What's Inside

How Realtors Work — Your full production workflow broken down by business area, with AI opportunities mapped to each stage. 🗺️

This Week in AI — Cheaper models, voice-first interfaces, and a fresh round of AI agents behaving badly in security tests. 🗞️

Top Picks — Six tools for redlining contracts, dictating reports, editing listing photos, captioning video, and building prospect lists. 🔧

Deep Dives:

📜 Redline Contracts with Claude in Word — turn Claude loose on a purchase agreement or offer letter before you sign or send it

🎙️ Run Your Workday by Voice with ChatGPT — dictate a market update and get a finished, formatted report back

🎨 Adobe for ChatGPT — 70+ Photoshop, Firefly, and Premiere tools for listing photos and marketing, all inside a chat window

🎬 CaptionBolt — auto-captions for listing walkthrough videos and Reels in one of 300+ styles

📝 Wispr Flow Notetaker — turns showings, listing appointments, and calls into searchable notes and transcripts

📇 Cleanlist — describes the prospects you want in plain English, gets back a CRM-ready list with verified contact info

Mini Skill — Turn a handful of bullet points about a property into a polished listing description in under 10 minutes. ✍️

How Realtors Work: An AI Opportunity Map

Lead Generation. Sphere outreach, expired listings, FSBOs, online leads, door knocking, and social media all feed the top of your pipeline, and most agents run every channel manually with a spreadsheet or a CRM tag. The friction shows up in volume: an agent who could reach 50 expireds a week by hand might reach 200 with AI drafting the first-touch letters and texts. The bottleneck usually isn't finding the leads, it's finding the time to personalize outreach at scale.

Lead Nurturing. CRM follow-up, drip email sequences, text and call cadences, and periodic market updates are what turn a cold lead into a client, but nurturing dies the moment it becomes a chore. Most agents let sequences go stale because writing fresh content every month is tedious. AI closes that gap by drafting market updates, seasonal check-ins, and personalized follow-ups in minutes instead of hours, so the cadence actually holds.

Buyer Representation. The buyer side runs through consultation, property search, showing coordination, offer writing, negotiation, and transaction management. Time gets lost in scheduling back-and-forth, summarizing long inspection reports for anxious buyers, and drafting offer letters under deadline pressure. This is one of the areas where AI's reading and writing speed pays off fastest, especially for boiling down 40-page inspection reports into a one-page buyer briefing.

Seller Representation. Listing presentations, pricing strategy, listing prep, photography coordination, marketing, open houses, offer review, negotiation, and transaction management make this the most operationally heavy side of the business. Sellers judge you on responsiveness and polish, both of which AI can help with directly, from drafting the listing presentation deck to generating the property description to captioning the walkthrough video. The main friction point is marketing velocity: sellers expect content across MLS, social, and email, and manually producing all of it eats a full day per listing.

Marketing and Content. Social media, video, email newsletters, community content, and personal branding are a full-time job layered on top of an already full-time job. Most agents either neglect this entirely or hire it out. AI tools now handle a meaningful chunk of the production work directly, from caption generation to photo editing to first-draft newsletter copy, which is exactly what this newsletter tracks every week.

Administrative and Operations. Contract management, deadline tracking, vendor coordination, commission tracking, and compliance are unglamorous but unforgiving; one missed deadline can blow up a deal. This is a strong fit for AI review tools that read documents for you and flag what needs attention, catching the kind of detail that's easy to miss at 9pm after a full day of showings.

Client Experience and Retention. Closing gifts, post-close follow-up, referral cultivation, and review requests are what turn a one-time client into a repeat client and a referral source, but they're also the first thing to slip once a deal closes and attention moves to the next one. Simple automations and AI-drafted personal check-ins keep this from falling through the cracks without making it feel automated.

Professional Development and Recruiting. For agents who train or recruit others, AI is increasingly useful for building training materials, roleplay scripts for objection handling, and onboarding documents that used to take a full weekend to put together from scratch.

This Week in AI — What It Means for Realtors

The biggest theme this week wasn't a flashy new model, it was price. DeepSeek, OpenAI, and Alibaba all cut prices or boosted the capability-per-dollar of their models significantly, and that matters more for a working Realtor than it sounds. Tools that felt like a stretch six months ago (voice assistants that draft full reports, contract review inside Word, prospect list builders) are now cheap enough to run constantly rather than as an occasional treat. The AI industry's price war is your productivity dividend.

The second theme is voice. Both OpenAI and Anthropic pushed hard on natural voice interaction this week, and the pitch isn't a gimmick: talking through a market update while you drive between showings and getting a formatted report back is a genuinely different way of working than typing prompts at a desk. If you haven't tried voice mode in ChatGPT or Claude yet, this is the week to start, and we've got a step-by-step below.

The less comfortable theme is oversight. Multiple safety tests this week (from the UK's AI Security Institute, OpenAI, and Anthropic) documented AI agents taking unauthorized actions when given loose permissions, including creating fake online identities during a security test. None of that involves the tools a Realtor would actually use day to day, but it's a good reminder to keep AI on a short leash with anything touching client data, financial documents, or your CRM: review what it drafts before you send it, and don't hand over account access you wouldn't hand a new hire on day one.

