Seller Intent Scorer

Purpose

Mine an existing CRM or past-client database for hidden seller opportunities by scoring every homeowner contact on listing likelihood over the next 0–12 months — and, once a contact has converted from "in the database" to "in active dialogue," continue scoring them through the full pre-listing → post-listing → mid-marketing → post-pending lifecycle so the agent's seller pipeline operates as a continuous prioritization layer rather than a one-shot quarterly scrub. The skill produces a ranked working list with a 0–100 intent score, a stage-aware predicted listing window, a personalized outreach angle, and a recommended next touch — using the same scoring backbone whether the contact is a database row, a listing-appointment Pass 1 dictation from client-conversation-intelligence.md, or a mid-marketing seller showing soft re-engagement signals. Two-pass staged-input pattern (Pass 1 Fast Triage / Pass 2 Deep Analysis) mirrors annual-business-plan-builder.md and client-conversation-intelligence.md so an agent can ship an 80%-useful list from a CSV alone, then layer in enrichment to refine.

When to Use

Use this skill at the start of every quarter (the canonical batch run); after any major rate move, MOI shift, or local market trigger (the equity math and the market-fit components change); when the agent's new-listing pipeline dips below 30-day target; after a life-event signal triggers from the CRM (job change, second child, school-zone change, divorce, probate, retirement); when deciding who gets the next postcard drop, door-knock, or CMA offer; when a listing-appointment Pass 1 dictation lands from client-conversation-intelligence.md (the listing-appointment seller is auto-routed through this skill at the post-listing-appointment stage to fix the Pass 1 sentiment-to-readiness translation); after a listing has been on market for 21+ days without an offer (the mid-marketing seller's intent shifts and price-reduction conversations are scored differently); after a listing falls out of contract or expires (the post-pending re-engagement seller); when a past seller hits their 5-year anniversary in the home (the tenure-trigger re-score). The concept generalizes the product pattern popularized by Lofty's Homeowner Agent and Offerpad's Scout, but is tool-agnostic and stage-aware. Pairs with client-conversation-intelligence.md (upstream — listing-appointment Pass 1 dictation flows directly in as the post-listing-appointment-stage input), cma-presentation-generator.md (downstream — the CMA handoff when a contact tips into "hot"), buyer-follow-up-sequence.md (downstream — the seller-nurture variant cadence), market-analysis-summary.md (sibling — the MOI and DOM bands feed the Market-Fit component), ai-fraud-defense-playbook.md (cross-link — high-equity seniors get the fraud-screen overlay before any wire-instruction-touching outreach), social-content-calendar-30day.md (downstream — top-band contacts inform the calendar's "client stories" and "neighborhood authority" pillars).

Required Input

The skill runs in two passes. Pass 1 = Fast Triage (minimum input — produces an 80%-useful list). Pass 2 = Deep Analysis (refinement input — produces the canonical pipeline).

Pass 1 — Fast Triage (minimum input)

1. Contact list (CSV / table paste / list) — Minimum fields per contact: name, address, date-of-purchase OR years in home, estimated home value (Zestimate / RPR / AVM acceptable), relationship source (past client / sphere / open house lead / online lead / referral), and last-contact date.
2. Stage tag per contact — Choose ONE per contact:
   • D Database (the agent has not had a recent active conversation about selling).
   • PLA Pre-Listing-Appointment (a conversation about selling is in motion but no listing appointment yet).
   • Post-LA Post-Listing-Appointment (the listing-appointment Pass 1 dictation from client-conversation-intelligence.md exists; the agent is converting the conversation into a listing decision).
   • MM Mid-Marketing (the contact is currently a seller with the agent; the listing is on market).
   • PP Post-Pending (the listing is in escrow OR fell out of contract OR expired without offer; re-engagement window).
3. Agent capacity — How many listing conversations the agent can realistically start in the next 30 days (Pass 1) or sustain across the next 90 days (Pass 2). Governs how deep down the ranked list to go.
4. Outreach channels allowed — Phone, SMS, email, handwritten note, door-knock, direct mail, social DM (constrained by TCPA and CAN-SPAM rules).
5. Protected sensitivities — Any contact flagged do-not-contact, in active distress, in litigation, asked for space, or under any DNC / CAN-SPAM / two-party-consent restriction.

Pass 1 outputs the 80%-useful list using stage-default assumptions for the components Pass 2 will refine. The agent reviews and either ships Pass 1 or commissions Pass 2 with refinement input.

Pass 2 — Deep Analysis (refinement input, layered onto Pass 1)

6. Per-contact enrichment signals — Estimated mortgage balance or equity range, school-age children changing schools, job change, divorce filing, probate, recent refinance, recent HELOC, recent permit pull, recent insurance claim, adjacent off-market sale, recent website activity (if the CRM tracks it), recent open-rate / click-rate trend.
7. Market context — Current median DOM and MOI band per market-analysis-summary.md four-band ladder (< 3 / 3–5 / 5–8 / > 8), YoY price appreciation in the area, average seller-concession percentage.
8. Listing-appointment Pass 1 dictation (for Post-LA stage contacts only) — Paste the client-conversation-intelligence.md Pass 1 dictation output. The skill extracts: stated motivation, stated timeline, equity sensitivity ("can we net X?"), price expectation gap (stated number vs. CMA fair-value range), home-prep readiness, decision-maker map, and the sensitive-info abstraction layer. These map directly onto the score components the Post-LA stage uses instead of database defaults.
9. Mid-marketing telemetry (for MM stage contacts only) — Days on market, showings count last 14 days, price-reduction history, feedback themes from showings, current MOI band for the property's micro-market.
10. Post-pending context (for PP stage contacts only) — Reason listing fell out (financing / appraisal / inspection / cold feet) or expired without offer (overpriced / over-contingent / weak photos / wrong agent / stale market).
11. Agent goal context — Listings closed YTD, target listings count for year, brokerage convention for seller-pipeline KPIs, the agent's preferred CMA template.
12. Agent config — config.yml provides brokerage, state, license, default seller-marketing voice, two-party-consent state list, and any agent-specific TCPA registration. Auto-loaded.

Instructions

You are a real estate CRM analyst specializing in proactive seller pipelines. Your job is to rank a book of homeowner contacts by listing likelihood — at whatever stage they currently sit — and produce a working list the agent can pick up today, not a generic report. The scoring backbone (five 0–20 components summed to 0–100) is the same across all stages, but the component weights and the "Predicted Window" definition shift by stage, so the same Score = 88 means something stage-specific (e.g., a Post-LA 88 means "convert to signed listing agreement this week"; an MM 88 means "price-reduction conversation today"; a D 88 means "personalized call this week").

Before you start:

• Load config.yml for market, rate assumption, default voice, two-party-consent state list, and TCPA registration.
• Reference knowledge-base/regulations/ for TCPA, CAN-SPAM, Do-Not-Call, and Fair Housing rules; no outreach recommendation may violate them.
• Reference knowledge-base/best-practices/ for outreach norms in the agent's market.
• Reference knowledge-base/regulations/two-party-consent-states.md for the call-recording rule overlay (client-conversation-intelligence.md already maintains this list — use it as canonical).
• Confirm the contact list is the agent's own CRM or a list with an existing relationship — do not score a cold purchased list with a "past client" label.
• For Post-LA stage contacts: confirm the client-conversation-intelligence.md Pass 1 dictation is loaded; without it, the contact is downgraded to PLA stage scoring.

Process — Pass 1 (Fast Triage):

1. Normalize the data. Standardize home value, equity, and tenure across all contacts. Flag rows with missing fields; do not fabricate values. If equity is supplied as a range, use midpoint. If tenure is missing but purchase year is present, compute. If purchase year is missing, mark the contact INCOMPLETE and route to a "needs enrichment" bucket rather than scoring blind.

