PWOOD GROUP
← All deployments
Commercial real estateExample build

Deal Flow Agent

Every day it reads property listings, public records and broker emails, and points out the off-market deals that fit what the fund buys.

Example business · Commercial real estate investment firm, NYC

1,900 listings → 14 dealsScreening live
Broker emailListing feedsPublic filingsSCORED VS THESISBrief3 matches

Outlier flagged: 31-yr hold, new LLC filed 9 days ago.

1,900

listings checked daily

14

good-fit deals a week

9 hrs

saved per analyst each week

Time to build

4 weeks

Typical build price

From $12,000

Running cost

$500–$1,000 a month

Pays for itself in

about 3 months

See it running

What it looks like on a phone.

The texts the agent sends and gets back, playing one real job from start to finish. Names redacted.

# acquisitions

Slack · 6:00 AM

D

Deal Flow Agent 6:00 AM

3 new fits from overnight. Ranked:

412 W 38th St

92

Long-held · new LLC

88 Bergen St

87

9.1% cap

1550 Jerome Ave

81

Lien cleared

👀 2✅ 1Open draft to owner →

In plain English

Like an analyst who reads every listing and filing in the city before 7 am.

An investment firm’s analysts spend their mornings reading listing sites, public records and broker emails to find the few buildings worth a look. The agent reads all of it every day, scores each property against what the fund buys and flags signs that an owner might sell quietly. Analysts start the day with a short list instead of a pile.

The problem

Before: analysts spent every morning reading listing sites and broker emails. The few deals that fit were a needle in a haystack.

Works inside the tools they already had

GmailSlackNotionAirtableClaude

What the agent does, step by step

  1. 1

    Reads the market

    It pulls listings, public filings and every broker email into one place.

  2. 2

    Scores each deal

    It rates each property on return, location, tenants and price per square foot, against what the fund looks for.

  3. 3

    Spots hidden chances

    It flags unusual signs, like a long-held building with a new company filing, that hint an owner may quietly sell.

  4. 4

    Briefs the team

    Each morning a ranked list lands in Slack. Each deal comes with comparable sales and a draft email to the owner.

What goes in, what comes out

It reads

  • Listing sites and broker emails
  • Public property and company records
  • The fund’s buying criteria
  • Past deals the team passed on, and why
Deal Flow Agent

It produces

  • ✓A ranked shortlist in Airtable each morning
  • ✓A one-page summary per property
  • ✓Owner-may-sell signals with the evidence
  • ✓A Slack digest for the team

What it actually sends

A real-format example of one message from the agent. This is what the team sees, in the tool they already use.

Slack · #deal-flow

Deal Flow Agent AGENT

6:45 AM

Today: 1,912 checked · 3 strong fits

Mixed-use, 14 unitsFit 91 · Brooklyn · $412/sq ft
WhyOwned 31 years, new company filed last week
Retail strip, 6 tenantsFit 84 · Queens · 2 leases end in 2027
Office, 40k sq ftFit 62 · outside target size
Skipped1,909 listings, reasons logged

One-pagers are in Airtable. Owner contact for the Brooklyn lead found in public records.

Open one-pagersAssign analystPass with reason

A normal day before

With the agent

7–10 AM: analysts read listings and emails

6:45 AM: ranked shortlist in Slack

Good deals found late, or missed

Quiet sale signals flagged the week they appear

Reasons for passing lost in email

Every pass logged, so the scoring learns

How we built it

Every build follows the same path: learn the job, connect the tools, run it next to your team, then switch it on.

  1. 1Week 1

    Write down the taste

    Interviewed the partners on what makes a good deal and reviewed 50 past yes and no decisions.

  2. 2Week 2

    Connect the sources

    Listing feeds, public records and the broker inbox, all into Airtable.

  3. 3Week 3

    Score and compare

    Ran the scoring on last quarter’s deals and checked it against what the team actually picked.

  4. 4Week 4

    Daily digest live

    The morning Slack digest started, with analysts marking each pick good or bad to sharpen it.

Your people still decide

  • ●Which deals to pursue
  • ●Any contact with owners or brokers
  • ●Pricing and offers

The agent asks before it acts on any of these. Every action it takes is logged, and you can switch it off at any time.

Builder’s note

A score alone is not useful to an analyst. Every pick comes with its evidence, the filing date, the years owned, the lease dates, so they can check it in a minute and trust it.

Phil Wood, founder, PWOOD GROUP

Example build based on a common workflow, not a named client. Figures show what the agent is designed to deliver.

Have a job like this? Book a free 30-minute call and we’ll tell you if an agent can do it.

Next: Dispatch Agent →

Give us 30 minutes. We’ll show you where the week goes.

A free video call. Tell us how your team spends a normal week. We’ll name the jobs an agent could take over, what that saves and roughly what it costs. If it isn’t worth it, we’ll say so, and you’ve spent nothing.

Example week

Inbox and follow-ups9.0 hrs/wk
Data entry7.0 hrs/wk
Reporting4.5 hrs/wk
Scheduling3.0 hrs/wk