Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/stkchan/googleanalytics4-publicdataset-ecommerce-dashboard-powerbi
This dashboard uses Power BI Desktop as a visualization tool by extracting data from Google BigQuery.
https://github.com/stkchan/googleanalytics4-publicdataset-ecommerce-dashboard-powerbi
analytics bigquery dashboard portfolio portfolio-project powerbi sql
Last synced: 4 months ago
JSON representation
This dashboard uses Power BI Desktop as a visualization tool by extracting data from Google BigQuery.
- Host: GitHub
- URL: https://github.com/stkchan/googleanalytics4-publicdataset-ecommerce-dashboard-powerbi
- Owner: stkchan
- Created: 2024-05-25T01:03:47.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2024-05-25T05:06:04.000Z (8 months ago)
- Last Synced: 2024-09-30T10:41:32.290Z (4 months ago)
- Topics: analytics, bigquery, dashboard, portfolio, portfolio-project, powerbi, sql
- Homepage:
- Size: 1.4 MB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# GoogleAnalytics4-PublicDataset-Ecommerce-Dashboard-PowerBi
## Table of Contents
* [Project Objective](#Project-Objective)
* [Dataset](#Dataset)
* [ETL Process](#ETL-Process)
* [DAX Formulas](#DAX-Formulas)
* [Question](#Question)
* [Conclusion](#Conclusion)
## Project Objective
In this project, we aim to find insights from the Google Merchandise Store GA4 eCommerce public dataset. The primary objective is to use ETL (Extract, Transform, Load) by extracting data in BigQuery and visualizing it in Power BI to generate insights based on defined goals.
## Dataset
The dataset for this project is from the [Google Merchandise Store](https://shop.googlemerchandisestore.com/), an online store that sells Google-branded merchandise. You can find the dataset [here](https://developers.google.com/analytics/bigquery/web-ecommerce-demo-dataset), The dataset, available through the BigQuery Public Datasets program, contains a sample of obfuscated BigQuery event export data for three months from 2020-11-01 to 2021-01-31. It includes many columns such as event_date, user_pseudo_id (client ID in Google Analytics 4), device_category, device_brand, geo_region, geo_city, and eCommerce dimensions such as transaction_id, item_id (SKU), item_name, item_category, item_revenue, purchase revenue, and more.
You can find GA4 BigQuery schema :[here](https://support.google.com/analytics/answer/7029846?hl=en)
## ETL Process
* **Overall** - In the first step, I extracted the columns from the dataset that I wanted to use and unnested some columns that are stored as arrays, which is the normal format in Google Analytics 4 BigQuery. You can see my scrip [SQL](https://github.com/stkchan/GoogleAnalytics4-PublicDataset-Ecommerce-Dashboard-PowerBi/blob/main/unnested_table.sql) After ensuring there are no nested columns, I will create more tables to be used in Power BI for analysis in the next step.
* **Traffic Performance report** - In this step, I created a table for analyzing traffic performance. You can see my script [SQL](https://github.com/stkchan/GoogleAnalytics4-PublicDataset-Ecommerce-Dashboard-PowerBi/blob/main/traffic_performance.sql) I calculated the total number of users, sessions, pageviews, engagement rate, and more by grouping by UTM source, medium, and campaign to monitor traffic channels, and identify which device category and web browser have the most users. This provides an overview of performance in terms of users, sessions, and pageviews over three months.
* **Ecommerce Funnel report** - I have calculated the number of events at each step before customers complete their purchase and transaction in the [Google Merchandise Store](https://shop.googlemerchandisestore.com/) This was done by grouping each traffic channel for each step (session_start, view_item, select_item, add_to_cart, begin_checkout, add_payment_info, purchase) SQL Script is [here](https://github.com/stkchan/GoogleAnalytics4-PublicDataset-Ecommerce-Dashboard-PowerBi/blob/main/ecommerce_funnel.sql)
- Note: session_start is an event in GA4 that occurs when users enter the website and initiate a session. A session will time out after 30 minutes of inactivity."
