How to Win Expired Listings With AI in 2026: The Agent’s Playbook

Updated October 2026.

An expired listing is the warmest cold lead in real estate: a homeowner who already decided to sell, already signed with an agent, and is now frustrated that it didn’t work. The intent is proven — only the strategy failed. This playbook shows you exactly how to win expired listings with AI in 2026: how to find them, how to prioritize which expireds are worth your time, how to personalize outreach at a scale no human can match, and how to be the first credible voice a frustrated seller hears — while staying on the right side of the calling and emailing rules.

Why expired listings are still gold in 2026

Most lead sources make you guess about intent. Expireds remove the guesswork — the seller raised their hand months ago. What’s changed in 2026 is the competition: the moment a listing expires, dozens of agents pull the same MLS data and start dialing. The winner is rarely the agent with the best script. It’s the agent who reaches the seller first with a message that actually addresses why the home didn’t sell — price, condition, photos, or marketing — and who keeps showing up after everyone else has given up.

That diagnosis-first, follow-up-forever approach is exactly where AI gives solo agents an edge that used to belong only to big teams with assistants. It will not place the call for you or replace your judgment, but it removes the two bottlenecks that cap most agents: how fast you can research and personalize, and how consistently you can follow up.

Where most agents lose the expired game

Three failure modes show up over and over. First, no prioritization: agents work expireds top-to-bottom by date instead of by winnability, wasting hours on overpriced, stubborn sellers while ignoring the motivated ones. Second, generic outreach: “I noticed your home didn’t sell” postcards read identically from every agent, so none of them stand out. Third, no persistence: most expired sellers relist within 90 days, but most agents stop following up after one or two attempts, so the seller who relists in week nine never hears from them again.

AI helps with all three — it scores, it personalizes, and it never forgets to follow up — but only if you feed it the right list and keep a human voice on the important touches.

Where to find expired listings (data sources)

Before any tool can score or personalize, you need the raw list. Expireds are easier to source than most off-market signals because the status is recorded in the MLS, but the phone numbers and mailing addresses usually are not — that is the gap the data services fill.

Source	What it gives you	Notes
Your MLS (Expired / Withdrawn / Cancelled status)	The listing itself: original and final price, days on market, price changes, photos, agent remarks	The primary, most authoritative source. Save a daily “expired/withdrawn/cancelled in the last 24h” search. Check your MLS rules before contacting — some restrict soliciting expireds.
Expired-lead services (REDX, Vulcan7, Landvoice)	Daily expired lists already matched to phone numbers and emails (skip-tracing), often with DNC flags	These exist because the MLS rarely exposes seller contact info. Compare coverage in your market and confirm each service scrubs against the federal DNC registry.
County recorder / assessor	Owner of record, mailing address, length of ownership, last sale price	Free and public. Best for direct mail and for estimating equity when you can’t get a phone number.
Public tax records	Assessed value, exemptions (owner-occupied vs. investor), delinquency flags	Helps you gauge equity and whether the owner lives there — both change how you open.
Skip-tracing / people-search tools	Current phone and email when the service list is thin	Verify before dialing; numbers go stale. Always cross-check against DNC before calling.

Whatever the source, the output is the same: a name, a way to reach them, and enough of the listing history to diagnose what went wrong. That is all AI needs to help you do the rest.

What makes an expired listing “winnable”

Not every expired is worth a letter, let alone a call. Three markers separate the ones that convert from the ones that waste your week:

• Healthy equity. A seller with real equity can afford to transact — drop the price, pay closing costs, or move. A seller who is underwater usually can’t, no matter how motivated. Ownership length and last sale price from public records are good proxies.
• A fixable reason it failed. If the home was priced within a few percent of comps and still didn’t sell, the gap was usually photos, condition, or marketing — all fixable, which gives you a credible pitch. A listing that was priced 20% over the market needs a hard pricing conversation, not a marketing one.
• Recency. An expired from the last 30 days is far warmer than one from six months ago, who has likely already relisted or given up. Work the freshest expireds first.

AI scoring bakes all three into a single ranking so you spend your hours where the odds are best instead of dialing the list top to bottom.

