Getting an apartment found by AI starts with the basics: make the property public, crawlable, easy to understand, and consistent across the web.
If you have already read the Dashboard chapter in Boost Occupancy, you know the goal: get the most important property performance numbers into one simple weekly view.
The challenge is usually not understanding the dashboard. The challenge is loading the data.
Property management systems like Entrata, Yardi, ResMan, RealPage, and AppFolio all produce useful reports, but they do not always organize the information the same way. Even within the same system, weekly reports can be hard to compare if the data is spread across multiple sections.
That is why we created the Boost Occupancy Weekly Property KPI Extractor.
This is an AI prompt designed to help ChatGPT read one or more weekly property performance reports and turn them into a clean dashboard-ready table. The prompt tells ChatGPT what sections to look for, how to identify the correct rows, how to normalize different labels, and how to format the final output.
The prompt is designed to extract the weekly KPIs that matter most for occupancy review, including:
The script is not looking for one exact report format. It is designed to interpret common sections such as Availability Summary, Property Pulse, Lead Activity, Lead Conversions, Application Activity, and Make Ready Status
The workflow is simple.
First, download or export your weekly property reports from Entrata or your property management system.
Next, open ChatGPT and paste in the Weekly Property KPI Extractor prompt.
Then upload one or more weekly reports.
Finally, ask ChatGPT to follow the prompt, extract the data, and create the combined KPI table.
The intended output is a standardized Excel table with one row per week and columns for each dashboard metric. The prompt specifically asks ChatGPT to export the result as Weekly Property KPI Table.xlsx and include a short summary noting any missing or inferred values.
This process saves time, but the bigger benefit is consistency.
Instead of manually digging through multiple reports every week, you can use AI to create a repeatable first pass. It can read the reports, find the totals, normalize the labels, and organize the numbers into the same structure each time.
That gives you a cleaner starting point for the dashboard.
You should still review the results before relying on them. AI can misread a report, especially if the file is poorly formatted or a section is missing. But when used correctly, it can dramatically reduce the manual work required to prepare weekly occupancy data.
The point is not to let AI make decisions for you.
The point is to let AI do the extraction work so you can spend more time reviewing the numbers, spotting trends, and asking better operating questions.
<!--
Boost Occupancy — Weekly Property KPI Extractor
-------------------------------------------------
Author: Boost Occupancy
Version: 1.0
License: © 2025 Boost Occupancy. All Rights Reserved.
Description:
This Markdown file defines an AI prompt specification for extracting
weekly KPI metrics from multifamily property performance reports.
Supported Platforms:
• Entrata
• Yardi
• ResMan
• RealPage
• AppFolio
• Other Excel-based PMS exports
Usage:
Paste this prompt into ChatGPT (or another GPT-based assistant)
before uploading one or more weekly property performance XLS/XLSX files.
The assistant will follow these rules to generate a standardized
KPI report across all weeks and platforms.
-->
## 🧩 Section Overview
Each PMS export contains one or more data blocks (sections).
GPT should locate and interpret these sections by title and structure:
| Section | Typical Title Example | Purpose |
|----------|-----------------------|----------|
| Availability Summary | `Availability (As of …)` | Occupancy and Pre-Lease metrics |
| Property Pulse | `Property Pulse (…)` | Leasing and movement activity |
| Lead Activity | `Lead Activity (…)` | Leads and Tours |
| Lead Conversions | `Lead Conversions (…)` | Applications |
| Application Activity | `Application Activity (…)` | Application Denials |
| Make Ready Status | `Make Ready Status (…)` | Vacant Ready / Not Ready breakdowns |
---
## 🧱 Extraction Rules
### 1️⃣ Locate and Interpret Sections
- Identify each section by its header (case-insensitive).
- Find the **“Total:”** row within that section.
- Read all numeric values under relevant column headers.
- If a section is missing, leave its metrics blank and note it in the summary.
