A real estate investor's working life runs on documents that don't speak to each other. The deal underwriting lives in Excel. The signed leases are PDFs in Drive. The broker emails are in your inbox. The property-management reports are PDFs in a different folder. The notes from the walkthrough you did six months ago are on a phone or in a notebook. When the lender asks why the cap rate on Property B looks different from what was modeled at acquisition, the answer lives across four of those silos and you have to assemble it on the fly.

AI notes won't replace your underwriting model or your property-management software. What they can do is hold the connective tissue — deal memos, lease abstracts, broker correspondence, walkthrough notes, market research — in one place where the agent can read across all of it. The same shape underwrites how [angel investors handle deal screening and portfolio tracking](/guides/founders-ceos/angel-investors-deal-screening/) — different asset class, very similar working pattern.

## A vault shaped around the portfolio

The shape that holds up across portfolio size is roughly: one top-level page per property or deal, with sub-pages for the underwriting memo, the lease abstracts, the operating notes, the broker and tenant correspondence, the inspection and walkthrough notes, and the disposition planning. A second top-level page for the portfolio view — the running pipeline, the market context, the lender relationships, the tax and entity structure.

Capy supports unlimited page nesting, so a heavy property — multifamily with twenty units, a value-add retail center — can fan out by unit, by tenant, or by phase without forcing structure on a stabilized single-tenant net lease. The whole vault is plain markdown. That matters because when you sit down to evaluate a new deal or answer a lender's question, you ask the agent to read across every relevant page in one query and write the answer in the time it takes to make coffee.

## Deal underwriting memos with the assumptions written down

The model in Excel handles the math. The thing the model doesn't hold is the rationale: why you used the rent comps from the four blocks east instead of the two blocks south, why you assumed the property tax reassessment would land at the lower end of the band, why you carried six months of vacancy on the upper-floor units. Six months later, when the assumption holds or doesn't, you can't quite remember why you made it.

A working setup: a deal underwriting memo sub-page for each property that holds the assumptions explicitly — three to five short sections, two paragraphs each. Why the rent assumptions, why the cap rate, why the exit timeline, why the construction or value-add cost estimates. After the deal closes, ask the agent to read the memo and the actual operating data after six months and after twelve, and to surface where reality is diverging from the assumptions. The conversation about whether to hold or sell becomes data-grounded instead of vibe-driven. (The same decision-rationale discipline is the spine of [contract negotiation with AI notes](/guides/founders-ceos/contract-negotiation-ai-notes/).)

## Lease abstracts the agent can actually search

Most real estate investors have leases sitting in a Drive folder. The leases are sometimes summarized in a leasing-broker abstract that's already drifted from the actual signed document. When the question comes up — does this lease have a CAM cap, does the tenant have a right of first refusal on the adjacent suite, what's the renewal option — the answer is in the lease and you have to open the PDF and read.

Drop the signed lease on the property page. It auto-converts to markdown via docstrange, which means the agent can read every page and treat it as searchable text the same as any other note. Ask the agent to summarize the lease in plain English with the unusual clauses called out, build a lease abstract page with the key economic and operational terms, and flag anything that contradicts the broker's summary. The conversion runs once per upload and the lease stays searchable from then on.

When the lender asks about the CAM structure across the portfolio, ask the agent to read every lease in the vault and produce a one-page comparison. The work that used to take a Saturday morning takes a few minutes.

## A pipeline database that's a database, not a forwarded email

Most active investors track deal pipeline in some combination of email-with-stars and a half-maintained spreadsheet. Both fail in the same way: the deal that needed a follow-up gets lost, the broker who hasn't sent anything in three months is forgotten, the deal you passed on for a specific reason gets re-introduced and you can't remember why you passed.

A working alternative: an inline pipeline database in the portfolio top-level page via the `:::database:::` directive — rows for property, broker, ask price, status, last contact, and outcome with rationale. The database lives directly inside the page, not in a separate tab. Each row links to a sub-page with the deeper context: the broker's offering memo as a markdown copy, the email thread, the underwriting notes if you took it that far.

When a deal gets re-introduced, ask the agent to search the pipeline and tell you whether you've seen it before, what the prior status was, and what the rationale was for the prior outcome. The conversation with the broker tightens to "here's what changed since last time," not "I think we looked at this."

## Broker and tenant correspondence held alongside the deal

Real estate is a relationship business and the relationships sprawl across email threads that nobody reads back through. The broker who's brought you three deals over five years. The tenant who's renewed twice and is approaching their next option. The lender whose terms you should remember without re-reading the engagement letter.

A working setup: a sub-page per significant relationship with the running thread of meaningful exchanges. After every call, email exchange, or in-person meeting, drop a one-line entry with the date and the substance. Two minutes per entry. Over a year, the page becomes the primary-source archive of the relationship.

Before your next conversation with that broker or tenant, ask the agent to read the page and draft a brief: what they last raised, what's open, what they care about. (The same context-holding habit is the spine of how [account managers keep client context from slipping](/guides/sales-accounts/account-managers-ai-notes-client-context/).)

## Market research and walkthrough notes that compound across deals

Market research for real estate is the kind of work that compounds badly across deals. You researched the submarket six months ago for a deal you passed on, and you'll research it again next month, because the prior research lives in a folder you've forgotten the name of.

A working setup: a market-context page per submarket with running notes — what you've read, what you've heard from brokers, what the rent and cap-rate trends look like, what the local-government action affecting the submarket has been. Drop entries as you encounter them. Two minutes each.

When the next deal in that submarket lands, ask the agent: read the submarket context across the last year and draft a one-page market summary for the deal memo. The memo is grounded in your actual prior research, which is more current and more specific than anything you'd get from a general market report.

The same submarket page is also where walkthrough notes accumulate. Record the walkthrough in Capy on your phone — voice notes as you walk through the property. The transcript comes back with speaker diarization if multiple people are present (labels like "Speaker 1: …") so you can tell who said what. Ask the agent to draft a walkthrough recap with sections for physical condition observations, market and neighborhood signal, and operational issues that need follow-up. Next time you're underwriting in the same neighborhood, the agent searches across walkthrough notes and surfaces the relevant context — you're not relying on memory.

## What this isn't

Capy doesn't replace your underwriting model, your property-management software, or your accounting system. The structured-data side of real estate investing still lives in the tools that handle structured data. Capy is for the unstructured side — the deal memos, the lease abstracts, the relationship threads, the walkthrough recaps, the market research — which is the part that's currently sprawled across email and Drive and a notebook.

It's also single-user by design. One investor, one vault. If your investing model is a multi-person partnership where each partner needs to edit the same artifacts with role-based permissions, that isn't this product. The shape that fits is the principal investor or asset manager running the personal connective layer alongside the structured tools. Capy doesn't claim regulatory certifications either — your fund's compliance posture stays your responsibility. Pricing tiers are on the [pricing page](/pricing/).

## A small first test

Pick the property you've owned the longest. Drop the signed lease, the most recent operating report, and your underwriting memo from acquisition into a Capy page. Ask the agent to write a one-page status summary: where reality has diverged from the underwriting assumptions, what the current operating signal suggests, and what the disposition timeline should look like given current market context. If the summary catches a divergence you'd otherwise have missed, that's the agent doing for you what your portfolio asks for and rarely gets.

[Try Docapybara free](/accounts/signup/). Load one property's documents and see what the agent does with them.