Demonstration. Marisol Patrimonio SL does not exist and every figure here is invented. The software is the real thing — every screen, every model, every document — and nothing can be saved.
Guide
Short answers, each with a door to the screen that does the work. The search box finds these too — type what you are trying to do.
Press / (or the search box) and type where you want to be — a deal, a tab of it, a document, a person, a note, or an entry from this guide. Two words in any order name the inside of a deal: “usera pnl” opens that P&L. Enter goes there.
Today is the worklist: your tasks, contract deadlines inside two weeks, overdue next actions, deals past their stage limit, the queue waiting on us, trading months nobody has entered, and money leaving in the next sixty days. The same list arrives by email at six each Madrid morning when there is anything in it.
Today →New deal on the pipeline (or type “new deal” in search). A lead starts small — a postcard with the facts, files and three verbs — and grows tabs as it moves through the stages. Name it by street and city, the way it is spoken about.
Pipeline →Import on the pipeline, beside CSV. Paste the rows straight out of Excel or choose the file — the columns are read in Spanish or English, and prices in either dialect (1.800.000,00 and 1,800,000.00 are the same money). You see every row and what it became before anything is written, and a building already in the book is skipped rather than added twice, so importing the same sheet again is safe.
Import →The stage decides the screen: a Lead is a postcard, Study adds the record and the market, an offer adds negotiation, and so on — surfaces arrive with their stage and leave if it falls back. Each stage has a time limit (a Lead rots in a week); a deal past its limit shows on Today as stale.
Pipeline →A deal carries three estimates — revenue a year, costs a year, all-in cost — and a quick take computed from them. A blank is unknown, never zero: a missing cost does not become an NOI equal to revenue. Market estimates default to the 75th percentile — the thesis is beating the median operator — and the percentile is selectable.
The offer wizard on the deal writes structured terms — entrada, instalments, balloon — and each round is kept. Coverage of a lender's schedule is a lender's test, not a negotiating argument.
Passing a deal asks for the reason, and the reason is kept — the pass list is where the thesis gets tested against what was declined. A parked deal sleeps without a verdict. Archived deals stay searchable but lose to every live match.
Pick deals on the pipeline with the checkboxes: one picked offers its documents; two or more offer the cash statement or a comparison — the pipeline's own figures aligned on the firm's paper, where a dash means the record does not carry it, which side by side is a finding.
Pipeline →The plan wizard on the deal is the one door to a modelled plan: rent roll, costs, financing structures, exit. The thesis wizard pins the argument the money is raised on, and the app says when the model drifts from it. An empty plan never overwrites good figures — a plan with no revenue lines is not a plan.
One layer at a time: the plan's own line beats the market pack, the pack beats the house. House holds the defaults every plan starts from — change a number there and every new plan inherits it; naming a line on the deal stops that line being inherited and leaves the rest alone.
House →A plan can hold several financing structures, but the deal names one as its reporting basis and every quoted figure follows it — the hero, the documents, the cash statement. Change the basis on the deal, not by editing figures to match.
The Sensitivity tab re-runs the model across price, exit and operating sweeps — once arras are signed the price axis closes and the operating questions stay open. The fallback panel prices the building let by the month, which is what the licence risk costs if the tourist use never arrives.
The Capacity tab judges the unit mix — standard and extended cases, beds per typology, hotel room types where the asset is one. A sofa bed is upside in a studio and standard in a one-bed; the bedroom count decides which.
The plan re-checks its own arithmetic on every run, and where a workbook has been exported it measures drift against what that export said. Investor-facing figures get a checking script, not an eyeball — if a document and the model disagree, the tie strip on the plan says so before anyone outside does.
Drop photos, videos and PDFs in the Inbox from any device. A photo that knows where it was taken files itself to the deal whose building it is; everything else waits as pending with the nearest deals offered, and a press can ask Claude to propose where it belongs. Nothing is deleted on filing — the trail always says where a file went and why.
Inbox →Filed to the wrong deal? Every row on a deal's Documents and Media tabs carries Move — send it to the deal it should have been on, or back to the inbox to be decided again. What was read off it travels with it, and the trail stays unbroken.
Read on a document sends it to Claude, which proposes what it says — dates, figures, parties — as changes you confirm or discard; nothing is written without your press. Each reading is a metered call and is kept, so the same document is never paid for twice by accident.
From the pipeline row or the deal: a one-pager (works off estimates, no plan needed), the investment summary, the deck, the LOI, the cash statement, the full business plan and the Excel model — all on the firm's paper. Fit to send lists what would print blank on each, before it goes out.
Fit to send →The pipeline and Holdings export as CSV in the Spanish dialect Excel here expects — semicolons, comma decimals. The export is the screen's current view, so filter first and take exactly what you see.
Commitments live on the deal's Raise tab; capital calls are written against them and Capital shows every call falling due across the book. The deal's cash ledger records money that actually moved — append and correct, never edit in place.
Capital →An investor reaches a deal by a seat on it or a stake in the vehicle that holds it — one rule, everywhere. Behind their door (portal sign-in or a minted link you can revoke) they get the statement — one sentence, three figures, a card per deal — and the full business plan read-only where one is modelled. Bank details appear only inside a card asking for money. They can also ask questions in words; the answers draw only on what their own pages already show.
One row per building the firm holds: whether it is trading, the month's verdict against the plan that raised the money, and any months still owed to the record. The verdict is the deal's own — this screen and the deal can never disagree about the same month. Capital plans money; Holdings says who owns what; Portfolio says how the buildings are doing.
Portfolio →The entity register — SPVs, vehicles, people — with ownership drawn as the tree it is, and the GP's own P&L across the book. An owner chip anywhere opens the entity. Deals join Holdings when the firm actually holds them, not when the pipeline hopes.
Holdings →From the month a deal's opening date passes it carries an Actuals tab: five figures a month, read off the channel manager and the bank — revenue and cleaning income as the platforms report them, occupancy, one costs figure, a note. The verdict names the assumption that broke, not the variance; Today asks for a month from the 10th after it closes.
Team holds who can sign in and as what; invites are sent from there and the activity log shows who did what and when they last signed in. Investors are not team — their door is the portal.
Team →Setup's spend card meters the paid calls — AI readings and market data — with a daily safety cap that resets each morning. Anything metered sits behind a deliberate press, never behind a page load.
Setup →Demonstration. Marisol Patrimonio SL does not exist and every figure here is invented. The software is the real thing — every screen, every model, every document — and nothing can be saved. Walk the ninety seconds →
The firmThe investor