DataSift

Data Strategy

Data Priority Pyramid

Everyone has the same lists. Almost nobody pulls them in order.

14 min read Companion tool: Doors Per Deal Framework
The framework

Every list, ranked against the same baseline.

The question is never where to get data. It is which pull finds a deal for the fewest doors worked. Doors per deal answers it: list size divided by the deals that list actually produced. Lower is better, and the spread from the sharpest signal to the list everyone starts with runs more than fifty to one.

317Doors per deal on high equity, a Priority 3 list
vs
6.2Doors per deal on a final judgment, Priority 1

That is the whole idea behind the priority ranking. Every list gets divided by the deals it produced. That number is measured against the county baseline: what marketing to the whole county costs you per deal. Three times that baseline or better is Priority 1. One and a half to three times is Priority 2. Anything under that is Priority 3, which means coverage, not efficiency.

You do not earn your way to the good data. You start at Priority 1 on day one, then add layers. An operator ground a whole market unscored at 200-plus doors per deal. The same market at AI score 90-plus: about 22 doors, $125,000 median gross, 56% margins. Same market, same phones. The only difference was which rows got pulled first.

Priority 1, 2, and 3. One work order.

This is the same ranking the Doors Per Deal Framework runs on your county, held as a pyramid. Priority 1 sits at the top because you pull it first. Priority 3 sits at the base because it is where the doors are, not where the deals are cheap.

Native lists and stacks. Every number is a median across the counties where that signal held a measurement, out of 3,144 counties of recorded investor purchases.

PRIORITY 3 Coverage and context PRIORITY 2 Second wave PRIORITY 1 Pull first width = how many doors sit in the tier

Click a tier. Work it top down. The apex is the smallest list and the cheapest deal. The base is the whole market, at the steepest cost per new deal.

Doors per deal · lists
Priority 16-56
Priority 239-112
Priority 3158-317

Lower is better. Median doors worked per closed investor purchase.

Doors per deal · AI score
Priority 1: score 90-10011-23
Priority 2: score 80-9039-44
Priority 3: score 50+64-78

Measured in Knox and Blount, TN, the reference market with a full off-market score.

Priority 1: pull first

Three times the county baseline, or better

Thirty-six list signals clear the bar: 11 native lists and 25 stacks. The court-driven distress lists sit at the top of it, at 6 to 20 doors per deal. A typical county baseline runs 100 to 300. A signal at 10x or better ranks here at any list size.

36Signals at Priority 1
6.2Best doors per deal
22.7xBest lift vs baseline

The Priority 1 lists

  • Notice of foreclosure, 6.8 doors per deal, 20.9x baseline (non-judicial states)
  • Final judgment, 6.2 doors per deal, 20.6x (judicial states)
  • Notice of default, 8.9 doors per deal, 20.2x (non-judicial states)
  • Probate, 16.4 doors per deal, 9.1x, measured in 38 counties
  • Lis pendens, 20.1 doors per deal, 8.7x (judicial states)
  • Judgment lien 14.3 doors at 8.2x, estate sale 32.9 at 5.2x
  • Vacant 44.7 doors at 3.7x, and it pulls on every plan
  • Other lien 44.7 at 3.4x, pre-probate 49.8 at 3.1x, out-of-state 56.2 at 3.0x
  • Sharpest stack: absentee plus notice of foreclosure, 6.4 doors per deal, 22.7x

Your state decides which court signal is real. Non-judicial states run notice of default and notice of foreclosure. Judicial states run lis pendens and final judgment, and the other pair drops to Priority 3 there. Access: every row except vacant needs SiftMap Pro data, included on the $499/mo Expert plan or a $297/mo add-on on the $149/mo Professional plan.

Priority 1: pull first

AI Investor Score 90 to 100

The top band is a Priority 1 pull by doctrine, and the measurement backs it. Knox County runs 22.9 doors per deal at 9.0x its baseline. Blount runs 10.8 doors at 24.8x. The band reaches about 27% of a county's investor sales, so it is a sharp pull, not a coverage pull.

11-23Doors per deal
9-25xLift vs baseline
27%Deal share reached

What the band gives you

  • Properties scored before they become an obvious, marketed list
  • Median reach of 27.0% of investor sales at 90 and above, across 37 measured counties
  • $125,000 median gross and 56% margins on the operator run that leaned on it
  • One county of off-market scoring included with the plan
  • Runs as its own pull, in parallel with the distress lists, never AND-stacked on top of them

The score is measured in Knox and Blount, TN, the one market with full off-market scored supply. Treat the two counties as a range, not a national median. Access: the $1,250/mo AI plan.

