Speed to Lead

Past-client referral automation

A Houston team had 847 past clients, never followed up. One 3-touch campaign got 23 ready movers ($195,500 GCI). See the playbook.

The Castillo Group is a three-agent real estate team in Houston’s Energy Corridor. In early 2025 they had been operating for six years and had closed 141 transactions across that span. Every client’s information was in their CRM: name, address, phone, close date, property type. The total contact count was 847.

They had never sent a single systematic follow-up to any of them.

Operator details anonymized. Based on a real LeadExploder account matching this profile.

Houston Texas real estate agent congratulating past clients at suburban home closing, warm handshake

Their business was 100 percent referral and repeat, which they considered a sign of health. What they had not calculated was how much of that repeat business was coming to them versus going somewhere else. A past client thinking about selling in 18 months does not wait around. They Google a name, they remember a yard sign, they ask a neighbor. If no agent is already in their inbox, they start fresh.

In March 2025, they ran their first reactivation campaign. Three touches over three weeks. 847 contacts received touch 1. By the end of the sequence, 180 had replied. Twenty-three indicated they were thinking about buying or selling within the next 12 months.

At an average GCI of $8,500 per transaction, that is $195,500 in potential pipeline from a single campaign to a list they already owned.

Why does the past-client database outperform every paid channel?

Every lead you buy from Zillow, Realtor.com, or Google Ads is a stranger. They do not know you, they are comparing you to three or four other agents simultaneously, and your cost to acquire them ranges from $35 to $400 per lead depending on the market.

A past client already closed with you. They remember your name. They associate you with one of the largest financial decisions of their life. The trust that took months to build during the transaction is still there; it just needs a signal to reactivate. The cost to send that signal is a few cents per SMS.

According to the National Association of Realtors 2024 Profile of Home Buyers and Sellers (nar.realtor), 20 to 30 percent of a mature team’s volume should come from past clients and referrals within 24 months of the close. When that number is below 20 percent, it almost always means the team has no systematic follow-up in place, not that the clients are unhappy.

Cleaning the list before you send a single message

Before the Castillo Group sent touch 1, they spent two hours on data hygiene. This step is skipped more often than any other, and it is the one most likely to cause problems.

Remove duplicates first. A CRM that has been populated over six years almost always has duplicate records: the same contact entered twice from different sources, or a couple entered once as individuals and once as a household. Sending two identical messages to the same person in the same sequence is the fastest way to kill a conversation before it starts. Most CRMs have a built-in merge tool. Use it.

Update phone numbers. Numbers go stale. People change carriers, move areas, or port numbers between lines. Before launching a reactivation campaign, run the list through a phone validation service. LeadExploder real estate account data from 2024-2025 shows that lists older than 18 months typically have a 12 to 18 percent invalid number rate. That is one in six contacts who will never receive touch 1, not because the message failed but because the number does not connect to anyone.

Scrub disconnected lines. A disconnected number that receives an SMS is flagged as undeliverable and can affect your sender score over time, reducing deliverability for the rest of the campaign. Run a quick check and remove disconnected numbers before launch. This takes about 20 minutes with an automated validation tool and protects the entire campaign.

Segment by recency. Contacts who closed within the last 24 months are warm. Contacts from three or more years ago are cold. The message for each group should be different. The warm group gets a check-in tone. The cold group gets a softer opener that acknowledges the time lapse. Mixing the two in the same sequence produces mediocre results for both.

What does the 3-touch reactivation sequence actually look like?

Real estate agent reviewing automated referral email sequence on computer, past client list visible, Houston Texas office

The Castillo Group’s sequence was simple on purpose. Each message was short enough to read in under 10 seconds, personal enough to feel like it came from the agent, and spaced far enough apart to avoid feeling like a campaign.

Touch 1 (Day 1): The personal check-in. No ask.