How AI helps you win expireds

Modern AI tools bring three capabilities to the expired game:

• Likelihood-to-relist scoring — predictive models weigh equity position, ownership length, local absorption rates, and pricing history to rank which expired sellers are most likely to relist soon, and at a realistic price. Instead of guessing, you work a ranked list.
• Research and personalization at speed — AI can summarize a listing’s history (original vs. final price, days on market, price cuts, photo and description quality) and draft a first touch that names the actual problem, in seconds rather than the ten minutes it would take you per lead.
• Persistent multi-touch sequencing — a 90-day cadence of value-first touches that runs automatically, so the seller who relists in week nine still hears from you in week nine. You escalate to a live call when engagement signals fire.

The same kind of engine that powers this workflow can also find sellers before they ever list — see our guide on how to use AI to predict when homeowners will sell to run both plays from one pipeline.

The AI expired-listing workflow, step by step

1. Pull today’s expireds and score them. Feed the fresh list into your predictive tool and sort by likelihood to relist. Skip the bottom half entirely; work the top 20% hard.
2. Diagnose before you dial. Have AI summarize the listing history — original price vs. final price, days on market, photo quality, description quality. Your opener should name the real problem, not recite a script.
3. Scrub against DNC. Before any call, run the numbers against the federal Do Not Call registry and your own suppression list. Mail and (compliant) email are always available; calls are not.
4. Send a personalized first touch within hours. A short message that shows you did homework (“Your home was priced within 2% of comps — the photos and marketing look like the gap”) beats “I have buyers in your area” every time.
5. Offer value, not a pitch. An updated equity and market analysis gives a frustrated seller a reason to talk to you. AI generates the first draft in minutes; you sanity-check the numbers.
6. Let the sequence run. Schedule the follow-up cadence and let automation carry it. Escalate to a call when engagement signals fire (opens, clicks, replies) — and only to numbers you’ve cleared to call.

Related: FSBOs are the other warm listing-side source — see how to win FSBO listings with AI.

Outreach scripts that actually get replies

Lead with the diagnosis, not the ask. These are starting points — personalize each one with the specific detail AI pulled from the listing, and keep an easy opt-out on anything digital.

1. Mailed letter or handwritten note

Hi [First name] — I saw your home on [Street] came off the market last week. I pulled the listing history, and from the outside it looks like the price was close to where comps landed; the photos and online marketing are where I’d focus next time. I put together a short, no-obligation breakdown of what I’d change and what it’s likely worth today. If it’s useful, I’m happy to drop it by or mail it — just let me know. Either way, I hope the next round goes smoothly. — [Name], [Brokerage], [Phone]

2. Voicemail (to a DNC-cleared number)

Hi [First name], this is [Name] with [Brokerage]. I specialize in homes that didn’t sell the first time, and I noticed yours on [Street] just expired. I already looked at the price history and the marketing — I think I can see what the gap was, and I’d rather show you than pitch you. If you want a quick, honest second opinion with no pressure, call or text me at [Phone]. Thanks, and good luck either way.

3. Email (CAN-SPAM compliant)

Subject: The gap on [Street] — and what I’d change

Hi [First name], your listing on [Street] came off the market recently. I reviewed the price history and the online presentation; the pricing looked reasonable, so I’d focus on photos, staging, and marketing reach next time. I’d be glad to send a short updated market analysis — no obligation, no pressure. Reply “yes” and I’ll put it together.
— [Name], [Brokerage], [Phone], [Physical mailing address]
If you’d rather not hear from me, reply “stop” and I’ll remove you immediately.

Compliance: DNC, TCPA, CAN-SPAM, MLS and fair housing

AI drafts and schedules the messages, but the legal obligations are yours. None of this is legal advice — confirm the specifics with your broker and counsel — but these are the rules that most often trip up expired outreach:

• Federal Do Not Call (DNC). A homeowner you’ve never done business with is a cold contact; the “established business relationship” exemption generally does not apply. Scrub every number against the national registry (and your internal list) before calling, and keep records.
• TCPA. Autodialers and prerecorded/artificial-voice messages to cell phones generally require prior express consent, which you don’t have from a cold expired. “Ringless voicemail” has been treated as a call under the TCPA in a number of cases — treat it as regulated, not a loophole. Manual dialing of a DNC-cleared number is the safer path.
• State mini-TCPA laws. Several states (for example Florida’s FTSA, plus Oklahoma and Washington) add stricter consent and calling-window rules than federal law. Check the rules in the state where the seller is located.
• CAN-SPAM. Commercial email needs a truthful subject line, a valid physical postal address, and a working opt-out that you honor promptly. Build these into every AI-drafted template.
• MLS and NAR rules. Some MLSs restrict or prohibit soliciting expired listings — check yours. Under NAR’s Code of Ethics (Article 16), you may not solicit a seller who is under an exclusive listing agreement; once the listing has genuinely expired or been cancelled that generally no longer applies, but confirm there’s no new agreement in place before you reach out.
• Fair housing. Target and word your outreach around the property and the pricing/marketing problem — never around the owner’s race, religion, national origin, familial status, disability, or any other protected class.