Priority 2: second wave

One and a half to three times baseline

Eight native lists. Real signals, not sharp ones. They run three to twenty times bigger than the Priority 1 lists. That size is exactly why they go second: the volume waits there until your first pull is worked out.

8Lists at Priority 2
39-112Doors per deal
1.6-4.8xLift vs baseline

The Priority 2 lists

  • Zombie, 38.7 doors per deal, 4.8x, but measured in only 16 counties
  • HOA lien, 59.3 doors per deal, 2.6x
  • Bankruptcy, 61.8 doors per deal, 2.5x
  • Tax delinquent, 69.5 doors per deal, 2.2x, measured in 427 counties
  • Bad credit, 78.4 doors per deal, 1.8x
  • Low income, 94.5 doors per deal, 1.6x
  • Tired landlord, 103.8 doors per deal, 1.7x, measured in 1,097 counties
  • Absentee, 112.3 doors per deal, 1.7x, measured in 1,966 counties and on every plan

Lift alone does not promote a row. Zombie carries a Priority 1 lift and still sits here, because 16 measured counties is a thin sample. Priority 1 also demands volume: 2% of county deal flow, or 25 deals.

Priority 2: second wave

AI Investor Score 80 to 90

The second band by doctrine. Knox runs 38.7 doors per deal at 5.3x baseline, Blount runs 43.5 at 6.2x. Cumulative reach at 80 and above is 39.5% of a county's investor sales. Widening the band buys about 12 points of coverage and costs you roughly half the sharpness.

39-44Doors per deal
5-6xLift vs baseline
39.5%Deal share at 80+

How to read the band

  • Still 5x or better against baseline, which would be Priority 1 on the list ladder
  • Ranked second because the 90 to 100 band is strictly cheaper per deal
  • The score is cumulative, so pull 80 to 90 as its own preset once 90 and above is dialed out
  • Best use: your first pull is worked through and you need next week's volume
  • Widen the band. Do not stack a distress list on top of it

Coverage numbers are medians across the 37 counties with a measured off-market score. Access: the $1,250/mo AI plan.

Priority 3: coverage and context

Under one and a half times baseline

Three lists, and they are the three most people pull first. High equity, senior, and free and clear are the biggest lists in any county and the slowest per deal. Senior and high equity run below the county baseline, which means you would find deals faster marketing to the entire county.

3Lists at Priority 3
158-317Doors per deal
0.6-1.3xLift vs baseline

What lands here

  • Free and clear, 157.5 doors per deal, 1.3x, measured in 2,155 counties
  • Senior, 263.3 doors per deal, 0.7x, measured in 1,520 counties
  • High equity, 317.1 doors per deal, 0.6x, measured in 1,387 counties
  • Judicial notice lists pulled in a non-judicial state, and the reverse: right list, wrong state
  • Thin samples, sliver-volume rows, and churn artifacts where doors per deal falls under 1

None of these are dead. They are filters, not lists. High equity plus notice of default is 9.4 doors per deal at 19.8x. Free and clear plus vacant is 39.5 doors at 4.3x. Stack them onto a distress trigger and they turn into Priority 1 rows.

Priority 3: coverage and context

AI Investor Score 50 and above

The coverage layer. Knox runs 63.8 doors per deal at 3.2x baseline and reaches 61.1% of its investor sales. Blount runs 78.1 doors at 3.4x and reaches 47.5%. Across the 37 measured counties, the median reach at 50 and above is 67.7%.

64-78Doors per deal
3.2-3.4xLift vs baseline
67.7%Median deal share

How to read the band

  • On lift alone it clears the Priority 1 bar. It sits here on purpose
  • The job of this band is reach, so judge it on the 67.7%, not the 3.2x
  • Driving for dollars layered on the union adds roughly 8 to 13 coverage points
  • Blended across the whole union you are working 100 to 180 doors per deal
  • Run it when the sharper bands are worked through and you want the whole market

Precision leads, coverage scales. Drop to 50 when your Priority 1 pull is producing and your callers have capacity, not before. Access: the $1,250/mo AI plan.

The stacking rule. Run the AI score and the distress lists as parallel pulls. Never AND-stack a distress list on top of the score. The score already prices in distress, so stacking shrinks the list without making it sharper. You end up near 48 doors per deal instead of 31. Stack distress signals on each other, and run the score band as its own separate preset.

