A real estate agency lives on replying fast and not losing anyone along the way. That's exactly what gets hard when enquiries arrive through the portals, your website, the phone and WhatsApp, when many of your buyers are abroad and writing at odd hours, each agent runs their listings their own way, and owners want to know how things are going. We build software and AI for real estate agencies that sits on top of the tools you already use and brings everything onto one screen.
We don't have a published case with an agency yet. What follows are the problems we know how to solve and what we would build, with the examples labelled as examples.
What slips through the cracks at an agency every week
Leads arrive through too many doors
Idealista, Fotocasa, your own site, phone calls, WhatsApp, email. Each channel has its own inbox and none of them tells the right agent. The buyer who wrote on Saturday from another time zone hears back on Tuesday, if at all.
Each agent's pipeline lives on their phone
Follow-ups live in their head, their spreadsheet or their chats. When they're on holiday, nobody knows who needed a call. When they leave, their contacts leave with them.
Owners want to know how it's going
How many viewings, what people said, whether the price needs another look. Writing that update for each owner takes longer than it should.
Every property's paperwork is scattered
Land registry extract, energy certificate, photos, floor plans, the listing agreement. Some in email, some in Google Drive, some in a folder only its creator can find.
What we build for a real estate agency
We don't start from scratch: the engines have been running for years. What we build for you is the part that is yours.
One screen for leads, listings and viewings
We connect the channels your enquiries come through with your listings and calendar. Each lead reaches the right agent with its history, and management sees in one place what's coming in, what's moving and what has been sitting still for days. That's our systems integration: you don't change tools, you stop chasing them.
A custom CRM, or yours properly connected
If you already have a real estate CRM, or a general one like HubSpot, and it works, we leave it alone and connect it to whatever is left out today. If what you have is a shared spreadsheet, we build the one that's missing, with your stages and your way of working. More on custom CRM development.
Replies in minutes, on WhatsApp, in the buyer's language
An assistant that answers the usual questions with each listing's information: price, size, community fees, whether pets are allowed. It suggests viewing times and hands you what needs a person. Behind it is an AI agent that understands what people write, not just which buttons they press.
Client paperwork without retyping
A buyer's or tenant's documents arrive by email or WhatsApp, the system sorts them and extracts the data, and a person reviews it before it enters your system. Stop typing, start approving.
Owner updates that write themselves
Viewings, feedback, enquiries and status for each property, in a summary built from what's already in the system. You check it and send it.
A typical real estate project
For example: an agency in Barcelona with a few agents, listings on the portals, many buyers writing in English from abroad, a CRM nobody fully updates and a WhatsApp group per development. The first project would be the leads-and-listings screen, connected to what they already use. Once that's working, it shows what comes next: if half the messages are the same questions, the WhatsApp assistant; if every owner update eats an afternoon, automating it.
If you also manage holiday or mid-term rentals, have a look at our work in holiday rental management, where we do have a real project under way.
How we work with agencies
We start with what's bothering you, not with which AI you want. We look at who does what, which tools they use and where enquiries get lost. You get a proposal: what we'll build, when, and for how much, fixed and in writing. We build the smallest useful version, your agents use it for real, and we improve it from there.
People stay where they matter: showing a flat, negotiating, closing. The software sorts, flags and answers the routine questions.