is an anonymized client. All data is simulated to reflect realistic marketing dynamics.
High-level marketing performance summary including:
- Total Revenue, Total Spend, Conversions, Avg ROAS
- Weekly Revenue vs Spend trend
- Monthly Revenue vs Spend comparison
- Year and Quarter filters
Key Insight: Company A generated $11.53M in revenue from $4.07M in media spend — delivering an overall ROAS of 2.83x over 104 weeks.
Deep dive into channel-level performance including:
- Average ROAS by channel
- Total Spend vs Revenue Contribution
- Weekly Spend trends by channel
- Adstock decay and carryover profile table
Key Insight: SMS delivers the highest ROAS (1.12) despite the lowest spend — an underleveraged channel. Social Media shows signs of oversaturation with the lowest ROAS (0.92).
MMM-based budget reallocation recommendations including:
- Current vs Optimized weekly spend by channel
- Budget reallocation % waterfall chart
- ROAS improvement comparison
- Full optimization summary table
Key Recommendation: Reallocate $2,450/week from Social Media (-20%) to Email (+16%) and SMS (+10%). Total budget remains unchanged at $39,168/week.
| Channel | Current | Optimized | Change |
|---|---|---|---|
| $7,807 | $9,050 | +16% ↑ | |
| SMS | $4,054 | $4,471 | +10% ↑ |
| Paid Search | $15,146 | $15,936 | +5% ↑ |
| Social Media | $12,161 | $9,711 | -20% ↓ |
Geo-lift test results measuring causal impact of email campaigns:
- Test vs Control region revenue comparison
- Weekly incremental lift over time
- Average lift by period (Pre-Treatment vs Treatment)
- Statistical significance confirmation
Key Finding: Email campaigns generated ~$2,954/week in incremental revenue during the treatment period — a 12.4% causal lift above baseline (p < 0.05).
| File | Description | Rows |
|---|---|---|
01_weekly_performance.csv |
Weekly revenue, spend, conversions | 104 |
02_channel_attribution.csv |
Channel-level spend, revenue, ROAS | 416 |
03_budget_optimization.csv |
Current vs optimized budget per channel | 4 |
04_geo_lift.csv |
Test vs control geo-lift data | 104 |
05_kpi_summary.csv |
High-level KPI summary | 10 |
01_weekly_performance[date] → 02_channel_attribution[date]
01_weekly_performance[date] → 04_geo_lift[date]
-- Overall ROAS
Avg ROAS =
DIVIDE(
SUM('01_weekly_performance'[revenue]),
SUM('01_weekly_performance'[total_spend]),
0
)
-- Incremental Lift (Treatment Period)
Avg Treatment Lift =
CALCULATE(
AVERAGE('04_geo_lift'[incremental_lift]),
'04_geo_lift'[is_treatment] = "Treatment"
)
-- Budget Reallocation Amount
Reallocation Amount =
SUMX(
FILTER('03_budget_optimization',
'03_budget_optimization'[change_pct] > 0),
'03_budget_optimization'[optimized_spend] -
'03_budget_optimization'[current_spend]
)
- Clone or download this repository
- Open Power BI Desktop
- Click Get Data → Text/CSV
- Load all 5 CSV files from the
data/folder - Set relationships:
01_weekly_performance[date]→02_channel_attribution[date]01_weekly_performance[date]→04_geo_lift[date]
- Build visuals following the screenshots
🔗 Related Project This dashboard is the visualization layer for the full MMM analysis: Company A — MMM & Incrementality Testing (Data simulated based on MMM model outputs from the R/Robyn project.)
| Tool | Purpose |
|---|---|
| Power BI Desktop | Dashboard development |
| DAX | Custom measures and calculations |
| Python | Data simulation |
| R + Robyn | Underlying MMM model |
Azar Taheri LinkedIn