Lead Response and Follow-Up

Drip Campaigns: Building Automated Follow-Up That Feels Personal

Build real estate drip campaigns and CRM action plans that read like a person wrote them, match the lead’s timeline and stop the moment someone replies.

AssistCall research teamPublished Reviewed by Faruque Amin, VP of Sales and Marketing7 min read

On this page
  1. Why most drips read like a machine wrote them
  2. Automation is already in the toolkit
  3. Start from the lead’s timeline, not yours
  4. Build one sequence properly, step by step
  5. Rewrite one message: before and after
  6. Personalization that is actually personal
  7. Stop rules and consent
  8. Example messages for a seller “someday” sequence
  9. Drip mistakes that undo the good work
  10. Measure messages, not just campaigns
  11. Frequently asked questions
  12. References
Lead Response and Follow-Up: Drip Campaigns: Building Automated Follow-Up That Feels PersonalArticle coverLEAD RESPONSE AND FOLLOW-UPDrip Campaigns:Building AutomatedFollow-Up That FeelsPersonalAva for Real Estate TeamsInsightsLead Response and Follow-Up: Drip Campaigns: Building Automated Follow-Up That Feels PersonalArticle coverLEAD RESPONSE AND FOLLOW-UPDrip Campaigns:Building AutomatedFollow-Up That FeelsPersonalAva for Real Estate Teams

Key takeaways

  • A drip feels personal when it uses what the lead actually told you, sounds like you and stops the moment they reply.
  • Seller timelines are long: the median seller thought seriously about selling for three to four months, and 27% for six months or more[1].
  • Half of AI-using REALTORS already use AI to draft emails and follow-ups[2]; edit every template yourself and never pass software off as a person.
  • Build stop rules before you build messages: replies, bookings, opt-outs and bounces should end or pause a sequence automatically[3].

Why most drips read like a machine wrote them

Leads can spot an automated sequence within two messages. The giveaways are familiar: a subject line that could have gone to anyone, a greeting with the wrong name, a reference to “your home search” when they asked about selling, and, worst of all, a cheerful check-in that arrives the day after they replied to say they bought with someone else. None of this is caused by automation itself. It comes from sequences built around the agent’s calendar instead of the lead’s situation.

A good drip campaign, often called an action plan in real estate CRMs, does three jobs. It makes sure no lead is forgotten, it delivers something useful at a sensible pace and it hands the conversation back to a human the moment the lead engages. If a sequence does not do the third job, it is not follow-up; it is noise.

Automation is already in the toolkit

Agents are not starting from zero. Among REALTORS who use AI, 52% use it to create emails and follow-ups, behind only listing descriptions at 75% and social media posts at 56%[2]. Drafting is not the hard part any more. The hard part is making the result sound like you and fit the person reading it.

What AI-using REALTORS use AI for
What AI-using REALTORS use AI forBar chart. Listing descriptions: 75%; Social media posts: 56%; Emails and follow-ups: 52%; Document review or summarizing: 27%.Listing descriptionsListing descriptions: 75%75%Social media postsSocial media posts: 56%56%Emails and follow-upsEmails and follow-ups: 52%52%Document review or summarizingDocument review or summarizing: 27%27%What AI-using REALTORS use AI forBar chart. Listing descriptions: 75%; Social media posts: 56%; Emails and follow-ups: 52%; Document review or summarizing: 27%.Listing descriptionsListing descriptions: 75%75%Social media postsSocial media posts: 56%56%Emails and follow-upsEmails and follow-ups: 52%52%Document review or summarizingDocument review or summarizing: 27%27%

Source: National Association of REALTORS®, 2026[2]

Show the data
ItemValue
Listing descriptions75%
Social media posts56%
Emails and follow-ups52%
Document review or summarizing27%

There is also a trust question. In Pew Research Center’s 2025 survey, 76% of Americans said it is extremely or very important to be able to tell whether content was made by AI or by people[4]. Use AI to draft if it helps, then rewrite in your own words, sign as yourself and never pretend a fully automated message was typed personally that minute.

Start from the lead’s timeline, not yours

The most common drip mistake is a single 30-day sequence for everyone. Sellers in particular move slowly. In Zillow’s 2026 seller survey, the median seller thought seriously about selling for three to less than four months before listing, about 6% thought about it for less than a month and 27% for at least six months[1]. A seller lead who says “maybe next year” needs a sequence measured in months, with long gaps and genuinely useful content, not daily emails that burn out in three weeks.

Example sequence shapes by lead type; adjust to your market
Lead typeLengthRhythmChannel mixHand back to a person when
New buyer, active14 daysDays 0, 1, 3, 5, 8, 14Call, text, emailAny reply or a saved-search view of the same home twice
Seller, “someday”6 monthsEvery 2 to 4 weeksEmail, occasional textReply, home value request or a change of timeline
Aged lead revival30 daysDays 0, 7, 14, 30Text, then emailAny reply, including “not now”
Past clientOngoingMonthly or quarterlyEmail, call on anniversariesAny mention of moving, a friend moving or a question
Renter thinking of buying9 monthsMonthly, extra touch 90 days before lease endEmail, textReply or a lender introduction request

Build one sequence properly, step by step

  1. Define the trigger

    Pick one specific event that starts the sequence, such as a new seller lead tagged “6+ months”. Vague triggers produce messages that do not fit.

  2. Write the stop rules first

    Decide what ends or pauses it: a reply, a booked appointment, an opt-out, a bounced email or a tag such as “working with another agent”.