What to look for in an expired-listing tool

There’s no single required platform, and the right choice depends on your market and how you already work. Rather than chase a brand, judge any tool against what the expired game actually demands:

• Fresh, accurate data. Does it pull expireds daily in your market, and how good are the matched phone numbers and emails? Coverage varies a lot by region — test with your own ZIP codes.
• Built-in DNC scrubbing. The best lead services flag or remove DNC numbers automatically. Confirm this before you rely on it; it’s your liability either way.
• Prioritization, not just a list. A ranked, likelihood-to-relist score is worth far more than a raw dump of every expired.
• Personalization that reads human. AI drafting should reference the real listing history, not fill in a generic template. Read the output before you trust it.
• Automated, compliant sequencing. Multi-touch cadences with easy opt-outs and suppression lists baked in.
• Honest pricing. Check the vendor’s current plans on their own site — expired-lead services and AI CRMs price very differently, and “per seat” vs. “per market” adds up.

For the full category comparison, see the best AI lead generation tools for real estate in 2026, and pair expireds with our guide to finding motivated sellers with AI for a complete listing-side pipeline.

Frequently asked questions

Are expired listings worth pursuing in 2026?

Yes — arguably more than ever. Elevated mortgage rates keep some overpriced listings from selling, which creates a steady flow of expireds with genuinely motivated owners. The National Association of Realtors consistently finds that most sellers interview only one agent before signing — so the agent who reaches an expired seller first, with a credible diagnosis, usually wins the relisting.

Where do I actually get expired listings?

Start with your MLS’s expired/withdrawn/cancelled status filter — that’s the authoritative source. Because the MLS rarely exposes seller phone numbers, most agents pair it with an expired-lead service (such as REDX, Vulcan7, or Landvoice) that matches the list to contact info and flags DNC numbers. County recorder and tax records fill in mailing addresses and equity when you only have a name.

Can I legally call expired sellers?

Only after you’ve scrubbed the number against the federal DNC registry and your own list. A homeowner you’ve never transacted with is a cold contact, so the established-business-relationship exemption generally doesn’t apply. Avoid autodialers and prerecorded messages to cell phones without consent, watch out for state mini-TCPA laws, and when in doubt, reach out by mail or compliant email instead.

Is AI outreach to expired sellers compliant?

It can be, if you configure it correctly. AI drafts and schedules the messages, but the compliance obligations are yours: respect DNC lists for calls, honor opt-outs on text and email, include a physical address and working unsubscribe in every email (CAN-SPAM), and load your platform’s suppression lists before launching a sequence.

What makes an expired listing “winnable”?

Three markers: healthy equity (the seller can afford to transact), a pricing gap that explains the failure (fixable with a better strategy), and recency (expired in the last 30 days, before they’ve committed to another agent). AI scoring bakes all three into a single ranking so you spend your time where the odds are best.

How many times should I follow up with an expired seller?

Plan for a 90-day cadence, because most expired sellers relist within that window and most agents quit after one or two touches. Mix the channels you’re cleared to use — mail, compliant email, and calls to DNC-scrubbed numbers — and lead each touch with value (an updated analysis, a marketing idea) rather than a repeat ask.

Bottom line

Winning expireds in 2026 is a speed-and-relevance game, and that’s precisely what AI is good at. Learn how to win expired listings with AI and you turn the most competitive lead source in real estate into a systematic pipeline: source the fresh list, score it by winnability, diagnose the failure, personalize the first touch, stay compliant, and let automation sustain the follow-up. The tools do the heavy lifting from prioritization to first draft — the judgment, the honest pricing conversation, and the relationship are still yours.

See how expireds fit your full stack in our complete guide to AI tools for real estate agents. And grab the free 2026 AI Toolkit for Real Estate Agents — 25 tools that win listings and close deals, in one shortlist.