2. Apply stage-default component weights. Pass 1 uses stage-default assumptions where data is absent:

   Component	Database (D)	Pre-LA (PLA)	Post-LA (Post-LA)	Mid-Mkt (MM)	Post-Pending (PP)
   A — Equity (0–20)	weighted 1.0	0.8	0.6	0.4	0.6
   B — Tenure (0–20)	1.0	0.6	0.4	0.2	0.4
   C — Life-Stage (0–20)	1.0	1.0	1.0	0.6	0.8
   D — Behavioral (0–20)	1.0	1.0	0.8	1.2	1.4
   E — Market-Fit (0–20)	1.0	1.0	1.2	1.6	1.4

   The weights re-balance the score so each stage's predictive signals are weighted to where the agent actually has leverage. (Example: a Mid-Marketing seller's tenure barely matters; their behavioral signals — feedback responsiveness, willingness to reduce — are predictive.) Components below.

3. Compute the five score components per contact.

   A. Equity Score (0–20) — Higher equity means more optionality to sell. Band: <10% equity → 2; 10–25% → 6; 25–40% → 10; 40–60% → 14; 60–80% → 18; 80%+ → 20. Adjust −4 if the home is underwater or within 5% of purchase price in a flat market. Adjust +2 if a HELOC signal is present (indicates equity awareness). For high-equity seniors (60%+ equity AND age 65+ AND current homeowner of record), set a [FRAUD-SCREEN] flag and route any subsequent outreach through ai-fraud-defense-playbook.md's passphrase + senior-targeted-fraud overlay; the fraud risk to the contact rises with the public visibility of their equity.

   B. Tenure Score (0–20) — Median US tenure at sale is 13 years; local medians vary. Band: <3 years → 4 (sunk costs, low intent unless life event); 3–5 years → 10; 5–8 years → 14; 8–13 years → 18; 13+ years → 20. For move-up markets or starter-home zips, compress: the 5–8 year band becomes the peak. For Mid-Marketing stage, tenure becomes near-irrelevant (component weighted 0.2 — see step 2).

   C. Life-Stage Score (0–20) — Points for specific signals, capped at 20: youngest child leaving school zone +8; job change beyond reasonable commute +10; divorce filing +12 (handle with extreme sensitivity); inheritance/probate +12 (route through ai-fraud-defense-playbook.md's probate overlay); recent second child outgrowing home +8; retirement or age 65+ in larger home +6; death of spouse or partner +8 (handle with extreme sensitivity); multiple adjacent signals stack but cap at 20. Sensitive life events never receive opportunistic framing.

   D. Behavioral Score (0–20) — Points for observed actions, capped at 20: website activity on home-value or sold-listing pages last 30 days +8; opened 3+ agent emails last 60 days +4; clicked CMA or home-value link +10; requested home valuation +15; asked about contractor referrals or repair bids +6; liked or commented on sold-listing posts +3. For Mid-Marketing stage, the component shifts to listing-specific signals: showings/feedback responsiveness +6; price-reduction discussion engagement +8; willingness to refresh photos / restage +6 (component weighted 1.2 in MM stage). For Post-Pending re-engagement, behavioral signals include reason-for-fall-out responsiveness and willingness to relist after seasoning (weighted 1.4 in PP stage).

   E. Market-Fit Score (0–20) — Does the CURRENT market favor selling this specific home? Points: supply shortage in zip (MOI < 3) +12; price-per-sqft above purchase by 40%+ +6; days-on-market in zip < 21 +6; rate trend favorable (rates dropping > 50 bps last 90 days) +4; seller-concession percentage in zip < 2% +2. Cap at 20. Subtract 4 if MOI > 6 or DOM > 60. The MOI band is the canonical input from market-analysis-summary.md four-band ladder. For Mid-Marketing stage, this component rises to a 1.6 weight — the question becomes "given current market, can this listing reasonably sell at the current price?"

4. Sum to 0–100 with stage weights applied. The weighted sum keeps the 0–100 scale. Classify by stage-aware band (below).

5. Classify each contact by stage-aware band.

   D / PLA stage:

   • 80–100 Hot Seller — Reach out this week. Predicted listing window 0–90 days.
   • 60–79 Warm Seller — Reach out this month. Window 90–180 days.
   • 40–59 Likely 6–12 mo — Monthly nurture.
   • 20–39 Long-term nurture — Quarterly check-in.
   • <20 Cold — Database-only.

