The Weekly Snapshot Habit

National noise, local questions

Every week brings a new headline about rates, prices or a "frozen" market. Your sphere sees it, and the first question they ask is some version of "should I be worried?" The national story rarely answers that. What answers it is what's happening on their streets.

That's the opening. The agent who calmly explains what the numbers look like locally, every single week, becomes the person people trust when they're finally ready to move. The hard part has always been time. That's where AI earns its place.

Five numbers, three posts

The weekly snapshot is simple on purpose. Pick one area you want to be known for, then pull the same five numbers from your MLS every week: new listings, homes that went under contract, closed sales, median days on market and how many listings had a price reduction.

Same five numbers, same day, every week. Consistency is what makes it work: your audience starts to expect it, and you start to see the trends before anyone else does.

Let it draft, you decide

Free tools like ChatGPT, Claude and Gemini are good at one thing in particular: turning a handful of facts into clear, readable copy. Give one your five numbers and last week's five, and in a minute you have a first draft of everything you need:

Copy these prompts

Paste each one into ChatGPT, Claude or Gemini, fill in the brackets, and you're done. Save them in a note on your phone so they're ready every Monday.

You're helping a real estate agent write a weekly local market update.
Use ONLY the numbers I give you. Do not add statistics, predictions or mortgage rates.
Tone: calm, confident, plain English, no hype.

Area: [neighborhood or zip code]
This week: new listings [#], went under contract [#], closed [#], median days on market [#], listings with a price cut [#]
Last week: new listings [#], went under contract [#], closed [#], median days on market [#], listings with a price cut [#]

Give me:
1. A 60-second video script I can read to camera
2. A 5-slide carousel outline, one short line per slide
3. A 2-sentence Instagram caption that ends with a question

Here are 3 captions I've written before:
[paste caption 1]
[paste caption 2]
[paste caption 3]

Rewrite the script, carousel and caption above so they sound like me.
Match my sentence length, my energy and the way I talk to my audience.

Review everything above before I publish it. Flag:
- Any number that wasn't in the list I gave you
- Any prediction about prices or rates
- Any wording that describes who a neighborhood is "for" or "ideal for"
List what you flagged and suggest a fix for each.

Then turn the same draft into a short text for your top 25 contacts: "Turn the caption into a friendly 2-sentence text message."

The sentence AI can't write

Here's what separates a snapshot people skim from one they save: one line that only you could have written. "I walked three listings in Kirkland Heights this week, and every seller was open to talking concessions." AI can summarize the numbers. It can't stand in a kitchen.

Fast is good. Careful is better.

• Keep client info out. Never paste names, addresses, financials or documents into a public AI tool.
• Check every number. If it didn't come from your MLS or a source you can name, it doesn't go in the post.
• Review for fair housing. Read the draft for any language about who a neighborhood is "for." You're responsible for what you publish, not the tool.
• Leave the mortgage math to your lender. Keep rates, payments and loan terms out of AI-drafted posts and send those questions to your loan officer.

Your first snapshot, start to finish

1. Pick your area. One neighborhood or zip code you want to be known for.
2. Pull the five numbers from your MLS for this week and last week.
3. Run Prompt 1 for your script, carousel outline and caption.
4. Run Prompts 2 and 3, then add your human line.
5. Post it, then send it as a short text or email to your top 25 contacts.