  3. Write the first three messages yourself

    Read them aloud. If you would not say it on the phone, rewrite it. Keep emails plain text with a normal signature.

  4. Use only merge fields you trust

    First name and area are usually reliable. Home type, price and timeline often are not. A wrong merge field reads worse than no personalization.

  5. Set send windows

    Send texts in daytime hours in the lead’s time zone, and never at a time you would not phone them.

  6. Test on yourself

    Enroll your own contact record and read every message on your phone before a single lead receives it.

  7. Review monthly

    Check replies and opt-outs per message. Rewrite or remove any message that never gets a reply.

Rewrite one message: before and after

Example, before: “Hi {first_name}! Just checking in to see how your home search is going. The market is moving fast, so let me know if you have any questions. I’m here to help with all your real estate needs!” It could be sent by anyone to anyone, it asks nothing specific and it ends with a slogan.

Example, after: “Hi Sam, it’s Jo. You mentioned wanting a yard for the dog and staying near Lincoln Elementary. Two homes came up this week that fit both; one has a fenced yard. Want the links, or shall I hold off until closer to your lease end in May?” It uses two facts the lead gave, offers something concrete and gives an easy way to say not yet.

The second version takes thirty seconds longer to set up, because it depends on fields you filled in during the first call. That is the real cost of a personal drip, and it is worth paying.

Personalization that is actually personal

Real personalization comes from what the lead told you, captured in fields your CRM can use. Three or four are enough: the area they care about, the type of home, the timeline and the reason for moving. A message that says “Here are three homes near the elementary school you mentioned” beats any number of clever subject lines.

  • Ask one extra question on every call and save the answer to a field.
  • Write variants of key messages for the main reasons people move: more space, a new job, downsizing.
  • Let the sequence reference real local events and data you update monthly, not evergreen filler.
  • Send from your own name and address, with replies coming to you, not to a no-reply inbox.

Every sequence needs automatic exits. A lead who replies should never get the next scheduled message; the reply should pause the sequence and create a task for you. A booked appointment should end it. For texts, opt-outs must work immediately: replying stop, quit, end, revoke, opt out, cancel or unsubscribe is treated as a reasonable way to revoke consent to automated texts and calls[3].

Automated voice is a separate risk. In February 2024 the FCC ruled that calls using AI-generated voices are “artificial” under the TCPA, and robocalls to consumers require prior express written consent, with AI voice calls held to the same standard[5]. Do not add prerecorded or AI voice drops to a drip without consent your counsel has reviewed.

Example messages for a seller “someday” sequence

Example, month 1 email, subject “Westbrook sales this month”: “Hi Carla, you mentioned you might sell next year, so here’s a quick look at what sold near you in October and how long each took. No action needed; I’ll send one of these each month. If your plans change, just reply.”

Example, month 3 text: “Hi Carla, Lee from Example Realty. A house like yours on Pine Street just went under contract. Happy to tell you what that might mean for your timing whenever it’s useful.”

Example, month 5 email, subject “Before spring”: “Hi Carla, if spring is still the plan, now is a good time to walk through what to fix and what to leave. Want me to come by for twenty minutes, no paperwork?”

Drip mistakes that undo the good work

  • Enrolling every lead in every campaign, so a single contact gets three overlapping sequences.
  • Sending listing alerts and a drip on the same day, which doubles the volume without adding anything.
  • Leaving old dates or seasons in evergreen templates, such as a “spring market” email that keeps going out in October.
  • Forgetting to remove a lead who became a client, so they keep receiving “are you still looking?” messages after closing.
  • Copying a vendor’s template library without reading it, then discovering a message that promises a service you do not offer.
  • Never checking the sent folder. Read a week of outgoing automated messages every month as if you were the lead.

Measure messages, not just campaigns

  1. Reply rate for each message in the sequence.
  2. Opt-outs and unsubscribes for each message.
  3. Appointments booked within 30 days of a reply.
  4. Leads who went quiet after a specific message.

Prune ruthlessly. A sequence with four excellent messages beats one with twelve mediocre ones, and every message you cut is one less chance to annoy someone who was going to call you anyway.

Frequently asked questions

How many messages should a real estate drip campaign have?

Fewer than you think. An active buyer sequence might have six touches over two weeks; a long seller nurture might send one useful message every few weeks for six months. Cut any message that never earns a reply.

How do I stop a drip when a lead replies?

Set a stop rule in your CRM so any reply pauses the sequence and creates a task for you. Test it by enrolling yourself and replying to the first message.

Can I use AI to write my follow-up emails?

You can use it to draft, but edit each template into your own words and sign as yourself. Many people want to know whether content was written by AI, so never present automated messages as personally typed.

Are automated texts to real estate leads legal?

Rules depend on consent and how messages are sent. Honor opt-outs such as STOP immediately, keep send times reasonable and have your broker or counsel review your consent wording.

References

  1. Zillow Research (2026). Sellers: Results from the Zillow Consumer Housing Trends Report 2026 (opens in a new tab) ↑

  2. National Association of REALTORS® (2026). REALTORS® Adopt Technology to Save Time and Improve the Client Experience, NAR Report Finds (2026 REALTORS® Technology Report) (opens in a new tab) ↑

  3. Legal Information Institute, Cornell Law School (eCFR text) (2026). 47 CFR § 64.1200 - Delivery restrictions (TCPA rules) (opens in a new tab) ↑

  4. Pew Research Center (2025). How Americans View AI and Its Impact on Human Abilities, Society (opens in a new tab) ↑

  5. Federal Communications Commission (2024). FCC Makes AI-Generated Voices in Robocalls Illegal (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo