AI-Generated Cold Call Scripts for UAE Real Estate (How to Use Them)

Most UAE brokers still rely on the same tired scripts that get cut off before the second sentence. In 2026, the brokers who consistently book viewings in Marina, Business Bay and JLT are the ones using AI to create calls that sound like a natural conversation rather than a sales pitch. This article shows exactly how to prompt an AI tool to generate cold-call scripts that respect RERA rules, reference current AED price bands, and actually get past the gatekeeper.

Why generic scripts fail in the UAE market

Buyers and tenants in Dubai and Abu Dhabi receive multiple calls every week. When the script opens with “I have a great opportunity,” the listener already knows what comes next. Local buyers check Bayut and Property Finder daily, so they can spot inflated claims in seconds. A script that mentions specific towers in JLT or current AED 1.8–2.4 million two-bedroom ranges in MBR City feels researched rather than robotic.

Key prompt elements that produce usable scripts

To move beyond generic output, include these five details in every prompt:

The exact neighbourhood and building type (e.g., Marina waterfront or Saadiyat Island townhouse).
Current price band in AED and payment plan structure.
Buyer profile (end-user family, investor seeking 7–8 % gross yield, or first-time tenant relocating from Abu Dhabi).
Relevant regulations: RERA escrow rules, DLD transfer fees, or DEWA connection timelines.
Call objective: secure a 15-minute discovery call, not an immediate viewing.

Adding these constraints forces the model to generate language that references real market conditions instead of vague benefits.

Sample prompt structure for a Marina apartment call

Use the following template and replace the bracketed sections with live data before each session:

“You are an experienced Dubai broker calling a UK-based investor who viewed a two-bedroom unit in Marina Mall Tower last month.”
“Current asking price is AED 2.15 million with a 60/40 payment plan. Service fees are AED 18 per sq ft.”
“Mention that RERA-registered escrow is already in place and DLD transfer can complete in 14 days.”
“Goal: book a 15-minute call to review comparable transactions on Bayut from the last 30 days.”

Run this prompt in ChatGPT or AgentsAI and you receive a 60–75 second script that includes pauses and natural follow-up questions.

Refining the output for tone and compliance

After the first generation, run a second prompt that asks the model to:

Remove any guarantee language that could breach RERA advertising rules.
Replace “best deal” with “current median price per square foot for similar units.”
Insert one relevant local landmark (e.g., proximity to Sheikh Zayed Road or Aljada community facilities).
Shorten sentences so the entire opening fits inside 18 seconds.

This second pass typically reduces script length by 25 % while increasing the number of booked calls in our experience.

Testing and iterating with live data

Record the first ten calls made with the new script. Note which lines trigger questions and which cause immediate hang-ups. Feed those observations back into the prompt as negative constraints (“avoid mentioning capital appreciation forecasts”). After three iterations, most brokers report that the same script works across JLT, Business Bay and MBR City with only minor neighbourhood swaps. Keep a shared folder of approved scripts so the whole team can update them when Etisalat or DEWA connection times change.

Stop typing. Start closing.

Generate property listings, follow-up emails, WhatsApp templates, and CMA reports in seconds. Free tier: 5 generations/month, no card needed. Try AgentsAI free →