About Me | Projects

Hi, I’m Titus — a data analyst and founder of Monroe Analytics. I love to help businesses eliminate guesswork by turning messy data into clean, strategic insights. With a degree in mathematics and statistics, I’ve spent years learning how to break down complex problems and build solutions that work — not just in theory, but in the real world of business.

Whether it’s building automated dashboards, forecasting performance, or streamlining manual processes, I specialize in solving problems at their core. My favorite work doesn’t just involve charts or code — it’s about understanding the moving parts of your business and applying the right logic to bring clarity and control. I use tools like Power BI, Python, and Excel not because they’re flashy, but because they help me deliver smarter solutions, faster.

If you don’t have a technical background — that’s perfectly okay. Many of my clients don’t. I focus on keeping things simple, effective, and grounded in what actually matters to you: saving time, making better decisions, and improving results. I’m here to listen, collaborate, and turn your challenges into custom-built tools you can trust.

As for the picture — that’s my brother (right) and me (left) shortly after our first look at Machu Picchu at the end of our 1 day Inca Trail hike. There is no better feeling than seeing something that magnificent after having to work for it!

A Few of My Projects

Here’s a sample of some of the projects I’ve worked on.
They reflect a few of the areas I specialize in — from forecasting and automation to data storytelling and performance tracking.

If you’re curious about any of these, want to brainstorm a challenge you’re facing, or just want to talk through what’s possible with your data, feel free to reach out. I’m always happy to chat.

This is just a starting point — I’ll continue updating the page as I take on more work and build out new solutions.

Budget Optimization

Data Analysis Hub

Centralized promotional strategy dashboard for executive visibility

Client Problem

A large organization managing complex promotional strategies across multiple channels and product categories need an updated unified system for tracking how funds were allocated and performing.

Solution Built

  • Developed a scalable Power BI dashboard hub to track promotional spend across several strategic buckets (e.g., incentive programs, discounts, reimbursements)
  • Integrated multiple disconnected data sources, including setup files, transaction data, and planning inputs, into a unified model
  • Built dedicated dashboard pages for:
    • Program utilization and historical claim trends
    • Budget vs. actual analysis by category and segment
    • Forecast accuracy and spend pacing insights
  • Complemented the dashboard with lightweight Python/Streamlit tools to automate data prep and reduce manual input requirements

Impact

  • Reduced manual reporting time through automation and dynamic dashboard design
  • Enabled leadership and finance teams to monitor promotional activity in near real-time
  • Supported strategic reallocation of promotional funds based on performance and pacing trends
  • Laid the foundation for future enhancements like alerting, forecasting, and scenario modeling

Tech Used: Power BI, DAX, Power Query, Excel, Streamlit (Python), pandas

Claims Report Automation

Streamlined claims processing for a multi-million dollar budget

Client Problem

The manual claims tracking process was time-consuming and error-prone, requiring analysts to merge Excel sheets weekly and identify claims on hold.

Solution Built

  • Created a Streamlit dashboard to process claims, merge mapping files, and identify discrepancies.
  • Automated detection of claims released from hold using uploaded files.
  • Added multi-tab export and download functionality for client reporting.

Impact

  • Reduced processing time from 2+ hours to less than 5 minutes
  • Improved accuracy in identifying system-related issues
  • Enhanced usability for non-technical team members through drag-and-drop simplicity

Tech Used: Python, Pandas, Streamlit, Excel Automation

Order Allocation Reporting

Automated insights on order activity, fulfillment pace, and stock behavior

Client Problem

The business needed visibility into how certain products were being allocated and fulfilled across product classes. Without a way to regularly track order volumes, shipping timelines, or the ratio of stock vs. custom orders, operations teams couldn’t identify patterns, plan inventory, or respond to demand changes effectively.

Solution Built

  • Created a script to process and merge open and shipped order data from Excel files
  • Engineered new metrics such as:
    • Average order age by class
    • Stock order ratio vs. made-to-order
    • Weekly shipping totals for historical comparisons
  • Automatically writes performance summaries to a master Excel tracker for weekly reporting
  • Included logic to isolate metrics for time periods like “two weeks ago” to allow side-by-side comparisons

Impact

  • Enabled consistent tracking of order health across product classes
  • Improved inventory conversations with clearer data on aging and stock behavior
  • Saved time by automating data cleaning, merging, and calculations each week
  • Supported decision-making for operations and supply chain teams through transparent performance metrics

Tech Used: Python, pandas, Excel Automation

Import Allocation Engine

Automated tool for prioritizing inbound supply against backlog demand

Client Problem

With limited supply arriving from external manufacturing sites, the operations team needed a faster, smarter way to assign incoming machines to backlogged customer orders. The manual process of checking inventory status, shipment timing, and order priority was complex and inconsistent — leading to delays, misallocation, and lost visibility.