And the ranking does not stop at the pull. When a category proves out here, that is your signal to go get it at the county directly, weeks before the provider version exists. That thought process runs through the rest of this page.

Two dials, never confuse them. Saturation sets the touch: first-to-market lists reach easily on cheap multi-touch, saturated lists need a higher-caliber caller and a premium touch. Sales-cycle speed sets the urgency: foreclosure runs a 45-day auction clock, probate and courthouse pipelines are long nurtures.

5-DAY DEAL FLOW CHALLENGE

Next live cohort: Monday, September 21 to September 25

5 days live with Ty, free. Built for investors who have already closed at least one deal and want a repeatable pipeline instead of a lucky one. Save your seat in the next cohort. Already in? Use this guide as your between-session refresher.

Every signal, one ranking

The whole ladder, in pull order

Fifty pull signals: 22 native lists, 25 stacks, and the three AI score bands, each judged against a county baseline of every single-family investor purchase. Filter it the same way the county tool does. Save your Priority 1 rows as their own presets in SiftMap, then work them top to bottom.

Signal Doors per deal Lift vs baseline Counties measured Priority Plan
Priority 1 pull first: 3x the county baseline or better, and 10x or better ranks here at any list size Priority 2 strong second wave: 1.5x to 3x baseline Priority 3 coverage and context: under 1.5x, regime-demoted, or too thin to trust yet All plans any tier  Expert plan SiftMap Pro data, included on Expert or a $297/mo add-on  AI plan Investor AI Score

Read the counties column before you trust a row. Notice of foreclosure carries the second-highest lift on the board off 28 measured counties. Absentee carries a 1.7x off 1,966. The first is a sharp signal you should verify in your own market; the second is a settled fact about a list everybody already has. Both belong in the plan, in that order.

Every number here is a median across the counties where that signal held a measurement. The basis is single-family, off-market, sold-to-investor deals over a six month window. Run your own county through the Doors Per Deal Framework to replace these medians with your local ranking.

From the ranking to the courthouse

Proved it in SiftMap? Go pull it at the source.

This is the move, and it is how to think about most of what is on this page. The ranking is the scouting step. It tells you which categories actually produce deals in your county. Once a category proves out, you stop renting it from a provider.

Every Priority 1 row above started life as a public record that publishes on a clock. The provider bought it, cleaned it, and sold it to you and everyone else. Pull the same record at the county and you are working it 30 to 90 days before it shows up in anyone's paid data. Same signal, proven by the same ranking, minus the queue.

It runs the other way too. When SiftMap coverage is low or zero for a category in your county, the county pull is not a backup. It is the only way in. The Doors Per Deal Framework flags those gaps county by county, and a list nobody can buy is a moat, not a dead end.

Pull first Priority A 4 sources
Foreclosure notices

Proves out in SiftMap as notice of foreclosure at 6.8 doors per deal and final judgment at 6.2. Also notice of default at 8.9 and lis pendens at 20.1. Four of the five sharpest rows on the whole ladder.

WindowDays 1-30 from the notice filing, before the auction CadencePull daily to weekly (fast in non-judicial states) OfficeCounty recorder in non-judicial states, court docket in judicial states

Your state decides which of the four is the real signal. Pull the pair your regime supports, and skip the other two.

Probate filings

Proves out as probate at 16.4 doors per deal, estate sale at 32.9, and pre-probate at 49.8. The free and clear plus probate stack runs 10.8 doors at 12.0x baseline.

Window30 to 180 days from filing CadencePull daily to weekly from the court OfficeProbate court or clerk of court

Contact the personal representative, never the deceased. This is the row where deep prospecting on the heir pays for itself.

Tax delinquent roll

Proves out as tax delinquent at 69.5 doors per deal, a Priority 2 row on its own. Stacked on vacant it is 54.9 doors and 4.3x, which lands it in Priority 1.

WindowOngoing: bills owed, no auction date yet CadenceMonthly rolling list OfficeCounty treasurer or tax collector

The roll is where the stack comes from. Pull it monthly, then cross it against your vacancy flags before you dial.

Tax sale list

No SiftMap row of its own. The closest proxy is the tax delinquent roll these properties graduate out of. At the sale stage, the source pull is the only way to work them.

