The 30-agent pilot: scoring, research briefs, and the 14-day sequence.

15 — Realtor Targeting Strategy (Micro-Account-Based Pilot)

Prepared 2026-07-22. Figures verified as of this date unless marked otherwise. Binding inputs: research_notes/DECISION_BRIEF.md (strategy), compliance file 05 (research_notes/compliance_outreach.md — cited below as 05 §A–§I), research_notes/local_market.md (agent landscape).

Operating premise (Model D, "Realtor Scenario Desk"): Mitchell does not cold-carpet-bomb the MLS roster. He runs a pilot of 30 named agents/teams, researched one at a time, approached with one specific, legitimate value offer each, on a manual, low-volume, fully compliant 14-day sequence. Positioning line, verbatim from the Decision Brief: "I'm not asking you to replace your preferred lender. Give me one difficult scenario or one buyer who needs a second look."

---

1. The pilot list of 30 — construction spec

Do not fabricate names into this plan. The list is built by Mitchell from public sources only (brokerage rosters, agents' own websites, Zillow/Realtor.com public profiles, public MLS-fed sold data on portals, LinkedIn public pages, Google reviews). No scraping tools, no data brokers, no purchased contact lists (05 §D2 platform-ToS ban; MASTER_PROMPT guardrails). Populate data/realtor_prospect_template.csv as the working file.

Slot allocation across the wedge (Tinley Park / Oak Forest / Orland Park / Lockport, anchored on the Harlem Ave corridor — see local_market.md §8). Revised after red-team review (M3): the warm network leads; the cold corridor follows. For a reserved newcomer with no book, 6–8 warm first at-bats are worth more than any cold source pool — and they enter the sequence FIRST (weeks 1–2), before any cold wave:

Production banding (target mix within the 30): - ~6 slots: newer agents (0–3 yrs, 2–8 sides/yr) — fellow-new-agent script #15; grow together. - ~18 slots: working mid-producers (8–25 sides/yr, buyer-side heavy) — the core target. Busy enough to have hard scenarios, not so big they have a captive lender team. - ~6 slots: top producers / team leads (e.g., RE/MAX 10's publicly celebrated producers — Jack Gawron, Dave Shalabi, John Charleston are named on the office's own public page) — approached with the top-producer script #14 and zero expectation of near-term referrals; the goal is one overflow/second-look scenario.

Crosstown Realtors conflict flag (scope corrected after red-team M3): Crosstown is the family company and runs a Zillow Flex team (Decision Brief). Flex tracks a Zillow Home Loans contact-rate KPI — Flex-team agents score 1 on criterion 12 and NEVER occupy pilot slots. That is where the flag ends: extending it into non-use of Hybrid-tier/non-Flex Crosstown agents — or of family introductions to agents at other brokerages — was over-caution the red team correctly called strategy malpractice for this specific person. Hybrid-tier and non-Flex Crosstown agents are pilot-eligible, occupy the warm slots above, and are flagged in compliance_notes with the family relationship noted.

---

2. Scoring model — 12 criteria, 1–5 rubrics, public data sources only

Score each candidate 12×(1–5); total 12–60. Enter the pilot at ≥40 with no automatic disqualifier. Automatic disqualifiers regardless of total: known Flex-team membership scoring 1 on #12, or any prior opt-out (suppression list, 05 §F).

Data hygiene: collect business information only — name, brokerage, office address, public business phone/email, public production signals. No personal addresses, no personal data beyond what the agent publishes about their business (MASTER_PROMPT L552). Log the source of every phone/email in the CSV (05 §F recordkeeping — 5-yr retention).

---

3. One-page research brief (per top prospect) — 11 fields

Complete before Day 1 of the sequence. Target 30 minutes per brief. Template (all fields from MASTER_PROMPT L554–566):

1. Public business profile — name, brokerage/office, years licensed (IDFPR public lookup), team status, coverage towns.
2. Recent activity — last-12-month sides summary: count band, buyer/list split, price band, towns.
3. Likely client mix — FTB / move-up / investor / self-employed inference, with the evidence (not a guess presented as fact).
4. One concrete observation — a real, specific, checkable fact ("your last four buyer sides were Tinley townhomes under $340k"). This is the anti-fake-familiarity mechanism: the opener references their business, never feigned personal connection.
5. One relevant Ready capability to verify — the product/capacity claim Mitchell must confirm with his manager BEFORE stating it to this agent (e.g., "can Ready pair Cook County DPA with FHA?"). Nothing unverified is ever said on a call.
6. One personalized value offer — chosen from: scenario desk (1-hr ack / 4-hr brief SLA), condo-association pre-screen (Limited Review sunset 2026-08-03), DPA-stack payment worksheet for a live sub-$375k buyer, Cook-vs-Will tax payment comparison for a listing.
7. Best outreach channel — office landline vs. email vs. LinkedIn, based on what they publish. Realtor cells are presumptively DNC-protected personal numbers (05 §A2): default to office line and email first.
8. First message — the actual personalized first-email draft (from file 16, script #3, personalized with field 4).
9. Follow-up sequence — the 14-day plan below, with any agent-specific modifications (e.g., skip in-person if no open houses scheduled).
10. Compliance note — channel constraints for THIS agent: number source and DNC status, any brokerage lender-affiliation sensitivity, no RESPA-adjacent offers.
11. Next action — the single next dated step, mirrored into the CSV next_action column.