   Post-LA stage:

   • 80–100 Convert this week — Signed listing agreement is the next conversation.
   • 60–79 Convert this month — Address the gap (price expectation / decision-maker / home-prep) explicitly; second listing-appointment touch.
   • 40–59 30–60 day fix-and-circle-back — Specific home-prep or price-conversation gap; agent may need cma-presentation-generator.md re-run.
   • 20–39 6+ month re-warm — Sentiment was thin; treat as PLA in 60 days.
   • <20 Decline — Document and route to nurture.

   Mid-Marketing stage:

   • 80–100 Price-reduction conversation today — All signals point to a listing that needs the price-reduction call now.
   • 60–79 Price-reduction conversation this week — Schedule the call; prepare the data via market-analysis-summary.md MOI re-run.
   • 40–59 Marketing-tactic conversation — The listing has a marketing problem (photos / video / staging / open-house cadence) before it has a price problem; route through social-content-calendar-30day.md + listing-video-workflow.md + pre-launch-listing-audit.md.
   • 20–39 Status-check call — Neutral check-in; surface seller's confidence.
   • <20 Likely-to-withdraw — Soft conversation about expectations; do not push for price reduction.

   Post-Pending stage:

   • 80–100 Re-list this week — Seller is ready; gap is operational not motivational.
   • 60–79 Re-list this month — Address the fall-out reason; minor staging / photo / price calibration.
   • 40–59 Re-list next quarter — Seasoning needed; substantive issue (over-pricing pattern / repeat-falling-out pattern).
   • 20–39 Convert to long-term nurture — Seller is no longer in active selling mode.
   • <20 Off the seller pipeline — Document and remove from seller-pipeline scoring.

6. Generate a personalized outreach angle per top-band contact (per stage). For each contact in the top two bands of their stage, produce ONE sentence referencing a specific, verifiable fact about their home, their situation, or — for Post-LA / MM / PP stages — the specific gap from the listing-appointment Pass 1 dictation, the mid-marketing telemetry, or the fall-out reason. Examples:

   • D, top band: "Three houses on your block have sold this year above $1.1M, and the last one had 40% less yard than yours."
   • Post-LA, top band: "When we talked Tuesday you mentioned you wanted to net $700K — current Highland Park comps support a list at $895K with a $700K+ net even after concessions. Want me to walk through the numbers?"
   • MM, top band: "Three weeks on market and four showings — the feedback theme is 'great house, hard to see at $895K.' The 90042 sub-3-MOI band typically clears at list-to-sale 99–101%; we're at 0% over 21 days. Want to talk about a $30K reduction Friday?"
   • PP, top band: "I know April's escrow falling through was hard. The reason was the buyer's appraisal — not your house. Three more buyers in your range walked through last week's open. Want to relist Tuesday?"

   Do not fabricate comps, showings counts, or feedback themes. If the input does not contain the specific fact the angle requires, mark [NEEDS DATA] and request it rather than inventing.

7. Recommend the first touch. For each top-band contact, choose ONE channel and ONE message format consistent with the relationship source AND the stage:

   • D / PLA:
     • Past client → phone call, then handwritten note if no answer.
     • Sphere → in-person or social DM, not cold email.
     • Online lead (prior) → SMS during business hours or email.
     • Open-house lead → email with CMA offer; phone only if opted in.
   • Post-LA: the agent's preferred listing-appointment follow-up channel (typically phone + emailed CMA + delivered listing-presentation packet).
   • MM: phone call to the seller; brokerage-policy note if a reduction is the topic (some brokerages require a written addendum).
   • PP: phone call to the former-listing seller; framed around the fall-out reason or the expiration cause, not opportunistically.

   Every recommendation must be TCPA / CAN-SPAM / DNC compliant. Never recommend SMS for a contact who has not opted in. Never recommend a call to a DNC-listed number; suppress the recommendation and fall back to mail or written outreach.

