Follow Up Ace

The AI that works your Follow Up Boss leads.

Ace tells every agent who to contact and what to say — drafts you’ll actually send, landing as a note + task on the contact. From Claude, ChatGPT, or right inside FUB. Learns your voice. Knows Fair Housing. Keep FUB — no migration.

Get Started Free No credit card · 60-second setup

60-second OAuth — keep FUB, no migration.

Free forever • No credit card • Cancel anytime • We don't train on your data • Your contacts stay in FUB

The exact workflow, animated in code — demo data. ▶ Watch the real screen recording — 3-min tour All 9 walkthroughs →
Follow Up Boss Claude ChatGPT Cursor Cline Windsurf Zed OpenCode VS Code Zillow Gmail Twilio Stripe ElevenLabs OpenAI Anthropic
FUB Marketplace certified

A new lead lands. Ace is already on it.

The 5-minute window decides the lead. Ace runs the clock, raises the alert, and has the draft waiting.

5-MinResponse Timer + In-App Alert
LiveCadence Advisor on Every Contact
9Lead Sources Auto-Detected
80+Fair Housing Patterns (Automated Sends)
0:00

A Zillow lead arrives in FUB. Ace logs it and starts the 5-minute timer.

0:14

Alert fires to the assigned agent — with why-now context on the contact.

Slip-alert escalation kicks in if it goes unaddressed past SLA.

0:31

The first-touch draft is ready — the property they inquired on, their search history, your tone.

0:44

Compliance-checked email out the door — you review, you send.

The agent owns the send; Ace builds the draft.

One extra Zillow closing typically nets $9,000–$12,000 in commission. Ace Pro is $660/year. One closing = 12 years of Ace Pro per seat.

Every draft is calibrated by peer-reviewed sales psychology.

Six research layers are injected into every draft before the LLM writes a word — each mapped to a specific study or benchmark dataset.

  • Speed-to-lead frame — MIT Lead Response Management Study (Oldroyd 2007): a 5-minute response is 21× more likely to qualify vs. 30 minutes. New-lead drafts are built for a sub-5-minute send — no “circle back tomorrow” language allowed.
  • Cialdini’s seven principles — reciprocity, commitment, social proof, authority, liking, scarcity, unity — plus Kahneman/Tversky loss-aversion framing. Every opportunity rule gets a tactical playbook: value first, one micro-yes, a specific time — never “let me know when works.”
  • Email craft, measured — Boomerang’s 40M-email study (75–100 words ≈ 51% reply rate); Gong’s 304K-email study (specific-time CTAs book 37% vs. 15%). Every email targets the 50–125 word window with two specific time options.
  • SMS craft, measured — TextUs: under 100 characters gets 2–5× the reply rate. Every SMS targets the 60–110 character sweet spot.
  • Your voice, not the AI’s — your personal voice/brand prompt is resolved into every drafting surface, so drafts read like you wrote them.
  • Fair Housing — regression-tested — 150+ prohibited terms across Fair Housing + licensing patterns (“good schools,” “family-friendly,” “up-and-coming”…), locked in CI: if coverage silently breaks, the build goes red.

One assistant. Same memory. Every surface.

Ace lives inside Follow Up Boss — and the same assistant, with the same memory, rides along in Claude, ChatGPT, and your pocket.

Inside Follow Up Boss

Free chat for every teammate. Voice for Pros.

AI chat embedded in the FUB sidebar — open a contact and it starts on a launchpad built from the model’s read of that person. Voice (Pro) is hands-free while driving and read-only by design: Ace answers and prepares, you tap Send. Daily Brief every morning; the CRM Knowledge Graph speaks your stages, action plans, and smart lists.

The full feature suite →
In Claude & ChatGPT

Your CRM follows you.

Ace works inside Claude, ChatGPT, Cursor, VS Code, and any MCP client — the apps already on your phone. Manage your pipeline while driving or between showings, and because AceMemory carries the thread across surfaces, the triage you start in Claude is already waiting when you open Follow Up Boss at your desk.

How the connectors work →
The MCP Server · Pro

170+ tools. One URL.

A real MCP server every modern AI client speaks — fetch_people, create_task, send_email_direct, bulk_change_stage, zillow_evaluate_lead — real actions through natural conversation. Paste one URL into Claude or ChatGPT, sign in with your FUB email, done.

The Follow Up Boss MCP server →

Plus the Ace Chrome Bridge (Pro): as you work in FUB it syncs the conversations on your screen to your own Ace account, so drafts are grounded in the real words exchanged — not just metadata. How the Bridge works →

For agents: who to call next.
For brokers: what’s actually moving revenue.