---

4. The 14-day outreach sequence — day by day

Rules that govern every day (compliance file 05 cross-references in brackets):

• All calls: manual, human-dialed, live voice, 8am–9pm — practically business hours. Open every call: name + "Ready Mortgage Lenders" + purpose + permission ask [05 §A3, §A4, §C — 815 ILCS 413/15]. No dialers, no RVM, no AI voice [05 §A3, §E].
• Numbers: scrub against the National DNC Registry unless verified office landline; registry access runs through Ready's account [05 §A2, §F4 — [UNVERIFIED whether Ready holds a Subscription Account Number — P0 question, file 27 P0-6]]. SAN fallback rule (red-team M4): until Ready's SAN exists, unscrubbed cells are NO-CALL under this plan's own rules — the sequence for those agents runs email → LinkedIn → verified office landline → in-person, with conversation expectations cut accordingly (the connect-rate assumptions behind "5 meaningful conversations/week" assume cell access). Ready obtaining a SAN (first 5 area codes are low-cost) is a Day-1 written request with an owner and date on the Decision Sheet (file 10 item 9b). Do not let the calling channel die silently — price the degraded channel into the week's targets and say so at the Friday review.
• Email: Day-1 email is 1:1 and genuinely personalized (low-risk under CAN-SPAM), but include the full footer anyway — Ready physical address, working unsubscribe, truthful subject — plus the ad disclosure block if any product content appears [05 §D, §H]. When in doubt, footer.
• Texts: NOT in the cold sequence at all. Text only after an agent has given Mitchell the number in a business context or engaged first [05 §B].
• LinkedIn: manual, personalized, no automation, no product/rate claims in DMs [05 §D2].
• Any opt-out, any wording, any channel → same-day entry on the cross-channel suppression list; all channels stop [05 §F].
• No gifts, no paying for anything of the agent's, ever [05 §G]. The only things delivered are Mitchell's own branded informational assets about his own services.

Six to seven touches max across four channels in 14 days. Any reply at any point exits the agent from the cold sequence into the relationship track (scripts #9–#12).

---

5. Weekly capacity math — 60–90 minute daily sprint

Editable assumptions, not facts (Decision Brief baseline parameters):

• Research brief: ~30 min per new agent (one-time).
• Sequence touches per agent across 14 days: ~55 min total (Day-1 email 15; call+VM 8; LinkedIn 5; asset personalization+send 12; second call 8; in-person amortized 0–15; close-out email 5) → round to ~60 min, i.e., ~30 min/agent/week while in sequence.
• Reply handling / warm follow-ups: reserve ~20% of the sprint.

Sprint budget: 75 min/day × 5 days = 375 min/wk; minus 20% reserve → ~300 min for cold-sequence work.

Load per cohort size (N = new agents entering per week; ~2N in sequence at steady state):

The naive steady-state doesn't fit, which is exactly why the ramp is phased:

• Weeks 1–2: 4 new agents/wk (briefs are the bottleneck; batch 2 briefs per day on Mon–Thu). Load ≈ 120 + 120–240 ≈ 240–360 min. Use the 90-min sprint days.
• Weeks 3–6: 5–6 new agents/wk once brief-writing is templated (~20 min/brief with the CSV pre-scored) and early cohorts have exited the sequence.
• Full pilot of 30 enters the sequence within ~6 weeks. Weeks 7–8: no new entries; sprint time shifts to relationship-track follow-ups and the 60-day parked-list review.
• Rule of thumb: never more than ~12 agents in the active 14-day sequence at once. Above that, personalization quality collapses — and personalization is the strategy.

---

6. Stop/continue rules (per agent)

Weekly review (Friday, 15 min, outside the sprint): update CSV statuses, verify every opt-out hit the suppression list, count the real KPIs — meaningful conversations, scenario requests, sequence completion rate — not dials (Decision Brief KPI north stars).

---

All outreach copy: honest new-LO positioning, no fake familiarity, no manipulative urgency, no borrowed experience. Anything Ready-specific stated to an agent must be verified with the manager first. NMLS placeholders [MITCHELL_NMLS] / [READY_NMLS] throughout; final disclosure blocks come from Ready compliance (05 sign-off item 1).