### 2️⃣ Field Mapping
| Target KPI | Source Section | Common Header Keywords |
|-------------|----------------|-------------------------|
| **Occupancy** | Availability | Occupied, Physical Occ % |
| **Pre-Lease** | Availability | Leased, Pre-Leased, Economic Occ % |
| **Leases** | Property Pulse | Leased |
| **Move-Ins** | Property Pulse | Move-Ins |
| **Move-Outs** | Property Pulse | Move-Outs |
| **Skips** | Property Pulse | Skips |
| **Evictions** | Property Pulse | Evictions |
| **Leads** | Lead Activity | New Leads, Inquiries, Traffic |
| **Tours** | Lead Activity | Visits, Tours, Showings, In-Person, Self-Guided |
| **Applications** | Lead Conversions | Application |
| **App Denials** | Application Activity | Denied, Denials, Declined Apps |
| **Vacant Ready** | Make Ready | Vacant Rented + Vacant Unrented (Ready) |
| **Vacant Not Ready** | Make Ready | Vacant Rented + Vacant Unrented (Not Ready) |
---
## 📏 Smart Scaling for Percentages
For **Occupancy** and **Pre-Lease**:
- If value ≤ 1 → treat as fraction → multiply by 100.
- If 1 < value ≤ 100 → use as-is.
- Clamp between 0–100%.
- Format with two decimals (e.g., `88.05%`).
If the value appears inflated (e.g., 8800%), divide by 100 automatically.
---
## 📊 Output Format
### Table Columns
`Week End | Occupancy | Pre-Lease | Leases | Move-Ins | Move-Outs | Skips | Evictions | Leads | Tours | Applications | App Denials | Vacant Ready | Vacant Not Ready`
### Formatting Rules
| Field | Format |
|--------|---------|
| Week End | mm/dd/yyyy |
| Occupancy / Pre-Lease | Percent (0.00%) |
| All Others | Integer |
| Sort Order | Ascending by Week End |
---
## ✅ Validation and Summary Notes
After processing, GPT must provide:
1. A **downloadable Excel file** named **Weekly Property KPI Table.xlsx**
2. A concise summary below the table, such as:
> ✅ Processed 9 reports successfully
> ⚠️ “Application Activity” missing in 2 reports
> ⚠️ “Tours” inferred from “Visits/Tours” headers
> 📊 Output saved as *Weekly Property KPI Table.xlsx*
---
## 💡 Best Practices
- Always look for the **Total:** row inside each section.
- Match by **meaning**, not by exact spelling (e.g., “Pre-Leased %” ≈ “Leased %”).
- If multiple “Application” sections exist, use the one that also contains “Lease” or “Conversion”.
- When unsure, infer based on proximity and structure (e.g., last numeric row before totals).
---
## 🪄 Example Instruction to GPT
> You are a data extraction assistant for **Boost Occupancy**.
> Read all uploaded weekly property performance files (Entrata, Yardi, RealPage, etc.).
> Follow the extraction and normalization rules in this prompt.
> Create a combined KPI table with one row per week and export it as an Excel file.
> Include a short summary of missing or inferred values.
---
## 🧠 About Boost Occupancy
**Boost Occupancy** helps apartment owners, operators, and asset managers fix, optimize, and forecast property occupancy faster.
This extraction framework is part of our broader **Occupancy Optimization Toolkit**, empowering multifamily professionals to understand performance metrics with speed and accuracy.
---
© 2025 **Boost Occupancy**. All Rights Reserved.
For licensing, usage, or collaboration inquiries, contact: **info@boostoccupancy.com**
Getting an apartment found by AI starts with the basics: make the property public, crawlable, easy to understand, and consistent across the web.
Asking “What’s blocking you?” helps leaders quickly uncover the real bottlenecks, support their team more effectively, and turn hidden friction into forward progress.
Use this secret shopping checklist to find missed calls, broken links, confusing listings, weak follow-up, and other issues that stop renters before they ever become leads.
Nathan R,
Asset Manager
