Power BI Dashboard
Analyze, visualize, and forecast YouTube video performance using a 1000-row dataset with 20+ channel metrics.
The YouTube Channel Analysis and Forecasting project combines Power BI and Python to perform data analysis, visualization, and predictive modeling on YouTube channel performance data.
Using a dataset of 1000 videos containing 20+ key metrics such as views, impressions, CTR, watch time, RPM, and estimated revenue, this project helps creators understand what drives engagement and monetization — and forecast how future videos might perform.
To build a real-time, interactive Power BI dashboard that helps YouTube creators analyze their video performance — including views, engagement, revenue, and audience behavior — and forecast future trends.
This dashboard converts raw YouTube data into insights that guide better content planning and growth strategies.
- To analyze YouTube video performance and identify key engagement factors.
- To visualize content trends, traffic sources, and monetization metrics through Power BI.
- To build a machine learning model that predicts video performance levels (Low, Medium, High).
- To enable “what-if” forecasting for impressions and revenue growth.
- To deliver real-time, interactive dashboards for data-driven decision-making.
- Dataset Name: https://github.com/Mahalakshmi0807/Youtube-Channel-Analytics-and-Forecasting-/blob/main/youtube_videos_1000.csv
- Rows: 1000 Columns: 20+ Primary Key:
video_id
| Column Name | Description |
|---|---|
video_id |
Unique video identifier (Primary Key) |
upload_date |
Date when video was uploaded |
category |
Type of content (Education, Gaming, Tech, etc.) |
video_length_sec |
Duration of video in seconds |
views |
Total number of views |
likes, comments, shares |
Engagement metrics |
impressions |
How many times the video was shown |
ctr_percent |
Click-through rate (%) |
avg_view_duration_sec |
Average viewing time per user |
watch_time_hours |
Total watch time in hours |
estimated_revenue_usd |
Estimated ad revenue |
rpm_usd |
Revenue per 1,000 views |
traffic_source_primary |
Primary source of traffic (Search, Suggested, Shorts, etc.) |
subs_gained |
Subscribers gained from video |
contains_sponsor |
Whether the video is sponsored |
is_shorts_video |
1 = Shorts video |
is_monetized |
Whether video is eligible for ads |
The dashboard answers key business questions such as: 📊 What are the total views, likes, comments, and revenue?
🎯 What is the average engagement rate (likes + comments + shares ÷ views)?
🏆 Which are the Top 10 performing videos?
🧩 Which categories generate the most views and engagement?
🚦 Which traffic source brings the most audience?
📈 How have views and revenue changed month by month?
🔮 Can we forecast future views for the next few months?
🤩 Which traffic source contributes most to overall revenue?
🎥 How do Shorts compare with long-form videos in terms of views and monetization?
💎 What is the relationship between impressions, CTR, and total views?
🛍️ How does sponsorship impact engagement and revenue?
📽️ What factors most influence high-performing videos (based on ML analysis)?
🔍 What if impressions increased by 10–30% — how much would revenue grow?
- Open Power BI Desktop
- Click Get Data → Text/CSV → youtube_videos_1000.csv → Load
-
Change data types:
published_date→ Dateviews,likes,comments,shares,revenue→ Whole Number / Decimal
-
Rename columns for clarity if needed.
