Free class · Clients and follow-up
Most agents only reach out at the obvious times. The opportunity lives in between. This class shows you how to turn a simple Sheet into a relationship memory, let AI do the organizing, research, and drafting, and find a real reason to reach out before anyone has to chase you.
Part 1
Relationships are where the business comes from. In NAR's 2025 Generational Trends report, 40% of buyers found their agent through a friend, neighbor, or relative, and 17%used an agent they'd worked with before. Every useful conversation with someone you already know is an at-bat.
A Google Sheet becomes an AI-native CRM when AI does the administration. The Sheet stays inspectable and portable. AI reads it, updates it, and prepares the next-action queue. People hate managing spreadsheets. AI is good at it, because it can follow a clear instruction to a specific row and column.
The operating loop
Raw signal → AI organizes → research → draft and CRM update → your review → you communicate → the next at-bat.
AI handles the clerical handoff, analysis, research, and drafting. You own the relationship, the relevance, the permission, the judgment, and the send. Every touchpoint gets recorded and feeds the next one, so you're building a stack, not scattering bricks.
Part 2
Clients expect to hear from you at deadlines and closings. That's the floor. Proactive communication says “I remembered what matters to you, noticed something changed, and did the work before you had to ask.”
Hand AI the raw note, voice memo, or contact card. Get back a clean record, a short personal note that proves you remember the conversation, and the next at-bat. Never auto-enroll them in anything.
Deliver the thing you promised: the answer, the local resource, the introduction. Verify it first, and ask one clear question. Not a generic nurture email.
Find the reason before you write the message. A verified local event, a market change, a seasonal home task, a vendor, or an open promise. No manufactured urgency, no "just checking in."
Read the last touch and the deal record, pull out promises and gaps, and send the update that answers the next question before they ask it. No legal, inspection, lending, tax, or pricing advice from AI.
Closing starts the longest communication window in the relationship. Close loose ends first, then offer help tied to what you actually know: trusted local vendors, seasonal maintenance, verified local events.
Ask for a review or referral after a real positive signal, like a completed win or explicit thanks. A calendar date alone never earns the ask.
Paste whatever you have, exactly as it is. A messy note, a voice-memo transcript, a business card and a line of context. Don't organize it first. That's the AI's job. You get back the extracted facts, what needs verifying, a short draft, the exact CRM update, and a send, hold, or no-action call.
Contact exchange to CRM handoff
Turn the raw contact signal below into a reviewed CRM handoff. Do the organization for me. Use only the source artifact and public sources you can cite. Do not search for private personal information. Do not infer protected-class, family, financial, health, or lifestyle details. RAW SOURCE ARTIFACT [Paste the note, transcript, voice-memo transcription, contact card plus context, email, or other client-safe source exactly as you have it.] Before producing the handoff: - extract the name, how we met, stated need, human detail, geography, channel permission, promises, timing, and unknowns - distinguish explicit facts from your interpretation - if one missing fact would make any communication inappropriate, especially identity or permission, ask no more than 2 short questions. Otherwise mark the gap VERIFY and continue Produce 5 separate blocks: 1. EXTRACTED SIGNAL - the organized contact facts, relationship context, need, permission, promise, and timing you extracted 2. RESEARCH / VERIFY - separate public research you can complete from facts that require me or the client - for completed research, include the source URL and as-of date - mark every unresolved fact VERIFY 3. DRAFT - under 90 words - prove I remember the actual conversation - give one useful next step - no fake urgency, no generic "checking in," no invented familiarity 4. CRM PATCH - CONTACTS fields: relationship type, lead source, current need, human detail, permission, next action, next-action date - TOUCHPOINTS fields: summary, client need, promise, owner, due date, next step, approval state, source notes - keep missing values as VERIFY instead of guessing 5. NEXT AT-BAT + HUMAN DECISION - name the next useful at-bat and why its timing is earned - recommend SEND, HOLD, or NO ACTION, and explain in one sentence - list what must be verified before any promised resource or claim is delivered Do not send anything or change the CRM. I will approve the draft and apply the patch.
Part 3
Before anything goes out, answer five questions. If the answer to “why now” is “because the sequence says so,” don't send it.
Inputs you control
Outputs you watch
More at-bats is the input. More business is the output. You can't will an output into existence, but you can audit inputs and learn which ones earn real conversations. Don't optimize for sends. Optimize for useful conversations without weakening trust. Start with 5 relationships this week and create 3 useful at-bats.
Free with a student account
Create a free student portal account to get the workbook, prompts, templates, and downloads for every class.
Create my free account →Already a student? The same link signs you in.
No. A Google Sheet with four tabs works, and it stays visible, editable, and portable. AI features inside Sheets depend on your Google plan and rollout, but you can also paste rows into any AI assistant. If you already use a CRM, the same process works there.
A message with a real, verified reason: something you promised, a local event with a confirmed date, a seasonal task tied to a fact you know about their home, or a real change in the market or their deal. Give value before you ask for anything. If the only reason is that a sequence says so, don't send it.
After an earned positive moment, like a completed win or explicit thanks, and only when nothing is left unresolved. Right after a good closing is a natural time. A calendar date alone doesn't earn the ask, and you never script what the client says.
Not in this method. AI organizes, researches, and drafts, and every draft stays a draft until you read it and send it yourself. Respect each contact's permission and opt-outs. Meeting someone at an open house isn't automatic consent to put them on a marketing list.
Track the inputs you control, like contacts with a dated next step, promises kept on time, and useful touches sent. Then watch the outputs, like replies, real conversations, referrals, and opt-outs. Ask AI which inputs line up with the best conversations, and don't claim cause and effect from a small sample.
Clients and follow-up
Be your client's first AI voice, stress-test AI answers together, and gate every claim before it reaches a deal.
Clients and follow-up
Find the one useful next action for a contact, draft only when permission and facts support it, and record what really happened.
Clients and follow-up
Reset client expectations with one boundary message, 8 ready replies in your voice, and a Monday update that gets ahead of check-in texts.
Library
The whole library of recorded classes, by topic.
Research sources
NAR's buyer and seller survey, the source of how buyers found their agent (Exhibit 4-4).
First-party help for using Gemini beside your spreadsheet, including plan and availability limits.
How the in-cell AI function works on a row you choose.
The rules any commercial email you draft still has to follow, including opt-outs.
The rules for asking clients for reviews without scripting or filtering them.
Why the prompts never infer protected-class or family details about a contact.
Product access, limits, local listing rules, platform policies, and laws can change. Re-check the linked authority before important or public work.
Keep building
AI Acceleration: Real Estate is a private, self-directed curriculum for practical AI work across content, communication, lead follow-up, CRM, client research, operations, and relationship management. Build it. Review it. Keep what works.
Explore the Skool community ↗Follow Ryan + AI Acceleration
No ranking, revenue, closing, or time-saving promise. The classes teach the work and the review systems around it.