All articles
Sales
October 20, 20268 min

AI Lead Qualification for Real Estate Sales Teams: Portal and Paid Social Enquiries

Most property sales teams are not short of enquiries. What they are short of is a quick way to tell which of those enquiries deserve a call today. A lead form from a social ad, a portal enquiry, a WhatsApp or Telegram message, an Instagram DM asking "price for a two-bed?" — they all land on the same few people, and a large part of the day goes on asking the same qualifying questions over and over.

This piece looks at how an AI agent can take on first-line lead qualification for a developer or an estate agency: what it asks, how it decides who goes to a sales consultant, and where it should be kept on a short lead.

UNIKA Solutions builds AI automation: lead handling, CRM workflows and agents. Running ad campaigns is the job of our sister brand Unika Agency. This article is only about what happens to an enquiry once it has arrived.

Why property enquiries are hard to triage

Real estate has a few traits that make lead handling heavier than in most sectors:

  • A long decision cycle. Weeks or months can pass between the first question and a reservation. Someone can be keen on a project and still be a year away from buying.
  • Many channels at once. Property portals, paid social lead forms, the website, messaging apps, Instagram and phone calls. The same buyer often turns up in two or three of them.
  • Thin data. An instant lead form usually gives you a name and a number. You do not know whether the person is an end user or an investor, what the budget is, or which area they have in mind.
  • The same questions all day. Price per square metre or square foot, handover date, payment plan, mortgage options, floor plans, finishing — your team answers these dozens of times a day.
  • More than one language. Buyers write in whatever language suits them, and in markets like Tashkent or the Gulf that can change mid-conversation.

The result is predictable: your best people spend time on browsers, while the buyer who wants a viewing this week waits until the evening for a reply.

What qualification means — and which parts an AI can own

Qualifying a lead means answering one question: should someone call this person now, and who? In property it usually rests on a handful of points:

  • Purpose: a home to live in, a rental investment or a resale play.
  • Budget and payment: cash, a developer payment plan or a mortgage.
  • Timeline: this month, within six months, or just looking.
  • Property: area, number of bedrooms, size, ready or off-plan.
  • Next step: a viewing, a visit to the sales centre, a call.

Most of this can be gathered in a short written exchange. That is what the AI agent does: it holds a brief conversation, asks one clarifying question at a time, answers routine questions from your approved knowledge base and records the answers on the lead. Pricing decisions, discounts, contract terms and the viewing itself stay with your people.

How it works, from enquiry to sales consultant

The flow we use in AI agent projects looks like this:

  1. The enquiry arrives — from a lead form, the website, a messaging app or a portal notification, depending on which channels can be connected technically.
  2. The agent makes first contact in a channel that suits the buyer, confirms which project they asked about and offers a couple of quick questions to narrow down options.
  3. It clarifies the brief — purpose, budget, payment method, timeline, area — conversationally, not as a ten-field form.
  4. It handles routine questions such as handover dates, payment plans and layouts, using only material you have signed off.
  5. It assigns a status, for example "ready to view", "interested, longer timeline", "not a fit" or "needs a person".
  6. It hands over to the CRM — amoCRM, Bitrix24 or any system with an API — with the answers and full conversation attached, so the consultant knows exactly where things stand.

The handover rules are the part that matters most, and they should be agreed with your head of sales up front: which combination of answers sends a lead straight to a consultant, which ones stay in nurture, and when the agent must bring in a person immediately — a complaint, an unusual deal, or a simple "can I speak to someone?".

Portal leads versus paid social leads

These two sources behave differently, so it pays to give them separate scripts.

Portal enquiries

The buyer came from a specific listing, so they already have context: the unit, the price, the location. The agent should confirm that listing straight away, answer the question about it and suggest close alternatives if it is not quite right. These enquiries tend to be nearer to a viewing.

Paid social enquiries

Instant forms take one tap, and some people barely remember submitting one. The first message needs to remind them which project it was, and qualification usually takes a few more steps. This is also where the agent saves your team the most time: it filters out the poor fits and surfaces the people who are genuinely looking.

Both need de-duplication. If someone enquires on a portal and then messages you on Telegram or WhatsApp, the CRM should hold one record, not two.

What this looks like in practice

We have a case with the same mechanics, though not from property. For SAYF DESIGN we set up an AI consultant for Instagram Direct and Messenger. It talks to people in their own language, clarifies what they want and, once a phone number is shared, creates a lead in amoCRM with the conversation history attached. From the second week, enquiries rose noticeably and their cost came down. The architecture is set out in the case study.

The key design decision carries straight over to real estate: a lead is created only once there is a contact and a clear intent in the conversation, not for every "hello". Your consultants work on enquiries, not noise.

Where a sales team should start

  1. Measure where you are. Weekly enquiries per channel, how many get a reply within the first hour, and what share turn out to be poor fits. To put a rough figure on slow replies, try the lost revenue calculator with your own numbers.
  2. Define a qualified buyer. Which answers on purpose, budget and timeline mean "call today".
  3. Build the knowledge base. Current prices, floor plans, payment plans and handover dates — with one named owner who keeps it up to date.
  4. Start with one channel. Say, paid social forms only, or out-of-hours only. Compare against how things worked before.
  5. Read the conversations. In the first weeks, spot-check them daily. Most issues are fixed by editing the knowledge base or the rules, not the code.

Frequently asked questions

Will an AI agent replace our sales consultants?

No. It covers first contact, clarifying questions and routine answers. Viewings, negotiation, objections and the deal itself stay with your team. The point is that consultants spend their time on people who are ready to talk.

Can it pick up enquiries from property portals?

That depends on the portal. Some provide exports or notifications that can be processed; with others, enquiries arrive as calls or direct messages. We check this at the start, before any build work.

What if a buyer asks something that is not in the knowledge base?

The agent must not invent a price, a discount or a handover date. The right behaviour is to say a consultant will confirm, and pass the conversation to a person with a note in the CRM.

How much does it cost?

It depends on the number of channels, your CRM, enquiry volume and how ready your knowledge base is. The first conversation is there to work that out. If you would like to talk through your process, tell us about it.

Subscribe to AI Insights

Case studies, tips and AI automation trends — weekly

Want similar results?

Get a free consultation and ROI calculation for your company

Free Consultation
AI Lead Qualification for Real Estate Sales Teams | UNIKA