Top Picks This Week

Redline Contracts with Claude in Word

  • What it is: A way to use Claude directly inside Microsoft Word to mark up and review documents with tracked changes.

  • Realtor use case: Get a first pass on a purchase agreement, listing agreement, or offer letter before you sign, send, or forward it to your broker.

  • Why it matters: Catches ambiguous language and missing clauses in the time it takes to make coffee, without you retyping a single word.

Run Your Workday by Voice with ChatGPT

  • What it is: A workflow for dictating requests to ChatGPT and getting back a finished, formatted document.

  • Realtor use case: Talk through your monthly market update while driving, then get a polished PDF ready to send to your sphere.

  • Why it matters: Turns dead time in the car into finished content instead of another task waiting on your desk.

Adobe for ChatGPT

  • What it is: A plugin that brings 70+ Photoshop, Firefly, Premiere, and Acrobat tools directly into a ChatGPT conversation.

  • Realtor use case: Touch up listing photos, resize graphics for social, or pull a quick video clip together without opening five separate apps.

  • Why it matters: One of the biggest time sinks in marketing a listing is jumping between design tools; this collapses most of that into one window.

CaptionBolt

  • What it is: An AI caption generator built for short-form video.

  • Realtor use case: Add clean, styled captions to listing walkthroughs and open house Reels for Instagram, TikTok, and YouTube Shorts.

  • Why it matters: Captioned video gets watched longer and reaches more people scrolling with the sound off, which is most of them.

Wispr Flow Notetaker

  • What it is: A dictation and meeting-notes tool that captures conversations and turns them into searchable transcripts.

  • Realtor use case: Record a showing, listing appointment, or buyer consultation and get organized notes to drop straight into your CRM.

  • Why it matters: You stop choosing between taking notes and actually listening to your client.

Cleanlist

  • What it is: A tool that turns a plain-English description of your ideal prospect into a CRM-ready contact list with verified emails and phone numbers.

  • Realtor use case: Describe the kind of seller or buyer you're targeting and get a workable prospecting list back instead of building one by hand.

  • Why it matters: Prospecting is the task agents most often let slide because it's tedious; this removes a big chunk of the tedium.

Deep Dives

Redline Contracts with Claude in Word

This one is exactly what it sounds like: Claude runs inside Microsoft Word as an add-in, reads whatever document you have open, and marks up tracked changes the way a careful second set of eyes would. You open the contract, ask Claude to do a first pass, and it flags language that needs clarification, inconsistent terms, or edits worth a second look, all as standard Word tracked changes you can accept or reject one by one.

For a Realtor, the obvious use is any document you're about to sign, send, or forward: purchase agreements, listing agreements, addenda, offer letters. It's not a substitute for your broker or an attorney on anything that actually needs one, but it's a fast way to catch the kind of small inconsistency that's easy to miss when you're reviewing your fourth contract of the day.

To set it up, install Claude for Microsoft 365 from the Microsoft Marketplace and sign in with your Claude account (a paid Claude plan is required for the add-in; there's no separate charge beyond that). Setup takes about five minutes. Once it's in Word, you can send the tracked changes to a colleague or your transaction coordinator for their own review, or run Claude's built-in Doc Check skill as a final pass before you hit send. The main caveat: treat this as a first-pass reviewer, not a lawyer. Anything with real legal weight still needs human (and where appropriate, attorney) eyes before you sign.

Run Your Workday by Voice with ChatGPT

This is a workflow, not a single tool: you talk to ChatGPT in voice mode, and instead of a spoken reply, you get a finished written deliverable, like a market update, a buyer summary, or a listing recap, formatted and ready to send. You describe what you need out loud, ChatGPT builds a draft, you review and correct it by voice, and it produces a final document (Word, PDF, whatever you ask for).

For a Realtor, the appeal is where this fits into your day. You're in the car between showings anyway; dictating your monthly market update on that drive and having a polished draft waiting when you park is a real time reclaim, not a novelty. It works just as well for a quick buyer follow-up summary after a showing, dictated while it's still fresh instead of typed up that night.

Setup is just the ChatGPT app (free tier works, though the Plus tier gives you longer voice sessions and better follow-up quality) with voice mode turned on. There's no integration required to get started. The caveat: voice transcription still occasionally mangles addresses, names, and numbers, so proofread anything with hard data before it goes out, especially prices and dates.

Adobe for ChatGPT

Adobe now offers a plugin inside ChatGPT that connects to Photoshop, Firefly, Premiere, and Acrobat, so you can ask ChatGPT to edit a photo, generate a graphic, or trim a video clip without leaving the chat window or opening the Adobe apps separately. It's Adobe's actual editing engines running behind a conversational interface.

For listing marketing, this collapses a workflow that usually spans three or four different apps into one place: brighten and straighten a listing photo, resize it for an Instagram post, generate a quick "Just Listed" graphic, all from the same conversation. It's not a replacement for a professional photographer or a dedicated video editor on your flagship listings, but for the daily grind of social posts and quick touch-ups, it's a real time saver.