* **Customer Journey by Pages** - In this step, I calculated the total number of users for each page on the website to determine which page has the most users. [SQL Script](https://github.com/stkchan/GoogleAnalytics4-PublicDataset-Ecommerce-Dashboard-PowerBi/blob/main/customerJourney_byPages.sql)
*
## DAX Formulas Used in Measures
**1.ConversionRate**
```DAX
ConversionRate = (DISTINCTCOUNT(dataset[transaction_id]) / DISTINCTCOUNT(dataset[user_pseudo_id])) * 100
```
**2.Repeat Purchase Rate**
```DAX
Repeat Purchase Rate =
VAR PurchasesPerCustomer =
SUMMARIZE(
FILTER(
dataset,
dataset[event_name] = "purchase"
),
dataset[user_pseudo_id],
"PurchaseCount", COUNT(dataset[event_name])
)
VAR RepeatCustomers =
COUNTROWS(
FILTER(
PurchasesPerCustomer,
[PurchaseCount] > 1
)
)
VAR TotalCustomers =
DISTINCTCOUNT(dataset[user_pseudo_id])
RETURN
DIVIDE(RepeatCustomers, TotalCustomers, 0) * 100
```
**3.Churn Rate**
```DAX
Start Customers =
CALCULATE(
DISTINCTCOUNT(dataset[user_pseudo_id]),
FILTER(
dataset,
dataset[event_date] >= DATE(2020, 11, 1) &&
dataset[event_date] <= DATE(2020, 11, 30) &&
dataset[event_name] = "purchase"
)
)
Retained Customers =
CALCULATE(
DISTINCTCOUNT(dataset[user_pseudo_id]),
FILTER(
dataset,
dataset[event_date] >= DATE(2021, 1, 1) &&
dataset[event_date] <= DATE(2021, 1, 31) &&
dataset[event_name] = "purchase" &&
dataset[user_pseudo_id] IN
CALCULATETABLE(
DISTINCT(dataset[user_pseudo_id]),
FILTER(
dataset,
dataset[event_date] >= DATE(2020, 11, 1) &&
dataset[event_date] <= DATE(2020, 11, 30) &&
dataset[event_name] = "purchase"
)
)
)
)
Customer Churn Rate =
VAR StartCust = [Start Customers]
VAR RetainedCust = [Retained Customers]
VAR LostCust = StartCust - RetainedCust
RETURN
IF(
StartCust = 0,
BLANK(),
DIVIDE(LostCust, StartCust) * 100
)
```
## Questions
The analysis that we are going to perform shall answer the following questions:
* Which traffic sources (e.g., organic, referral, direct) contribute the most to user sessions, pageviews, and engagement rates?
* How does user behavior (total users, sessions, pageviews) vary across different device categories (desktop, mobile, tablet)?
* What is the average session duration and engagement time across different campaigns, and which campaigns have the highest engagement rates?
* How does the engagement rate correlate with the type of web browser used by visitors (e.g., Chrome, Safari, Firefox)?
* Which item categories or specific items generate the most revenue and transactions, and what is the average purchase revenue for each category?
* What are the conversion rates and repeat purchase rates for different campaigns, and how do they impact overall revenue?
* What are the common user journeys on the website, and which pages have the highest drop-off rates during the customer journey?
## Conclusion
* Traffic Source and Engagement: Analyzing the traffic sources reveals that certain channels, such as organic and referral, significantly contribute to user sessions and engagement. Understanding these patterns helps identify the most effective channels for driving quality traffic to the website.
* Revenue and Transaction Insights: The analysis of revenue and transactions highlights which item categories and specific products generate the most revenue. It also sheds light on the effectiveness of different campaigns in terms of conversion rates and repeat purchases, providing valuable insights for optimizing marketing strategies.
* Customer Journey and Funnel Optimization: By examining the customer journey, we can identify key pages with high user drop-off rates and stages in the e-commerce funnel where users are most likely to abandon their journey. This information is crucial for improving the website's user experience and increasing the conversion rate by addressing pain points in the customer journey.