Lease Renewal Analysis

Portfolio Project Summary

Role: Data Analyst Tools: PostgreSQL, SQL, Exploratory Data Analysis Dataset Size: 79,850 rows · 18 columns

This project investigates the factors that influence whether residents renew their leases. The goal was to translate raw lease data into clear business insights, quantify renewal drivers, and recommend actions that improve tenant retention.

---

Business Problem

Property management leadership needed to understand:

• What behaviors and conditions lead to lease renewal
• Which factors discourage renewal
• How confident the organization can be in these findings
• What additional data is required to improve decision‑making

---

Data Overview

• Binary indicator dataset (0/1) for all variables except lease ID
• Covers rent changes, lease tenure, demographics, violations, survey sentiment, payment behavior, and HOA status
• Key data gap: Age group missing for 42,467 leases (all zeros instead of nulls)

---

Methodology

1. Data Cleaning & Validation (PostgreSQL)

• Reviewed tables for spelling issues, whitespace, and type mismatches
• Checked for nulls and inconsistencies
• Confirmed missing age values were encoded as zeros

2. Exploratory Analysis

SQL queries were written to compute:

• Renewal counts and renewal rate
• Rent change distributions
• Lease term segmentation (1st, 2nd, 3rd term)
• Age‑group trends
• Violations, survey feedback, payment history, and HOA impact

3. Insight Generation

KPIs and percentage breakdowns were translated into stakeholder‑ready findings and business recommendations.

---

Key Insights

Renewal Drivers

• Residents with no rent increase renewed most frequently (excluding new leases).

• First‑term residents produced the highest renewal counts.

• Ages 30–39 showed the strongest renewal concentration.

• Residents with prior fines or late payments still renewed at higher counts than those without.

• Non‑mandatory HOA leases generated more renewals overall.

• Rent‑increase behavior varied by tenure:

  • 10% increase → more third‑term renewals than second‑term
  • 20% increase → more second‑term renewals than third‑term

• Positive survey responses were most common among first‑term renewals.

Non‑Renewal Signals

• 20% rent increases led to the most non‑renewals, followed by 10% increases, then no increase.

• First‑term residents were most likely to leave.

• Highest non‑renewal age groups (descending):

  1. 60+
  2. Under 24
  3. 50–59
  4. 25–29
  5. 40–49
  6. 30–39

• Non‑renewals were more common among residents with:

  • Prior fine violations
  • Negative survey feedback
  • Late payment history

• Higher non‑renewal counts also appeared where HOA was not mandatory, suggesting HOA communities may support retention.

---

Business Recommendations

• Collect missing age demographic data to strengthen segmentation.
• Investigate residents who did not renew despite stable rent.
• Analyze churn among first‑term tenants.
• Reevaluate fine/violation policies and their tenant impact.
• Perform deeper survey sentiment analysis.
• Study location‑level retention trends.
• Gather tenant and employee sentiment for qualitative insight.

---

Portfolio Value

This project demonstrates the ability to:

• Clean and validate large operational datasets
• Write analytical SQL for KPI generation
• Translate data into business‑ready insights
• Deliver actionable retention strategy recommendations

---

How To Use / Reproduce Analysis

1. Import the lease dataset into PostgreSQL.
2. Validate schema, data types, and missing values.
3. Run SQL aggregations to compute renewal KPIs and segment trends.
4. Compare renewal vs. non‑renewal behavior across rent, tenure, demographics, and sentiment.
5. Translate findings into retention strategy recommendations.
6. Extend analysis with demographic, geographic, and sentiment data where available.