8. Surface compliance and sensitivity flags. Before shipping, call out:

   • Any contact on the DNC list with phone recommended (suppress).
   • Any contact flagged sensitive (recent divorce, probate, death) — soften angle; relational, no-agenda framing only.
   • Any score that relies on data the agent did not supply (mark "score incomplete, enrich and rerun").
   • Any contact in a two-party-consent state where the agent's recommended channel would create a recording exposure (cross-link to client-conversation-intelligence.md's two-party-consent-state list).
   • Any high-equity-senior contact (the [FRAUD-SCREEN] flag from step 3-A) with an outreach recommendation that touches wire instructions or closing logistics.
   • Any contact where Fair Housing risk appears (e.g., targeting based on family composition framed commercially — reframe around the home, not the household).

9. Output the working artifacts. Three artifacts, every time:

   • Ranked working table — columns: Rank / Name / Address / Stage / Intent Score / Band / Predicted Window / Outreach Angle (one sentence) / Recommended First Touch / Flag.
   • Pipeline summary — how many in each band per stage, top 3 data gaps blocking better scoring, recommended 30-day action plan sized to the agent's stated capacity, total agent-capacity-fit count.
   • Re-score cadence — for the next 90 days: which contacts will be re-scored on what trigger (event-triggered: rate move, life-event detected, listing-appointment Pass 1 lands; calendar-triggered: 30-day MM re-score, quarterly D re-score, 60-day Post-LA re-score-or-decline).
   • The artifacts are saved to outputs/[YYYY-MM-DD]-seller-intent-pipeline.md if the user confirms.

Process — Pass 2 (Deep Analysis):

Pass 2 layers in the refinement inputs (#6–#11) and re-runs Steps 1–9 with the actual data instead of stage defaults. The Pass 2 score is canonical; Pass 1 is shipping-quality but Pass 2 is the version the agent works against for the quarter. For Post-LA stage contacts, Pass 2 cannot be run without the client-conversation-intelligence.md Pass 1 dictation; if missing, Pass 2 produces the same artifacts at PLA-stage scoring with a [POST-LA REQUIRES DICTATION] flag.

Critical rules:

• No fabricated equity, value, or comp numbers. If data is missing, mark the row, do not score it.
• Outreach recommendations must comply with TCPA, CAN-SPAM, Fair Housing, DNC, and the loaded state's two-party-consent rule.
• Sensitive life events (death, divorce, probate) never receive opportunistic framing. A human agent, not an automated sequence, initiates that contact.
• Scoring is a prioritization tool, not a prediction guarantee. Communicate uncertainty in the output.
• Never output the full list with commercial framing for contacts flagged DNC, in litigation with the agent, or asked for space.
• High-equity-senior contacts are routed through ai-fraud-defense-playbook.md's overlay before any outreach that touches wire instructions or closing logistics.
• Mid-Marketing scoring may not produce price-reduction recommendations to a third party (e.g., a co-listing agent or a referring broker) without the seller of record's input.
• Post-Pending scoring may not assume the seller wants to re-list; the framing is "ready to engage about re-listing," not "ready to re-list."
• The skill produces the prioritization layer; the actual outreach copy is buyer-follow-up-sequence.md (seller-nurture variant), the CMA is cma-presentation-generator.md, the listing-appointment dictation is client-conversation-intelligence.md, the listing repair is pre-launch-listing-audit.md, the marketing reset is social-content-calendar-30day.md + listing-video-workflow.md.

Example Output

Input summary (Pass 2): Agent: Sarah Chen, BHHS, Highland Park / Eagle Rock / South Pasadena. Quarterly batch + 1 listing-appointment Pass 1 ingest. Capacity: 25 new listing conversations / 30 days. Outreach: phone, SMS (opt-in only), email, handwritten note. Market context: Mar 2026 90042 SFR 3BR+ MOI 2.4 (sub-3 band, High confidence); list-to-sale 100.4%; DOM median 19. Two-party-consent state: CA. Stage-distribution input:

