End-to-End SQL + BI Dashboard Project
Part of the Data Analyst Roadmap
In short
This is the project recruiters ask about most: raw data in, a clean dashboard out, with SQL doing the heavy lifting in between. It proves you can do the actual job.
The flagship portfolio piece — pull data with SQL, build a dashboard in a BI tool, and document the business question it answers, showing the full toolchain working together.
Think of it like this
Running a stall at the college fest
You buy stock, set up the table, price items, then keep a little chalkboard showing what's selling. By evening you know which item to restock tomorrow. An end-to-end project is that whole stall: data sourcing, cleaning, SQL and a board people can read.
Watch it happen
The build plan
1/6Pick a public dataset with several related tables, like an e-commerce sample with orders, items and customers.
The big ideas
What you need to know
Start with three questions
Pick questions a manager would ask, like "Which region grew fastest?" Build the dashboard to answer them.
Use a real database
Doing the SQL in a proper database, not only in Excel, shows the skill employers screen for.
One clean page
A focused page with KPIs on top and a few charts beats ten crowded tabs.
Show your work
Put queries in a GitHub repo and link the live dashboard. Reviewers want to see both.
See it in code
Walk through the code
1CREATE VIEW monthly_sales AS2SELECT3 DATE_TRUNC('month', o.order_date) AS month,4 c.region,5 COUNT(DISTINCT o.order_id) AS orders,6 SUM(oi.quantity * oi.unit_price) AS revenue,7 SUM(oi.quantity * oi.unit_price)8 / COUNT(DISTINCT o.order_id) AS avg_order_value9FROM orders o10JOIN order_items oi ON oi.order_id = o.order_id11JOIN customers c ON c.customer_id = o.customer_id12WHERE o.status = 'delivered'13GROUP BY 1, 2;1/4A view gives the dashboard one stable, tested source to read from.
One clean view feeds most of the dashboard. The BI tool just reads it, so all the logic lives in reviewed SQL.
Save yourself the pain
Common mistakes
Oops
Building twenty charts because the tool makes it easy.
Do this instead
Answer three to five clear questions. Each chart should have a title stating its finding.
Oops
Doing all calculations inside the BI tool with no SQL to show.
Do this instead
Put the logic in SQL views and keep the BI layer simple. It's easier to review and reuse.
Your mission
First milestone: one working view
Quick check
Your dashboard shows more orders than the orders table has rows. What's the likely cause?
Go deeper
- 1Pick a business question with genuine substance, not just 'here's a random chart'.
- 2Write the SQL first, then build the dashboard, keeping both in your portfolio repo.
- 3Include a short write-up: the question, your approach, and what you found.
More projects at this stop
Exploratory Data Analysis (EDA) on a Public Dataset
A/B Test Analysis Case Study
Business Metrics Deep-Dive Report
Data Cleaning Case Study (Messy Dataset)
Python + Pandas Analysis Notebook
Portfolio Website or GitHub Showcase
Stuck on this topic? Ask an Insider
Get 1:1 guidance from people who've walked this exact path — on the InsideEdge app.