Hi [First Name], it’s [Agent] from The Castillo Group. We closed [X] years ago now and I realized I hadn’t checked in in a while. Hope you and the family are doing well. Let me know if you ever need anything on the real estate side.

The goal of touch 1 is a reply, not a conversion. Any reply, including “thanks, all good!” is a win. It opens the channel and confirms the number is active.

Touch 2 (Day 10): The neighborhood update. One concrete data point.

[First Name], quick market note for your area: homes in [Neighborhood/Zip] are averaging [X] days on market this spring, up/down from last year. Inventory is [tight/opening up]. Happy to pull a full report if you’re curious what your place is worth today.

Touch 2 provides value with no ask attached. It positions the agent as a market resource. Contacts who were not thinking about real estate start thinking about it.

Touch 3 (Day 21): The direct ask.

[First Name], wrapping up my spring check-ins. Simple question: do you know anyone thinking about buying or selling in the next 6 to 12 months? I’d love to help someone you trust. And if it’s you, even better.

Touch 3 is the referral ask. It is direct and brief. The phrase “someone you trust” is intentional: it frames the referral as doing someone a favor, not as a sales transaction.

How to handle “how did you get my number?”

With 847 contacts spanning six years, some recipients will not immediately remember the agent or the transaction. A small percentage will reply with some version of: “Who is this?” or “How did you get my number?”

The response script that works is honest and short:

Hi [First Name], of course. We closed on your home at [Address] back in [Year]. I’m [Agent Name] from The Castillo Group. Wanted to check in since it’s been a while. Sorry if the message came out of the blue.

The key components of that response: confirm the specific transaction, use the address not just the neighborhood, give the full name and team name, and apologize briefly for the surprise without being defensive. This response resolves the concern in almost every case. The contact is reminded of a real event they participated in, which re-establishes trust immediately.

Do not say “I got your number from our records” as the opener. It sounds vague. The specific address and year does the work.

If a contact explicitly asks to be removed from outreach, remove them immediately and reply to confirm. LeadExploder automates this: an opt-out reply triggers an unsubscribe tag and removes the contact from all active sequences. This is not optional; it is both a legal requirement and a reputation protection move.

What to do with the 23 warm prospects after they identify themselves

Real estate agent reviewing referral revenue from past client automation over 12 months, Houston Texas

The 180 replies sorted into five categories: friendly, market-curious, not now, referral, and ready-to-move. The 23 who flagged as ready-to-move did not all say the same thing. Some said “actually, we’ve been thinking about selling.” Others said “my daughter is looking to buy in our neighborhood.” A few said “we’re planning to move in about a year.”

Each of these requires a different next step.

Active sellers (timeline under 6 months): The agent calls same-day, schedules a listing consultation within the week, and moves the contact out of the nurture sequence and into an active pipeline. This is not an automated next step. The automation’s job is done; the agent takes over.

Active buyers (timeline under 6 months): The agent calls within 24 hours, confirms search criteria, and connects them with an automated lead follow-up sequence specific to buyers. If the buyer is in a neighboring area, the agent can route them to a preferred partner agent and collect a referral fee.

Referral contacts (pointing to someone else): The agent sends a personal reply thanking them and asks for an introduction: “Would you be comfortable connecting me with them directly? I can reach out with a quick introduction and take it from there.” The goal is a warm introduction, not a cold message.

Timeline of 6 to 18 months: Move these contacts into a multi-year nurture sequence designed to stay present without pressure. Market updates every 3 to 4 weeks, one personal check-in per quarter. The sequence does not drop them; it just reduces frequency and adjusts tone to match where they are.

Timeline over 18 months: Tag as long-hold and set a calendar reminder for the agent to re-engage manually at the 12-month mark. The automation keeps them on a low-frequency sequence in the meantime.

The 90-minute investment the agent made in running the campaign produces 23 sorted conversations with a clear next step for each. No inbox full of unread texts. No list of people to call with no context.