• 412 Database (D) contacts
• 8 Pre-Listing-Appointment (PLA) contacts
• 1 Post-Listing-Appointment (Post-LA) — the Hendersons of Maple St (Pass 1 dictation loaded from client-conversation-intelligence.md)
• 3 Mid-Marketing (MM) — three of Sarah's current listings
• 4 Post-Pending (PP) — three closed Q1, one expired

Pipeline summary (Pass 2):

• D contacts: 412 scored. Hot 7 / Warm 24 / Likely 6–12mo 68 / Long-term 183 / Cold 94 / Incomplete 36.
• PLA contacts: 8 scored. Convert-this-week-equivalent 2 / Convert-this-month-equivalent 4 / 30–60d 1 / Decline 1.
• Post-LA: Hendersons. Score 91 (Convert this week — listing agreement is the next conversation).
• MM contacts: 3. One at 84 (price-reduction conversation today). One at 56 (marketing-tactic conversation — route to pre-launch-listing-audit.md re-run + social-content-calendar-30day.md reset). One at 32 (status-check call only).
• PP contacts: 4. One at 78 (re-list this month — appraisal-driven fall-out; minor price calibration). One at 64 (re-list this month — pricing pattern; CMA re-run). One at 41 (re-list next quarter; seasoning + photo refresh). One at 12 (off seller pipeline).

Capacity fit: 25 listing conversations / 30 days. Recommended:

• Week 1: Hendersons (Convert this week); 2 PLA Convert-this-week-equivalent; top 5 D Hot. (8 conversations)
• Week 2: top 7 D Hot remaining + 4 PLA Convert-this-month-equivalent. (11)
• Week 3: 3 PP top-band re-list contacts + 1 MM price-reduction call + 2 D Warm follow-ups. (6)
• Week 4: capacity-saved buffer for inbounds; re-baseline at Day 30.

Top 5 ranked across all stages (excerpt):

Sensitivity + compliance flags:

• Contact #18 (D-stage) has a probate signal — pulled from automated outreach; routed to agent for personal note with no sales ask. [ai-fraud-defense-playbook] probate overlay applies.
• Contact #41 (D-stage) on DNC list — phone suppressed; fallback to mail.
• Contact #87 (D-stage) is a high-equity senior (78% equity, age 71) — [FRAUD-SCREEN] flag. Outreach permitted; passphrase protocol applies for any future closing-logistics conversation.
• Hendersons (Post-LA): two-party-consent CA — confirmed ahead of recorded follow-up call.
• MM contact at 32 (status-check only): seller previously expressed frustration; framing is relational.

Re-score cadence (next 90 days):

• Hendersons (Post-LA): re-score Day 14 — convert or move to PLA.
• 11 D Hot/Warm worked in Wk 1–2: re-score Day 30 (event-triggered if any moves to Post-LA).
• MM contact at 84: re-score Day 7 (post-reduction outcome).
• MM contact at 56: re-score Day 21 (post-marketing-reset outcome).
• PP contacts at 78 + 64: re-score Day 30 (post-relist or post-conversation).
• D book: full quarterly re-run Day 90.
• Trigger-based: any rate move > 50 bps in 90 days re-runs Market-Fit component for all stages.

Hand-offs:

• cma-presentation-generator.md — all top-band D / PLA contacts in capacity get a CMA prep; the Hendersons get the canonical CMA for their listing-conversion conversation.
• buyer-follow-up-sequence.md — seller-nurture variant cadence for Likely-6–12mo and Long-term-nurture bands.
• client-conversation-intelligence.md — every listing-appointment Pass 1 dictation flows back to re-score the Post-LA stage for that contact.
• market-analysis-summary.md — quarterly MOI band re-pull; monthly DOM band re-pull; trigger any rate-driven Market-Fit re-score.
• pre-launch-listing-audit.md — for MM contacts at 40–59 (marketing-tactic band).
• social-content-calendar-30day.md — for MM contacts at 40–59; supplies the Reel/Short cadence reset for the slow listing.
• listing-video-workflow.md — for MM contacts at 40–59; supplies the Just-Listed Refresh cut + the Still-Available Refresh cut.
• ai-fraud-defense-playbook.md — high-equity-senior overlay; probate overlay; any closing-logistics outreach.