Window60 to 90 days before the auction, once the list publishes CadenceAnnual list, published as legal notices 3 to 4 weeks pre-sale OfficeTreasurer, plus the legal notices in the paper of record
Pull next Priority B 2 sources
Eviction filings

Proves out as tired landlord at 103.8 doors per deal across 1,097 counties. The provider list infers the landlord is tired. The filing is the landlord telling the court.

WindowImmediate: the landlord, as plaintiff, is motivated right now CadencePull weekly from the court docket OfficeDistrict or magistrate court docket

Two or more filings in 12 months is your own tired-landlord definition, built from the trigger instead of a proxy.

Code violations

Proves out as vacant at 44.7 doors per deal and 3.7x, and other lien at 44.7 and 3.4x. Both are Priority 1, and both are downstream of a violation someone filed first.

WindowBefore the compliance deadline, typically 10 to 30 days CadenceWeekly to monthly OfficeUsually city-level, not county, and often a records request

This is the one that breaks the county pattern. Budget for a records request and a slower turnaround.

Pull last Priority C 1 source
Divorce filings

No SiftMap row, and no provider sells one. The ladder cannot rank what it cannot see, which is exactly why the competition here is the lowest of the seven.

Window60 to 180 days from filing, while the property sits on the marital asset schedule CadencePull weekly OfficeCircuit or family court docket

Pull it last because it is slow to build, not because it is weak. Nobody else in your county is working it.

The Doors Per Deal Framework carries the real offices, portals, phone numbers, and record-request paths for your specific county, grouped in this same pull order. Run your county, read your Priority 1 rows, then work down this list to find where each one is actually filed.

The Targeting Matrix

Priority tells you what to pull. This tells you how to touch it. The matrix plots freshness against competition: the fresher and less contested the pull, the cheaper the touch that works. Two Priority 1 rows can need completely different callers, and saturated never means dead.

The Targeting Matrix DATA FRESHNESS Fresh Stale COMPETITION LEVEL Low High FIRST MOVER Fresh data + low saturation SiftMap precision pull, obituary, same-day filings Lowest saturation | easy reach, cheap touch works This is where the money is. HIDDEN GEM Fresh data + high saturation Foreclosure: already heavily marketed Needs a higher-caliber caller + premium touch Winnable with a better caller. CROWDED RACE Stale data + low saturation Old lists nobody wants for a reason High volume, cold trigger, narrow with doors/deal Low ROI. Narrow it or skip it. DEAD MONEY Stale data + high saturation Recycled lists from data brokers Highest saturation | 20-50+ investors, least reachable Most investors live here.

First Mover: Your highest-ROI zone

Fresh, low-saturation records with almost no prior marketing. The SiftMap AI Investor Score and the native obituary flag live here: you reach the owner first, willingness is high, and cheaper multi-touch works. This is where every operator should start.

Action: Pull the AI Investor Score at 90 and above and the obituary flag, skip trace, and dial the same day. Add same-day courthouse pulls for consistency once you are producing.

Hidden Gem: Winnable with a better caller

Fresh data, heavily marketed already: foreclosure, where the lis pendens fired and everyone hit the owner. High saturation, fast clock. Your edge is a higher-caliber caller and a premium touch.

Action: Put your best closer on it, lead with a premium touch (door knock, handwritten mail), and move fast, because these are bottom-of-funnel deals on a 45-day clock.

Crowded Race: Low ROI territory

Old data few others want either. Low saturation does not save a stale record with a cold trigger. Only worth it if you narrow a high-volume list (senior, vacant, tired landlord) by doors per deal.

Action: Narrow with the AI Investor Score before you dial, or add a fresh trigger (a new code violation on an old tax-delinquent property). Otherwise spend your time on fresher data.

Dead Money: Where most investors are stuck

Recycled broker lists. Stale records, and 20 to 50 other investors calling the same people. This is the most saturated, least reachable data there is. If this is your primary source, the pyramid is your escape route.

Action: Stop buying pre-built lists. Pull the SiftMap precision score and start first, in the First Mover quadrant.

Click a quadrant above to see your action plan. Most investors are in Dead Money and think they are in First Mover.

Priority 1 in Action: Kasim Delake's Courthouse Engine

A W-2 employee in Delaware built a six-figure wholesaling operation on foreclosure and probate filings pulled straight from the county. Free records, skip tracing, sequential marketing. He never worked anything but Priority 1 rows, and he did it before there was a ranking that named them.

The Setup

His ad account died during COVID; a mentor showed him the courthouse method. Market: Delaware plus MD, VA, PA, NJ satellites. Data: county foreclosure and probate filings. Cost: zero. The Phase 2: The Delegator path, run deep.