Plug it in once. Ace scores every contact, ranks who to action next, and tells you weekly which sources, agents, and stages are stalling or paying off. One flat monthly rate covers your whole team — auto-quoted live from your FUB database size.

6 AI fields, written into FUB

Ace Win Score + Churn Risk free on every account; Expected GCI (est.), Opportunity, Seller Tier, and Last Analyzed with Trove — refreshed nightly and on every webhook.

Command Center

Every at-risk deal and neglected hot lead in one queue, priced in estimated commission — sorted by dollars at stake, each with a ready next move.

AI Signals & Seller Radar

Nightly anomaly detectors catch stalls and gone-quiet deals; Seller Radar scans your database every night for the likely sellers already in it.

Explore Ace Trove Flat monthly, from $49 · see tiers

Per-seat pricing. No per-message tax.

Free to start. Two seat tiers. One account-wide add-on. No commitment, no credit card to begin.

Free $0

Ace chat inside FUB for every teammate — 15 AI messages/day, Fair Housing-trained drafts, live Win Scores.

Ace $25/seat/mo

Unlimited messages, Daily Brief, activity dashboard, and real-time notifications for working agents.

Ace Trove — the account-wide intelligence layer — is a flat monthly add-on from $49/mo, sized to your FUB database. One subscription covers the whole team.

Others charge $449–$1,500+/mo for real estate AI. Full comparison on the pricing page and /compare.

What is Follow Up Ace?

Follow Up Ace is an AI assistant for real estate agents and teams who use the Follow Up Boss CRM. It connects to your Follow Up Boss account, analyzes your leads and conversations, tells each agent who to contact next, and writes ready-to-send follow-up drafts that land as a note and task on the contact — inside Follow Up Boss, from Claude or ChatGPT, or in the standalone Ace Companion on desktop and mobile.

If you choose to connect your Gmail account, Follow Up Ace uses that access only to send the emails you review and approve — from your own address, to your own CRM contacts. It does not read your inbox. Your Google data is never sold, never used for advertising, and never shared with third parties. Follow Up Ace’s use of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements. See our Privacy Policy and Terms of Service.

Frequently Asked Questions

Six calibration layers, each mapped to a specific peer-reviewed study or industry benchmark dataset, get injected into every draft before the LLM writes a word:

  • Speed-to-lead frame — MIT Lead Response Management Study (Oldroyd 2007): 5-min response → 21× qualification odds. New-lead drafts are built for sub-5-minute send.
  • Cialdini’s seven principles — reciprocity, commitment, social proof, authority, liking, scarcity, unity. Each opportunity rule has a tactical playbook.
  • Loss aversion & specificity — Kahneman/Tversky prospect theory; Janiszewski & Uy (2008) precise anchors. Drafts use specific times, not “let me know when works.”
  • Email craft, measured — Boomerang 40M-email study (75–100 words = ~51% reply); Gong 304K-email study (specific-time CTAs book 37% vs. 15%); Voegele direct-mail research (>90% read the P.S.).
  • SMS craft, measured — TextUs data (<100 chars = 2–5× reply rate); Sales.co 2M-message analysis (+78% informal tone); Udext binary-close data (60–75% vs. 10–15% open-ended).
  • Fair Housing + licensing scan — 150+ prohibited terms enforced upfront in the prompt AND scanned in the output. Regression-tested in CI.

Plus your stated voice / brand prompt — resolved per-draft so the message reads like you wrote it. A daily cloud agent compounds the moat: audits coverage every morning, adds one new research finding when something concrete and citable lands, writes the change to a public evolution log.

Ace Pro provides an MCP server at https://followupace.com/mcp — one URL that every modern AI client speaks. For Claude (web/Desktop/Code), open Customize → Connectors → Add custom connector and paste it. For ChatGPT, toggle Developer mode in Settings → Apps and add it as a custom MCP server (Plus/Pro: read-only; Business/Enterprise: full read+write). Cursor and VS Code support one-click deeplink installs. Cline, Windsurf, Zed, OpenClaw, Continue, and Goose all read the same JSON config. Sign in with your FUB email and you get 170+ tools — fetch_people, create_task, send_email_direct, bulk_change_stage, zillow_evaluate_lead, fetchMortgageRates, and more — all through natural conversation. Per-client install pages live at /install/<client>/. Power users can also install the Claude Desktop DXT (advanced) for the full tool surface via stdio.

FUB's native AI offers basic text suggestions. Ace gives you 170+ MCP tools that take real action — bulk_change_stage moves hundreds of contacts instantly, generate_message drafts personalized emails using contact history, zillow_speed_to_lead_check classifies lead urgency in seconds, and ace_engine_brief gives leadership a prioritized action summary across the entire team. It's the difference between autocomplete and a full AI operations layer.

