AmuraAMURA Software
Service · AI agents · Real estate & property

AI agents for real estate that don’t let a lead go cold.

Lead qualification from portals, viewings and availability agent, multilingual owner support and tenant-incident triage, all wired into your CRM and the portals you already use.

The workflows, examples and figures on this page are illustrative composites and modelled targets, not measured client results. In a real project, we define the baseline, thresholds and human review with your data before rollout.

What we solve

A real-estate lead that takes two hours to get a reply is already lost.

Idealista, Fotocasa, your own site, WhatsApp, missed calls. Leads arriving out of hours, in five languages, mixing real intent with casual browsers. Your agents spend more time copying data between the portal and the CRM than closing viewings. Meanwhile the international owner is asking about the contract and the tenant in flat 3C is reporting a leak, and nobody hears about it until Monday.

We design agents that qualify the lead the moment it lands, propose viewings against the agent’s real calendar, reply to owners in their language with the property’s official information and prioritise tenant incidents by urgency. When a case calls for human judgement, negotiation, offer, conflict, they escalate to the agent or property manager with the history already organised.

Implementation contract

How it runs in production

Workflow and actors

The agent receives the lead, checks the current listing, qualifies it and logs it in the CRM; for viewings it checks calendars and availability, and for incidents it creates the work order and alerts the right manager.

Systems and data

It works across authorised portal feeds, CRM, calendars, property listings, contracts, maintenance and KYC documents, with access separated by office, portfolio and role.

Exceptions and risks

Negotiations, offers, financial decisions, regulatory uncertainty, sensitive documents and emergencies outside the rules escalate. It does not confirm price, licence status or availability without a current source.

Human review

The salesperson retains negotiation and offer acceptance; the property manager decides on owner and incident matters, and legal reviews KYC, contracts and regulatory rules before any commitment.

Implementation pattern

We combine event-driven triage, retrieval from approved listings and documents, scheduling and urgency rules, and bounded actions while preserving traceability and portfolio isolation.

Relevant integration

The implementation uses available portal feeds or professional areas and authorised CRM, calendar and maintenance APIs. We validate the actual capabilities and permissions of each account.

What we build for this sector

Use cases that ship to production.

See full catalogue →
Sales

Lead qualification from portals

Picks up leads from Idealista, Fotocasa, web and forms, replies in under 60 seconds with the real listing, drops the unqualified and logs the rest in the CRM with scoring and reason.

Modelled target: first response < 60 s · 24/7
Commercial

Viewings and availability agent

Cross-references lead intent with the agent's real calendar and property availability, proposes three concrete slots, confirms and sends a reminder. Only re-books when a person is needed.

Modelled target: +34% confirmed viewings
Owners

Multilingual owner support

EN, ES, DE, FR and IT by default. Answers questions about contracts, settlements, maintenance and reports, citing the document source and escalating to the manager whenever there's a financial decision.

Modelled target: 5 languages · reply < 2 min
Operations

Tenant incident triage

Reads the incident, decides whether it's urgent (leak, electrical, safety), opens a work order in the maintenance system, assigns the trade if pre-approved and notifies the property manager with a summary.

Modelled target: < 3 min to assignment for urgent
Documentation

KYC and contract document assistant

Collects KYC documentation, validates format and validity, pre-fills contracts and addenda against the approved template and leaves the legal team only the final review, not the transcription.

Modelled target: −58% time on document prep
Illustrative composite scenario · modelled figures and targets, not client results

An agency managing 450 properties.

Coastal mixed sales-and-rental agency with international portfolio, 14-strong commercial team and CRM connected to Idealista and Fotocasa. Illustrative composite scenario; the figures below are a modelled baseline and targets, not measured client results.
Modelled baseline

Baseline assumption: average first reply time to a portal lead of 2 h 40 min during office hours, with no reply until the next day outside office hours. Each agent would spend roughly 90 minutes a day copying data from portals into the CRM and coordinating viewings on WhatsApp. Weekend incidents would pile up until Monday.

Modelled target

Modelled operating target: the agent would reply to portal leads in under 60 seconds, in the lead's language and with the listing detail. It would qualify and log them in the CRM with scoring, propose real viewing slots and aim to limit pre-confirmation agent intervention to 22% of cases. Urgent weekend incidents would be routed to the on-call trade in under 3 minutes.

Modelled target: −81% average first response time on leads
Frequently asked

What clients ask us

  • 01

    Does it integrate with our CRM and the portals (Idealista, Fotocasa)?

    Yes. We work on top of HubSpot, Salesforce, Pipedrive, Inmovilla, Witei, Sooprema and the most common real-estate CRMs via API. Idealista and Fotocasa come in via feed or their pro area depending on your account. You don't have to change CRM or portal to start.

  • 02

    How does the agent handle owner and tenant data?

    Data sits on EU infrastructure, with isolation per client, GDPR applied from the start and no model provider training on your data. We define the retention policy with you and treat sensitive data (ID, IBAN, KYC payslips) with extra encryption and restricted access.

  • 03

    Can it make pricing decisions or close deals on its own?

    No. The agent qualifies, informs, schedules and prepares documentation, but any negotiation, counter-offer or offer acceptance always stays with the agent or the manager. We configure strict limits on what it can commit to, and every action is logged.

  • 04

    What about urban planning, tourist licences and rent-controlled areas?

    The agent knows the commercial policy you set: if a property is in a rent-controlled area, without a current tourist licence or under a price cap, it reflects that in its replies and does not schedule viewings that breach the rules. These rules are loaded at the start and updated with your legal team.

Trust

Safe, traceable AI,
enterprise-ready.

We design for privacy from the start, human control, traceability, usage limits, permissioning and documentation. For sensitive processes, we help assess risk and applicable obligations under GDPR and the EU AI Act.

  • 01We never train models on your data without explicit authorization.
  • 02Human review built-in for processes where risk demands it.
  • 03Traceability: prompts, sources, permissions, errors and metrics, all documented.
  • 04Privacy, security and control integrated from day one.
  • 05Solutions engineered to be maintained, audited and improved over time.
GDPREU AI ActAEPDISO 27001 readyEU data residency
Personal diagnosis

We work with
few clients.

Every engagement is led personally by one of the partners. If there's a fit, you get a personal first read of your case within one business day, not a canned demo.

How we work
  1. 01Tell us which process eats your time
  2. 02Personal reply within one business day
  3. 0320-minute call, no demo, no pitch
Start the conversation →