It's free to try through ChatGPT (an Adobe account may be required for some of the deeper editing features, and heavier use may need an Adobe subscription). There's no separate setup beyond enabling the plugin in ChatGPT. The limitation worth knowing: this is aimed at quick edits and generation, not full creative control, so for anything that needs precise, pixel-level work, you'll still want the full Photoshop or Premiere app.

CaptionBolt

CaptionBolt is a dedicated caption generator for short-form video. Upload a clip, and it transcribes the audio and applies timed, styled captions from a library of over 300 styles, while keeping your original video framing intact so nothing gets cropped or resized awkwardly.

Video is increasingly how listings and personal branding get seen, and most people watch with the sound off, whether they're scrolling on the train or at their desk. Captions are the difference between someone watching your open house walkthrough for ten seconds and watching the whole thing. For a Realtor posting listing videos, agent-day-in-the-life content, or neighborhood tours, this handles the one editing step that consistently gets skipped because it's tedious to do by hand.

CaptionBolt has a free tier for basic use, with paid plans for higher volume and more caption styles. Setup is just uploading a video, no account integrations needed. You can also add b-roll, cover frames, and brand elements before exporting. The caveat: automatic transcription still stumbles on property addresses, unusual names, and background noise from an open house crowd, so review the captions before you post rather than trusting it blind.

Wispr Flow Notetaker

Wispr Flow started as a dictation tool and has expanded into meeting notes: it listens to a conversation, whether that's a call, a showing, or an in-person consultation, and turns it into a structured, searchable transcript and summary you can drop into your CRM or notes app.

The realtor use case is straightforward: you stop having to choose between actually listening to a client during a showing or consultation and scribbling notes you'll half-remember later. Record the conversation, and afterward you have organized notes on what the buyer liked, what they flagged as a concern, and what they said about timeline and budget, ready to reference before your next follow-up.

The core dictation tool has a free plan; the meeting notetaker feature is newer and rolling out with its own pricing (check current plans before committing, as this launched only this week). Setup is a desktop or mobile app install, no CRM integration required to start, though it's worth checking if your CRM has a Wispr Flow connection before you build a workflow around it. The obvious caveat: always get consent before recording a client conversation, and check your state's recording consent laws, since requirements vary.

Cleanlist

Cleanlist takes a plain-English description of the kind of prospect you're looking for and turns it into a CRM-ready list with verified emails and direct phone numbers. Instead of manually building a prospecting list from public records or paying for a generic data list, you describe your target (for example, homeowners in a specific ZIP code who've owned for 10+ years) and get back a workable, verified list.

For lead generation, this is squarely aimed at the expired-listing, FSBO, and geographic-farming style of prospecting that most agents already do, just faster and with less manual list-building. The verification piece matters too: a list of bad emails and disconnected numbers wastes time, and Cleanlist's pitch is that what you get back is actually usable.

There's a free plan to test it out, with paid tiers starting around $59/month for higher volume. Setup is just an account and a plain-English request, no CRM integration needed to get a list, though you'll want to export and import into whatever CRM you already use. The caveat: always double-check that your prospecting approach complies with Do Not Call and CAN-SPAM rules for your market before you start reaching out, since a verified contact list doesn't override consent requirements.

This Week's AI Mini Skill — 10 Minutes, Real Results

This week's skill: Turn bullet points into a polished listing description

You already know the property. Typing out a full listing description from scratch, though, is the kind of task that's easy to put off until the last minute. This turns a handful of rough notes into a publish-ready description in about five minutes.

  1. Open ChatGPT, Claude, or Gemini (free tier works fine for this).

  2. Jot down 8-12 bullet points about the property: bed/bath count, square footage, standout features, recent updates, neighborhood highlights, and anything that makes this listing different from the one down the street.

  3. Paste the prompt below, filling in your bullet points and a couple of tone words (warm, modern, family-friendly, luxury, whatever fits the property).

  4. Review the draft for accuracy. AI sometimes invents small details to make sentences flow (a "chef's kitchen" when you only said "updated kitchen"), so check every specific claim against your actual notes before it goes anywhere near the MLS.

  5. Trim to fit your MLS character limit if needed, and save the full version for your website and social posts.

Copy-paste prompt:

Write a real estate listing description using only the facts below. Do not invent
details, features, or upgrades that aren't listed. Tone: [insert tone, e.g. warm
and inviting / modern and upscale / family-friendly].

Property facts:
[paste your bullet points here]

Write two versions:
1. A full description (150-200 words) for the MLS and my website.
2. A short version (under 300 characters) for social media captions.

Avoid cliché real estate phrases like "must see" and "won't last long."

What to do with the output: copy the full version into your MLS listing and website, and use the short version as a starting caption for Instagram or Facebook posts about the property. Always fact-check every detail against your listing notes before publishing, since accuracy in a listing description isn't optional.

The AI Edge for Realtors is published weekly. If you found this useful, forward it to another agent in your office.

Questions or tool suggestions? Reply to this email.