Not at all. If you can type a question into ChatGPT, you can use Ace. There's nothing to code. For Claude, paste one URL into Customize → Connectors. For ChatGPT, paste one URL into Settings → Apps (Developer mode). For Cursor and VS Code, click a one-click install link. For Cline, Windsurf, Zed, OpenClaw, Continue, or Goose, drop one JSON block into the client's config file. Sign in with your FUB email and you're done.

Yes. Your contacts, conversations, and notes stay in Follow Up Boss — Ace reads them through FUB's official API over encrypted connections. The scores, insights, and memory Ace generates are kept in your account's isolated workspace, and personal details like names, emails, and phone numbers are tokenized before any AI model sees them. We never train public models on your data.

Yes — 7 dedicated Zillow MCP tools. zillow_speed_to_lead_check classifies urgency (5-minute gold standard, 15-minute max). zillow_evaluate_lead returns priority scoring (HIGH/MEDIUM/LOW), recommended first-contact channel, and a compliance-safe outreach draft. zillow_qualify_contact runs BANT+M scoring (Budget, Authority, Need, Timeline, Motivation — 0-100 scale). zillow_generate_followup returns the full 10-day Zillow play — Day 1 outreach, Days 2-4 value-add, Days 5-7 market data, Days 8-10 re-engagement — that the AI drafts each day's outreach against using your live FUB context, ready to send in one tap. zillow_check_stage enforces 11 lifecycle stages (New Lead through Closed). Cadence engine caps at 3 contacts/day with 9PM-8AM quiet hours.

Yes, the MCP integration requires your own subscription to Claude Pro or ChatGPT Plus. Ace Pro provides the secure connector that links them to your Follow Up Boss data. The built-in Ace chat assistant works without any external AI subscription.

Voice chat works like talking to Siri or Alexa. Just tap the microphone and say what you need: "What's on my calendar today?" or "Summarize the leads from yesterday's open house and draft a follow-up email I can review." Voice is read-only by design — Ace answers, summarizes, and prepares; you tap Send when you're ready.

Ace writes 6 model-grounded custom fields into Follow Up Boss for every analyzed contact. Free on every account: customAceWinScore (calibrated 0–100 probability the contact converts to a closed deal) and customAceChurnRisk (Low / Medium / High). With Ace Trove: customAceExpectedGCI (est. — calibrated win probability × your account's real median commission), customAceOpportunity (At Risk / Active Buyer / Active Seller / Likely Seller / Nurture), customAceSellerTier (Very High → None), and customAceLastAnalyzed. Values refresh nightly from the model suite and in real time as FUB webhooks fire. In June 2026 the legacy Ace Score / Tier / Status / summary fields were consolidated into this lean set — those signals are still computed internally and power the dashboards. Flat monthly pricing from $49/mo based on database size.

Yes. Ace is free forever with 15 AI messages per day, live Ace Win Scores on your contacts, and knowledge base access. Upgrade to Ace ($25/mo) for unlimited messages, your activity dashboard, and real-time notifications. Or go Pro ($55/mo) for voice, MCP Connectors, and the Zillow Playbook. Ace Trove is a separate account-wide add-on (flat monthly, $49–$899 based on FUB database size) that powers the Command Center queue, anomaly detectors, AceMemory, and the paid AI custom fields written into FUB — Expected GCI (est.), Opportunity, Seller Tier, Last Analyzed. No credit card required to start.

Absolutely. Each agent on your team can be on a different plan — Free, Ace, or Pro. Admins can upgrade team members or let agents upgrade themselves. No logins needed — the embedded AI agent appears right in Follow Up Boss.

Yes. Every call, text, email, note and appointment is attributed to the deal it lands under — activity inside an open deal's window counts as deal work pushing a closing, everything else as lead-gen work pushing toward the next deal. That powers the Quiet Deal Radar (open deals with no attributed activity in 7+ days) and Team Cockpit's Effort Split (each agent's deal-work vs lead-gen share) — and clients mid-transaction get tighter gone-quiet alerts instead of generic re-engagement nudges. Part of Ace Trove.

Ask your favorite AI about Follow Up Ace

Don’t take our word for it. Ask an AI you already trust.

Each button opens a new tab with the question pre-filled. The answer is the AI’s own — we don’t write it.

CINC charges $1,500/mo. kvCORE charges $499.

Know who to work next.
Starts free.

Every agent gets a ranked queue and a ready-to-send draft — inside the Follow Up Boss you already run. No credit card to start.

Free forever • No credit card • Setup in under 3 minutes