List Stacking

What it means

Every prospecting list is an inference. A tax-delinquent pull says an owner has fallen behind with the county. An absentee pull says they don't live at the property. A vacancy flag says mail has stopped being collected. Each one on its own is a guess about motivation, and each one is wide — a mid-sized county can return tens of thousands of records on a single filter, and the overwhelming majority of those owners will never sell to anybody.

List stacking is the practice of combining those pulls and ranking owners by how many of them they appear on. An owner on one list is a maybe. An owner who is absentee and tax-delinquent and flagged vacant is three independent record systems agreeing that a property has become a burden to somebody who isn't there to deal with it. The technique exists because that intersection behaves nothing like the union it came from.

Signal count beats list size

The instinct when a campaign underperforms is to pull a bigger list. Stacking is the opposite instinct: pull several lists and then throw most of each one away. A 40,000-record county pull stacked against two other pulls can leave well under a thousand owners — and those are a different population, not a random sample of the 40,000.

The effect shows up in three places at once. Conversations get easier, because a stacked owner usually has a problem they will admit to. Follow-up gets cheaper, because a list of that size can be worked properly by one person on a real cadence, where 40,000 records can only be sprayed. And the data bill drops, which is the part most operators discover last and care about most.

How stacking is actually done

1. Pull each source on its own cadence. Recorded events like pre-foreclosure filings and notices of default change when the county records them. Status attributes like absentee ownership or modeled equity move slowly. Stacking a monthly event feed against a quarterly attribute pull is normal; stacking two snapshots taken a year apart is how stale records sneak into a list you believe is high-signal.
2. Normalize before you compare. The same property arrives as "123 N Main St", "123 North Main Street" and "123 N MAIN ST UNIT 2" depending on the source. Matching has to happen on a normalized address or parcel key, not on the string a human typed.
3. Count appearances, then set a threshold. Two-or-more is the usual working line. Three-or-more is a call list you can finish in an afternoon.
4. Exclude as deliberately as you include. Subtracting a list is as useful as adding one — the owners you already worked last quarter, the ZIPs outside your buy box, the property types you would never buy.
5. Trace the survivors, not the inputs. This is the step order that decides the economics, and it is the section below.

What stacking is not

Stacking compares the records inside the lists you have built. It is not universal identity resolution across every dataset in the country, and it will not by itself tell you that the "J Smith" on your vacancy pull and the "John Smith Family Trust" on your tax pull are the same human being — that is a different problem, and it belongs to relational mapping and the owner-identification work that sits upstream of any list.

It is also not a motivation score. Signal count is a proxy, and proxies drift: a property can be absentee, high-equity and flagged vacant because the owner is midway through a renovation and perfectly content. A stack tells you who is worth a conversation, not what the conversation will contain.

Why stacking is a cost decision before it is a marketing one

Skip tracing is priced per record, so the size of the list you trace is the budget. Trace three raw pulls of 40,000 each and you are paying for 120,000 lookups, nearly all of them for owners you will never call twice. Stack first and trace only what survives, and the same campaign costs a small fraction of that — with a better list at the end of it, not a worse one.

That ordering — stack, then trace — is the most common thing experienced teams do that beginners don't. It is also why keeping stacking and batch skip tracing in two separate tools is quietly expensive: every round trip between them is an export, a re-import, and another chance to pay for the same record twice.

How Ava Data handles list stacking

List Stacking is a built-in step in Ava Data, sitting between List Builder and the trace. You select the source lists you want to combine, set the minimum number of appearances a record must have to qualify, optionally exclude selected source lists, and save the result as its own list — which then behaves like any other list in your account. A duplicate-run warning flags when you are about to process a file or list you have already run, which is the most common way credits get spent twice on one record.

The workflow that follows is the whole reason for having both halves in one place:

• Build the inputs. Pull the quick lists for your market — tax-delinquent, vacant, out-of-state owner, tired landlord and the rest of the quick-list set — by state, county, city or ZIP.
• Stack them. Set the appearance threshold to two or three, exclude anything you have already worked, and save the overlap.
• Trace only the stack. 1 credit (2¢) per phone match, 2 credits (4¢) for phone plus email, 10 credits (20¢) for a Deep Search that also returns related persons. You are charged only when a search returns a match, so a stacked list of 900 records costs at most 900 phone credits — about $18 — against a ceiling of roughly $2,400 to trace the three 40,000-record pulls it came from.
• Export what you need. Up to 35 selected columns, with your column preferences saved for the next run, or push the owners you reach into Follow Up Boss.

The identity graph behind the trace refreshes every 24 hours, so a stack you built last month and re-trace today returns the current number rather than the one that disconnected in between. The $9/month plan includes 100 credits — enough to stack a county, trace the top of the result, and find out whether the signal-count argument holds in your own market before committing a full list to it.

Stacking compares the records held in your Ava Data lists; it is not a claim to resolve every semantically similar record, and the duplicate-run warning is a warning rather than a guarantee.

For developers: example API call

If you'd rather click than code, the Ava Data dashboard returns the same data without a single line of JSON. The snippet below is for teams wiring Ava Data into a CRM, dialer, or AI pipeline.

Content-Type: multipart/form-data

file=@stacked-3-signal.csv
dataTypes=["phone","email"]

{
  "success": true,
  "data": {
    "jobId": "job_4b17ce",
    "recordCount": 912,
    "dataTypes": ["phone", "email"],
    "creditsPerMatch": 2,
    "status": "processing"
  }
}

Related terms

• Motivated Seller Leads — Motivated seller leads are property owners with a situational reason to sell quickly or flexibly — pre-foreclosure, inheritance, tax delinquency, tired landlords — and generating them means combining the right list with contact data that actually reaches the owner.
• Distressed Property Leads — Distressed property leads are owners of homes facing financial or physical distress — pre-foreclosure, tax-delinquent, code-violation, vacant, or inherited — and are the highest-converting segment for wholesalers and cash buyers.
• Batch Skip Tracing — Batch skip tracing — also called bulk skip tracing — runs a skip trace over a whole list of records at once, typically a CSV of names and addresses, and returns matched contact information as a downloadable file or through an async API.

Frequently asked questions

What is list stacking in real estate?

List stacking is combining several property-owner lists into one and keeping the owners that appear on more than one of them. Instead of working a single wide list, you rank owners by how many independent signals they carry — absentee plus tax-delinquent plus vacant, for example — and work the overlap first.

How many lists should I stack?

Three to five inputs is where most operators land. Two lists often leave an overlap still too wide to work properly; past about five you are usually adding sources that correlate with ones you already have — absentee and out-of-state owner overlap heavily by construction — so the extra signal is smaller than it looks.

Should I skip trace before or after stacking?

After, essentially always. Skip tracing is priced per record, so tracing the raw inputs means paying for every owner you are about to discard. Stacking first cuts the traced volume by an order of magnitude on a typical county pull. With Ava Data charging only on a match, a stacked list of around 900 records costs at most 900 phone credits rather than the tens of thousands of lookups the raw pulls would have needed.

Does list stacking work if my lists come from different providers?

Yes, provided they can be normalized to a common address or parcel key first — that normalization is what makes the comparison meaningful. Ava Data's List Stacking combines the lists held in your account, so a list pulled somewhere else has to be brought in before it can be stacked against a List Builder pull.

Is a stacked list always better than a single list?

It is a better use of a fixed budget, which is not quite the same claim. Signal count is a proxy for motivation, and proxies drift — a property can be absentee, high-equity and vacant because the owner is renovating it. What stacking reliably gives you is a shorter list you can actually work on a real follow-up cadence, at a fraction of the data cost of the pulls it came from.