The System

Every week, Kasim pulls fresh records from the courthouse (online or in person). Here is his exact flow:

County Records Google Sheet Upload to DataSift Skip Trace Niche Sequential (3-day call/text/mail)

Day 1-3: Call, text, and mail every new record. Cannot reach them? Second skip trace with new numbers. Still nothing? Try relative phone numbers. Last resort: direct mail only for ongoing touches.

The Result

"All of the county data and marketing that way is the life."

$100K+ wholesaling business built on two Priority 1 signals: foreclosure filings at 6.8 doors per deal and probate at 16.4. Kasim now runs Facebook ads and PPC alongside his courthouse pulls, but the courthouse engine through DataSift remains the backbone.

Kasim Delake case study interview
Kasim Delake: $100K+ From Free County Data

Kasim proves that two Priority 1 rows, worked to the bottom, build a real business. Today I would hand him the stacks too. Absentee plus notice of foreclosure runs 6.4 doors per deal. Free and clear plus probate runs 10.8. Same records, tighter pulls. The pyramid is a work order, not a ladder you have to climb.

How far down the pyramid your phase should go

Everybody starts at Priority 1. What changes by phase is how far down you are staffed to work, and whether the AI band belongs in your stack yet.

Phase 1: The Operator

Priority 1 only, lists view

Stand up the flow: one county, one list, sequential marketing live. Take the top three or four Priority 1 rows your state supports, skip trace them, and dial the same day. In a non-judicial state that is notice of default and notice of foreclosure at 7 to 9 doors per deal. In a judicial state it is lis pendens and final judgment. Probate at 16.4 doors works everywhere.

Budget: $499/month (Expert plan: SiftMap Pro plus unlimited skip trace), or from about $297/month as a SiftMap Pro add-on on the $149 Professional plan.

First hire: Data Manager at $500-700/month once you close your first deal.

Rule: One county, the sharpest rows, worked to the bottom. You do not need an AI plan to start.

Phase 2: The Delegator

Priority 1 in full, then Priority 2

Multiply touches: 4 full attempts on every record, mail to non-responders, data and admin off your plate. Work the whole Priority 1 block, stacks included, before you widen. Free and clear plus probate runs 10.8 doors per deal, out-of-state plus vacant runs 25.2. Those are stacks, not lists, and they are the cheapest volume you have left.

Budget: $500-2,000/month

Team: Data Manager + 1-2 callers

Rule: Stack distress signals on each other. If you add the AI band, run it as a parallel pull, never AND-stacked onto a distress list.

Phase 3: The Manager

All three priorities, both views

Scale the team on overflow: handoff system first, then admin, prospectors, lead manager, closer. You are the first phase with the caller capacity to justify Priority 3. Run the AI score at 50 and above for the 67.7% reach. Treat high equity and free and clear as filters on a distress trigger, not lists you dial straight.

Budget: $3,000-10,000+/month

Team: Full stack: Data Manager, Callers, Lead Manager, Closer, Sales Manager

Rule: Never skip a rung. Precision leads, coverage scales. Lead with Priority 3 instead of Priority 1 and your doors per deal and your margins both collapse.

Phase 4: The Owner

Priority 1 only, worked to the bone

Own the machine: deep-prospect every unreached door and tie in the obituary layer. That is a notice-of-death flag on the owner, added weekly nationwide, and about as uncontested as data gets. Estate sale at 32.9 doors per deal and pre-probate at 49.8 are the rows that reward this depth. Know every property by name.

Budget: $200-1,000/month on top of your running stack

Team: Data Manager (niche-trained)

Rule: Never go wide. Your margin comes from knowing more about fewer properties than anyone else, and from deep-prospecting the heir on the sensitive rows.

Starting at the base versus starting at the apex

Most investors start on the three lists that rank Priority 3, then blame the phones. The gap between 317 doors per deal and 6 is not talent or tonality. It is the order the rows got pulled.

Starts at Priority 3
Starts at Priority 1
First pull
High equity, senior, free and clear
The court signals their state supports
Doors per deal
158 to 317
6 to 20
Lift vs county baseline
0.6 to 1.3x, at or below it
8 to 21x
Who else is calling
20-50+ investors
1-3 investors
Data freshness
30-90 days old
0-7 days from filing
How they use the AI score
Stacked on top of a distress list
Run as its own parallel pull