How a real estate team automated auction sourcing and legal document analysis, and reviewed 5× more properties without adding people.
RE Innovation is a real estate team in Spain that buys properties through bank foreclosure auctions.
Before bidding, the team has to understand exactly what they are buying: ownership history, outstanding debt, legal issues and the risks attached to the property.
Most of that work was done manually. Properties were tracked in Excel and Google Sheets, while the documents for each auction were opened and reviewed one by one.
Every new property meant another set of documents to go through. Ownership records. Existing debt. Court documents. Legal risks.
An analyst had to open the files, read them and piece together the situation before the team could decide whether the property was worth bidding on. A single property could take around 10 hours to review.
And auctions don't wait. In some cases, the team was still analysing the property when the deadline closed.
10 hours per property. One analyst. One property. Documents reviewed by hand.
Auctions closed during review. Sometimes the deadline arrived before the analysis was finished.
The data lived in sheets. Excel, Google Sheets, separate documents and team notes.
The problem wasn't finding properties. It was processing them fast enough to make a decision.
The obvious idea was to use AI to read the documents.
But that only solved one part of the job.
Someone would still have to find the new property, copy the details into a sheet, find the right documents, upload them for analysis and then move the result somewhere the team could actually use it. You'd save time reading PDFs and keep the rest of the manual process.
So we stopped looking at document analysis as a separate task and looked at the whole process — from a new auction appearing on the bank's website to the team deciding whether to bid.
Every morning, the system checks the auction sources for new properties. When it finds one, the property is added to the review pipeline and the relevant auction documents are downloaded automatically.
Those documents are then sent to Gemini for the first pass. It extracts the information the analysts repeatedly look for — ownership details, debt, legal issues and other risks — and prepares a structured summary. The summary is saved with the property in the team's CRM.
So when an analyst opens a new property, they don't start with a folder of raw PDFs. They start with the property details and a first structured review already in front of them.
The team still decides whether to bid. AI just gets them to that decision much faster.
AI didn't make the investment decision. It removed the work that slowed the decision down.
The first project was supposed to solve property review.
It did.
But once the team stopped spending hours reading documents, they started noticing all the other repetitive work around every auction: creating folders, moving files, updating statuses, sending internal updates and archiving finished properties. Around four hours of manual work were still going into this every day.
RE Innovation came back and asked us to automate the rest of the workflow.
Full workflow on autopilot
4 hours/day of routine work removedProperty setup, Drive folders, AI summaries, Telegram updates and archiving automated end-to-end.
Read Phase 2 arrow_forwardShow us how the process works today. We'll look at it with you and tell you where we'd start.
Let's talk →30 minutes. No pitch.
95% faster paperwork per car
Read case arrow_forward Car Amigo · AutomotiveYour whole fleet. One system.
Read case arrow_forward 3D Device · E-commerce60 managers. One sales process.
Read case arrow_forward Yolo Real Estate · Real Estate10× more leads. Same team.
Read case arrow_forward Chelo · Trading PsychologyInstagram leads on autopilot
Read case arrow_forward All Case StudiesSee every result across the team
View all arrow_forward