How does the system handle replies?

This is where the automation pays for itself. When 180 contacts reply across a 3-week sequence, a human would need to triage 180 conversations and decide which to prioritize. Without a system, the highest-intent replies get buried under the friendly but low-urgency ones.

LeadExploder routes every reply into a conversation thread and applies a simple tag based on the content: buyer, seller, timeline, referral, or not now. The 23 contacts who indicated active intent get flagged immediately and trigger a notification to the agent with the full conversation context. The agent calls those 23 first and lets the system handle the 157 courtesy replies with an automated response.

The agent worked for about 90 minutes of their actual time across the full 3-week campaign. The system did the outreach, monitored for replies, and sorted the pipeline.

What does the ROI math look like?

The Castillo Group’s campaign produced 23 identified warm prospects from a list of 847. Not all 23 will transact in the next 12 months. Real estate timelines slip. But even at a 40 percent close rate on those 23, that is 9.2 transactions.

At $8,500 average GCI: $78,200 in closed commission from one campaign, to a list they owned, with no paid media spend.

If they run the same reactivation once a year and layer in quarterly market updates between cycles, the cumulative effect compounds. Contacts who were not ready in March may be ready in October. The database gets warmer every time it receives a relevant touch.

Teams that run consistent past-client programs report that within two to three years, 25 to 35 percent of their new listings come from the existing database without any paid advertising.

What to do this week

Pull your full contact list from your CRM. Filter for anyone who closed more than 90 days ago and has not received an outreach from you in the last six months. That is your reactivation list.

Before sending anything, run the hygiene pass: deduplicate, validate phone numbers, and segment by recency. Budget two hours for this. It is not glamorous but it directly determines how many people receive your first message.

Write the three templates above with your name, your specific neighborhoods, and one real market data point. Load them into a sequence. Set touch 1 to send Monday, touch 2 at day 10, touch 3 at day 21.

Then let it run. You do not need to monitor it. You need to be ready to call the people who reply with timing.

The database is already there. Most teams just never activate it.

Book a demo and see past-client reactivation running live.


Alex Rocha is the founder of Mastodon Marketing, a Houston-based growth agency that runs marketing for service businesses across 70+ client sites. He built LeadExploder as the operating system he wished his clients had on day one. Learn more about Alex →

Frequently asked questions

How often should a real estate team contact past clients?

At minimum, three times per year with a purpose behind each touch: a market update relevant to their neighborhood, a check-in around the anniversary of their purchase, and a direct ask once per year. Teams that systematize all three consistently see 20 to 30 percent of their closed volume come from past-client referrals within 24 months.

What is the best way to reactivate a cold real estate database?

A 3-touch sequence spread over three weeks works well. Touch 1 is a personal check-in with no ask. Touch 2 is a neighborhood-specific market update with one concrete data point. Touch 3 is a direct question: do you know anyone thinking about buying or selling? Keep each message under 160 characters for SMS deliverability.

What response rate should I expect from a past-client reactivation campaign?

Response rates vary by list age and how recently you engaged the contacts. For a warm list with no prior automation, 18 to 25 percent reply rates on a 3-touch SMS sequence are realistic. For a colder list (2 or more years since last contact), expect 8 to 14 percent. Either number produces meaningful pipeline from a list you already own.

Does automating past-client outreach feel impersonal to the client?

Only if it is written impersonally. Messages that reference the specific neighborhood, use the client's first name, and arrive from the agent's personal number read as attentive, not automated. Most recipients assume the agent typed it. The goal is to write the message the agent would send if they had time to send it manually.

More on Speed to Lead

5-Minute Lead Response: 2026 Buyer Data

5-Minute Lead Response: 2026 Buyer Data

Read →
Mortgage drip campaigns that close

Mortgage drip campaigns that close

Read →
Mortgage pipeline automation, mapped

Mortgage pipeline automation, mapped

Read →
Book my live demo