Human-centered design systems and product experiences
Core services
01UI/UX Design02Web Development03Software Development042D/3D Experiences05App Development06AI Agents & Automation07Website DesignTechnical services
Frontend Engineering
React DevelopmentNext.js DevelopmentWeb ApplicationsEnterprise DashboardsSaaS PlatformsBackend & Infrastructure
API DevelopmentCloud InfrastructureDevOps & CI/CDBackend SystemsCustom SoftwareIndustries we serve
Technology & Software
AI Real Estate Matching Agent Development
The right homes for each buyer
An agent for the first conversation with a property enquirer.
An agent for the first conversation with a property enquirer. It establishes what they need and can afford, matches that to your current listings, explains why each property fits from facts in the listing, and passes a qualified enquirer to the right agent.
Property enquiries arrive from portals at all hours, most about a listing that is already gone. The enquirer asked about one flat. What they need is a two-bedroom place near a school under a certain rent, and you have three. Finding that out takes a conversation nobody has time for at eleven at night, so the lead gets an automatic reply and goes to the next agency.
This page describes a workflow we propose to build as a custom project. It does not exist as software yet. It would reuse the question-by-question qualification and the CRM record from our lead qualification demonstration, and the catalogue lookup from our conversation demonstrations. Reading listings, matching requirements to properties and portal connections are not built.
Six Things That Slow This Down
Enquiries about gone listings
The advert is still on the portal and the property is let.
One property asked about, another needed
The enquirer's real requirements never come out.
Leads answered the next day
By then they have booked a viewing elsewhere.
Descriptions that oversell
"Close to transport" for a twenty-minute walk.
Budget never checked
Viewings are arranged for people who cannot proceed.
Sensitive criteria
Requests that would discriminate between applicants.
Six jobs it carries through
01Requirement capture
Finds out what the enquirer needs.
- Buy or rent, area and budget
- Bedrooms, features and timing
- Must-haves separated from preferences
- Asked one question at a time
02Listing match
Matches needs to live listings.
- Reads current availability
- Filters on must-haves
- Ranks on preferences
- Says what ruled a listing out
03Explained matches
Says why each property fits.
- Two or three properties, not twenty
- The facts from each listing
- Where it falls short
- A link to the full listing
04Readiness check
Establishes whether the enquirer can proceed.
- Timeline and position
- Finance in principle, as stated by them
- Documents your process requires
- Recorded in their own words
05Agent handover
Passes the enquirer to a person.
- To the agent for that area
- With requirements and matches
- With a suggested next step
- A viewing request, if they ask
06Fair treatment
Applies the same process to everyone.
- The same questions for every enquirer
- No filtering on personal characteristics
- Refuses discriminatory criteria
- A record of what was asked and shown
What We Need to Scope This
- The process as it runs today, in a few lines.
- The systems it touches, for example listings, portals, CRM, channels.
- Rough monthly volume, and who approves what.
How One Request Moves Through
An example of one case, in 6 steps. The same steps play in the picture at the top of this page.
Enquiry arrives
Late in the evening a portal lead asks about a flat that was let last week.
Needs established
The agent says so, then asks about area, budget, bedrooms and move-in date.
Listings matched
Three available flats meet the must-haves. One exceeds the budget slightly and is marked.
Matches explained
Each is described from its listing, including the one drawback the listing shows.
Readiness noted
The enquirer says they can move next month and have references ready.
Handed over
The lettings agent for that area receives the enquirer, the requirements and the shortlist.
Where It Connects
The systems named here are the kind this agent works with. Which ones it connects to in your case is settled in the scope.
Listings
- Your property management or listing system
Portals
- Lead feeds from the portals you advertise on
CRM
- Enquirers, requirements and history
Channels
- Website chat
Agents
- Who covers which area and property type
Policy
- Fair housing and anti-discrimination rules where you operate
Facts From the Listing, the Same Process for Everyone
The agent describes a property only with what its listing states, and says where a match falls short. It does not value a property, advise on a price or negotiate.
Housing is regulated against discrimination in most places. The agent asks every enquirer the same questions, never filters on personal characteristics and declines criteria that would do so.
- Only currently available listings
- Descriptions from listing facts
- Shortfalls stated
- No valuation or negotiation
- The same questions for everyone
- Handover to a licensed agent
Asked Before the First Call
Lead qualification decides whether an enquiry fits your business. This agent goes a step further for property: it matches the person's requirements against your live inventory and explains the matches, so the human agent starts with a shortlist.
Matching ends when the enquirer has a shortlist and an agent. The viewing agent takes over from there: arranging access, reminders and what happens after the visit.
No. It reports what a listing states and what the enquirer says they need. Valuation, advice and negotiation are regulated activities in many markets and stay with your licensed agents.
Not as software yet. This page describes a proposed workflow. It would reuse the question-by-question qualification and the CRM record from our lead qualification demonstration, and the catalogue lookup from our conversation demonstrations. Reading listings, matching requirements to properties and portal connections are not built. A project starts by mapping your process, then proving the first steps on test data before anything touches a live system.
Look at Last Month's Portal Leads
Count how many asked about a property that was already gone, and how many of those were offered something else.