Go to Modeling → New Measure and add these formulas:
Total Views = SUM('youtube_videos_1000'[views])
Total Likes = SUM('youtube_videos_1000'[likes])
Total Comments = SUM('youtube_videos_1000'[comments])
Total Shares = SUM('youtube_videos_1000'[shares])
Total Revenue = SUM('youtube_videos_1000'[estimated_revenue_usd])
Engagement Rate =
DIVIDE(
[Total Likes] + [Total Comments] + [Total Shares],
[Total Views],
0
)
Go to Modeling → New Table and paste:
Date =
ADDCOLUMNS(
CALENDAR(
MIN('youtube_videos_1000'[published_date]),
MAX('youtube_videos_1000'[published_date]) + 90
),
"Year", YEAR([Date]),
"Month", FORMAT([Date], "MMM"),
"YearMonth", FORMAT([Date], "YYYY-MM")
)
Then:
- Go to Model view
- Connect
Date[Date]→youtube_videos_1000[published_date] - Right-click Date Table → Mark as Date Table → select
[Date]
- Total Views
- Total Likes
- Total Comments
- Engagement Rate (%)
- Total Revenue ($)
-
Bar Chart → Top 10 videos by Views
-
Pie Chart → Views by Category
-
Line Chart → Views by Month (Date table)
-
Donut Chart → Traffic Source by Views
-
Table → All videos with columns:
- Title
- Category
- Views
- Likes
- Comments
- Estimated Revenue
- Engagement Rate
- Select your Views by Month line chart
- Go to Analytics Pane → Forecast → Add Forecast
- Set Forecast length = 3 months
| Section | Visual | Purpose |
|---|---|---|
| Header | Text “YouTube Channel Analysis Dashboard” | Title |
| Row 1 | KPI Cards | Key performance metrics |
| Row 2 | Bar + Pie Charts | Top videos + category comparison |
| Row 3 | Line + Donut Charts | Monthly trend + Traffic source |
| Row 4 | Table | Detailed video-level data |
| Row 5 | Forecast | Predict future performance |
🎨 Design tips:
- Use white cards, light background, shadows
- Use red (YouTube color) for highlights
- Add YouTube logo on the top corner
After creating the dashboard, we can see:
- 🔝 Top 10 videos bring almost 80% of total views
- 📚 Educational videos have the highest engagement
- 💬 Higher likes and comments lead to higher revenue
- 🌍 Suggested Videos and YouTube Search are main traffic sources
- 📈 Forecast shows steady growth for next months
- ⏱️ Shorter videos (under 5 mins) have higher engagement rate
We get the information that are,
- Education and Tech categories generated the highest total views and revenue.
- Suggested and Browse traffic sources were the most effective for audience reach.
- Shorts videos gained more views but had lower RPM compared to long-form content.
- Videos with higher CTR and average view duration performed significantly better.
- Sponsorship had a mixed impact — slightly higher views but inconsistent RPM.
- Machine Learning results showed that impressions, watch rate, and engagement were top predictors of success.
- What-if analysis revealed that a 20% increase in impressions could raise estimated revenue by ~18%.
| Category | Tools / Libraries |
|---|---|
| Data Cleaning & Analysis | Python, Pandas, NumPy |
| Visualization | Power BI |
| Dashboard / BI | Power BI Desktop & Service |
| Environment | Jupyter Notebook, VS Code |
| Version Control | Git, GitHub |
Focus Areas: Data Visualization • Predictive Analytics • Realtime Dashboards • BI Reporting
- 📘
youtube_videos_1000.csv— Dataset - 📊
Youtube Analytics Dashboard.pbix— Whole datas and Dashboard in powerbi - 🗂️
YT channel Analytics Dashboard screenshot— Screenshot of the Dashboard - 📊
Project report— The whole report of the project - 📘
README.md— Full documentation
The YouTube Channel Analysis and Forecasting project demonstrates how data analytics, visualization, and machine learning can work together to improve decision-making in digital media.
By leveraging Power BI dashboards and predictive models, creators can identify what drives success, optimize upload strategies, and plan future content more effectively.
This project not only enhances analytical skills but also showcases the integration of business intelligence tools with AI-driven forecasting — turning YouTube data into actionable insights for channel growth.
Maha
📍 Villupuram, Tamil Nadu
💼 BCA Graduate | MBA Aspirant | Data & Analytics Enthusiast
📧 maharagupathi05@gmail.com
🌟 Transforming YouTube analytics into actionable insights using Data Science and Power BI. 🌟