The AI Foundation for Real Estate Investing
Your investment data, structured in a pre-built real estate database and ready for your enterprise AI. Get consistent answers you can trust.
The problem
Most firms use AI. Few trust what it tells them.
The problem isn't the AI. It's the data underneath: Excel models with a new version at every turn, terms locked in PDF agreements, siloed systems that each hold part of the picture, and knowledge that leaves when people do.
Read the research: why AI fails without trusted investment data →The solution
From your investment data to answers you can trust.
No migration and no data project first. Your systems stay where they are.
- 01
Your models and documents
Excel models, loan and lease PDFs, KPIs and trackers, financials.
- 02
Flexible ingestion
Upload files, add them in the app, or send them straight from Claude or ChatGPT.
- 03
Structured foundation
One structured real estate record, reviewed by your team, with history on every change.
- 04
Trusted AI answers
Ask in Claude, ChatGPT or the Reapfolio app, with answers drawn from that reviewed data.
Speed
Analysts analyze instead of assembling data.
Consistency
Everyone works from the same reviewed record, so answers line up across the firm.
Insight
Maturities, rollover and exposure ready before the IC asks.
Durability
Knowledge stays when people move on.
Product in action
Ask about any investment. Get answers from reviewed data.
IllustrativeSample answers from an AI assistant such as Claude or ChatGPT connected to Reapfolio, using our synthetic demo portfolio, lightly edited. Pick a topic and scroll for the full analysis.
plot debt maturity for my portfolio
Your portfolio has 24 loans totaling roughly $790.6M in debt, concentrated on floating-rate development loans with banks and GSEs.
A few things stand out:
- 2026 is the crunch year — $350.8M matures, over 40% of total debt, spread across 11 loans (mostly industrial: Miami $67M, Richmond $44M, Rochester $38M).
- 2027 is the second wave — $202M, this time skewed toward residential (Stamford, Urbandale, Kalamazoo, Franklin, Atlanta).
- 2025 already has $134M concentrated in just two loans — Georgia Industrial ($68.8M) and Austin Industrial ($64.7M) — both with extended maturity options into 2027.
- The tail is light: 2028–2032 combined is only $104M across 4 loans.
Worth flagging that Richmond Industrial ($44M) and the other 2026 maturities may need refinancing conversations if they haven't already been addressed.
debt terms for georgia industrial
Here are the debt terms for Georgia Industrial:
| Term | Detail |
|---|---|
| Loan name | Georgia Industrial Development Loan |
| Lender | Demo Lender |
| Loan type | Development Loan |
| Interest rate type | Floating |
| Principal at closing | $68,800,000 |
| Maturity balance | $68,800,000 |
| Origination date | June 9, 2022 |
| Initial maturity date | June 9, 2025 |
| Fully extended maturity | June 9, 2027 |
| Loan guaranty | Yes |
So it's a $68.8M floating-rate development loan from Demo Lender whose initial maturity was June 9, 2025, but it carries an extension option pushing the fully extended maturity out to June 2027 — that's likely why it still showed up in the 2025 bucket in the chart above rather than being marked as already resolved. Worth confirming with the deal team whether the extension has been formally exercised.
Illustrative example. Not an actual screen from Claude or ChatGPT. All company names and data are fictitious; any resemblance to actual entities is coincidental.
Customer story
From pilot to trusted data source.
A U.S.-based real estate private equity firm with $1B+ in AUM and a diversified portfolio.
01 PROBLEM
The firm's investment, debt, leasing and performance data was scattered across Excel models and PDFs in different formats, and every quarterly reforecast added another version.
02 WHAT WE BUILT
- Expanded the data model to the firm's needs across investments, debt and leasing.
- Scalable extraction from underwriting and reforecast models, comparable across periods and investments.
- Organizational knowledge preserved, with context normalized across property types for fund- and portfolio-level analytics.
03 RESULT
We proved the numbers in Reapfolio match the firm's own models, investment by investment, then went from a single-sector pilot to 300+ investments in under two months.
Reapfolio is now in daily use across the firm's asset-management team.
Why Reapfolio
An institutional technology team, without building one.
Purpose-built for real estate
A pre-built data model for investments, debt, leases and KPIs, linked from property to fund.
Works with your AI tools
Claude, ChatGPT and the Reapfolio app, and the foundation stays as AI models change.
Your data stays yours
Reviewed by your team, and exportable at any time.
Grows with your AI vision
We extend the data model and build new capabilities with you, as part of one subscription.
Available now
- Answers anywhere
- Property to fund
- Your knowledge base
- Documents on demand
On the roadmap
- Excel scenario analysis
- IC memo drafts
- Tell us your use case →
Why not just use ChatGPT or Claude on our files?
You should use them, with Reapfolio underneath. On their own they read raw files and can answer differently each time. Connected to Reapfolio, they draw on one structured record your team has reviewed.
Do we have to replace Excel, our CRM or dashboards?
No. Your systems stay where they are. Reapfolio fills the gap they leave: getting the data in, structured and reviewed.
How do we know the answers are right?
Data from your models is extracted the same way every turn and checked before your team reviews it. If something looks wrong, correct it from Claude or ChatGPT, the Reapfolio app or an Excel upload, and every answer uses the corrected data.
How long does it take?
No migration. No data project first. A pilot is set up in less than a week.
Let's talk aboutyour enterprise AI vision.
Start with a conversation. If it's a fit, we scope a pilot with you: a subset of your investments, loans, leases and Excel models, set up in less than a week.
- 01Add your data, as it is today
- 02Share your underwriting and reforecast models
- 03Ask from day one, in the tools your team uses