The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I built a Food Delivery Time Prediction project, and the most important thing I learned wasn't about the model.
It was about the data.
Before building any prediction model, I focused on understanding the operational patterns behind delivery performance and identifying the factors that actually influence delivery time.
Here’s what the workflow involved:
Dataset:
1,000 food delivery records across variables such as distance, weather, traffic level, vehicle type, courier experience, preparation time, and delivery duration.
Exploratory Data Analysis & Preprocessing:
• Handled missing values across Weather, Traffic_Level, and Time_of_Day
• Identified distribution patterns and operational trends within delivery timings
• Analyzed the impact of variables like traffic conditions, distance, and preparation time on delivery performance
• Performed data cleaning, transformation, and feature preparation for modeling
Models & Optimization:
• Built baseline and predictive models using Linear Regression and Random Forest Regressor
• Applied Hyperparameter Tuning using GridSearchCV with 5-fold cross-validation
• Optimized model performance using R² score evaluation
Key Takeaway:
Clean exploratory analysis and strong data understanding often contribute more to model performance than the algorithm itself.
Tools & Technologies:
Python | Pandas | NumPy | Matplotlib | Seaborn | Scikit-learn
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Here’s a polished version that keeps the story strong without sounding too casual:
I started working with Nomad Internet about three years ago, when their conversion rate was just 0.2%.
We began with a full conversion rate audit, then redesigned the website, improved the messaging, and created a customer qualifier that helped guide visitors to the right solution while also capturing leads. From there, our work continued across the customer journey, including landing pages, offers, ads, lead generation, and conversion strategy.
The results were significant. Nomad Internet grew from a very low-converting website into a business converting at 2–3% overall.
During that time, the company even asked me to step in as CEO, which I did for a period of time to help guide the business through a critical stage of growth. Ultimately, I chose to return to the work I enjoy most: helping businesses improve their marketing, websites, messaging, and conversion performance.
Today, Nomad Internet is a much stronger company and is now positioned for a potential sale.
🚀 Built an Automated Power BI Dashboard with Real-Time File Refresh Workflow
Recently, I worked on a complete Power BI automation project where I transformed messy raw CSV data into a fully interactive business dashboard.
This project included:
✅ Data Cleaning with Power Query
✅ Handling messy and mixed date formats
✅ Data transformation & automation
✅ KPI Dashboard Creation
✅ Sales & Revenue Analysis
✅ Dynamic Visualizations
✅ Folder-Based Automated Refresh System
✅ Bangladesh Regional Map Integration
The most exciting part of this project was the automation workflow.
Instead of manually rebuilding reports every time new data arrives, I configured Power BI to automatically detect updated files from a folder source. Once new CSV files are added, the dashboard can refresh and update business insights instantly.
During the project, I also solved several real-world data problems like:
inconsistent date formats
null values
mixed discount scales
data type issues
automation refresh errors
This project helped me strengthen my skills in:
📊 Power BI
📈 Data Visualization
🧹 Data Cleaning
⚡ Automation Workflow
📂 Power Query
🐍 Data Analytics
As a Data Analyst, I enjoy building dashboards that are not only visually appealing but also practical for real business decision-making.
If you need help with:
Power BI Dashboards
Data Cleaning
Automated Reporting
Business Insights
Excel & CSV Data Processing
feel free to connect with me or knock me anytime.
#PowerBI #DataAnalytics #BusinessIntelligence #Dashboard #DataVisualization #PowerQuery #Automation #DataCleaning #Analytics #DataAnalyst #DataScience #CSV #Excel #Reporting #BusinessAnalytics
Data Systems DesignGrowth MarketingBusiness Workflow AutomationMicrosoft Power BIGoogle GeminiChatGPT
Shipping Cost Drivers , Root Cause Analysis
Built a root-cause Tableau dashboard analyzing shipping cost drivers across carriers, destination zones, fees, and package weight. The analysis found that parcel carriers bill 2.5x actual package weight, creating a $922K annual excess billing impact, while Zone 6 costs 65% more per lb than Zone 2. The dashboard also includes carrier cost breakdowns and a fuel surcharge sensitivity model to estimate cost exposure from fuel increases up to 50%.
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
portfolioreview
The best portfolios tell a story, not just show a grid. Share yours for feedback.
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?