Solution Built

  • Designed and deployed a web-based application using Streamlit and Python that automatically matches inventory to open orders using layered business logic
  • Integrated multiple data sources (inventory reports, shipment data, order backlogs) and transformed them into a unified allocation model
  • Applied custom ranking and matching algorithms to evaluate units by availability status, expected arrival date, configuration, and order urgency
  • Created dynamic filtering options, model exclusions, and traceable output to improve control and transparency across teams
  • Exported a multi-tab Excel workbook with color-coded outputs, data validation, and prioritization guidance for operational teams

Impact

  • Reduced manual allocation effort by automating key logic across thousands of rows of data
  • Improved speed and consistency in backlog resolution through clear, repeatable rules
  • Enabled proactive inventory management by surfacing mismatches and unfilled demand in real time
  • Provided a scalable tool to support weekly allocation cycles across multiple product lines

Tech Used: Python, Streamlit, pandas, Excel (xlsxwriter)

Retail Forecasting Engine

Blended predictive model for short-term retail demand across multiple product lines

Client Problem

The business needed a reliable, dynamic forecasting solution to estimate short-term retail performance across key product categories. Traditional models lacked transparency, and monthly projections were often misaligned with real-time sales data, leading to missed expectations and reactive planning.

Solution Built

  • Developed a custom forecasting engine that combines three independent inputs to produce a consensus retail prediction:
    • Machine learning-based time series modeling
    • Weighted historical pattern extrapolation
    • External model integration from complementary internal tools
  • Incorporated logic for:
    • Daily and monthly weight-based normalization
    • Rolling forecasts by product group
    • Live adjustment based on current month-to-date performance
  • Automated output into formatted Excel dashboards to support monthly planning reviews

Impact

  • Improved forecast visibility by providing live, mid-month retail predictions across three major product lines
  • Enhanced accuracy and trust in forecast numbers by blending quantitative rigor with business-aligned logic
  • Saved hours of manual modeling and cleanup each forecast cycle
  • Created a scalable and adaptable forecasting system that can be reused month-to-month

Tech Used: Python, Prophet, pandas, Excel (xlsxwriter), time series weighting

Inventory Optimization with Data Analysis

Projected $75,000+ savings annually for a mid-sized retailer through smarter inventory management *Sample Project - not for a real company*

Client Problem

Retailers often overstock or understock inventory, leading to high carrying costs or missed sales. This project targeted a 12-month dataset covering 50 products across 5 regions to optimize inventory levels.

Solution Built

  • Analyzed 3,000+ records including stock, sales, lead times, and demand patterns
  • Calculated turnover rates, stock-to-sales ratios, and fulfillment rates
  • Used ABC analysis, trend modeling, and dynamic reorder point logic
  • Developed automated stock alerts and visualized insights in Python

Impact

  • Reduced overstock by 25% (50,000 units); reduced stockouts by 30%
  • Saved $50,000/year in carrying costs and recovered $25,000 in missed sales
  • Cut manual stock checks by 60% with automated alerts
  • Improved fulfillment rate to 95%+

Tech Used: Python, Pandas, Matplotlib

Customer Segmentation & Revenue Prediction

Predicted increased promotional revenue by $250K+ through machine learning-driven insights. *Sample Project - not for a real company*

Client Problem

Retailers often struggle to target promotions effectively. This project analyzed 10,000 transactions to better understand customers and predict how promotions impact revenue.

Solution Built

  • Clustered 1,000 customers using K-means based on age, income, and behavior
  • Identified 4 clear customer segments for personalized marketing
  • Trained a Random Forest model to predict promotion-driven revenue with 85% accuracy
  • Built custom insights for targeted promotional strategies

Impact

  • $250K–$300K annual revenue increase with segment-specific promotions
  • Reduced promo waste by 20% ($30K/year savings)
  • Improved promo targeting by 40% using clear segment profiles
  • Model performance: R² = 0.85, MAE = $10

Tech Used: Python, scikit-learn, pandas, matplotlib

Scroll to Top