Is This Property a
Good Short-Term
Rental Investment?

How does it work

A note on method, for investors and for anyone submitting a property.

When you enter a property address, you receive a multi-section short-term-rental investment report within a few minutes. This page explains exactly where every number in that report comes from — and, just as importantly, where it does not.

The principle: sourced, not invented

The single most important thing to understand about this system is that it does not make up property facts. Bedroom counts, square footage, year built, lot size, tax figures and assessed values are not estimated, guessed, or inferred from what a property "looks like". They are retrieved from public property records keyed to the address you provide.

This is a deliberate architectural decision, not a marketing claim. Our analysis engine operates under an explicit instruction: any property fact that has not been supplied to it must be printed as "Not provided — verify" rather than filled in with a plausible-sounding number. A report with an honest gap is more useful than a report with an invented figure, because an invented figure is indistinguishable from a real one on the page.

What happens when you submit an address

1. Record retrieval. The address is sent to a licensed property-data provider, which returns the property record: configuration, interior size, year built, lot size, property tax, assessed value, HOA dues where recorded, an automated valuation, and a long-term rent estimate.
2. Imagery. Street-level imagery is retrieved for the address where coverage exists. Where it does not, the report simply omits the image rather than substituting a placeholder.
3. Analysis. The retrieved facts are passed to our scoring engine, which produces the written assessment, the six scoring components, and a revenue estimate.
4. Underwriting. Every financial figure — debt service, operating expenses, net operating income, cap rate, DSCR, cash-on-cash return, the sensitivity matrix and the five-year projection — is calculated arithmetically from the purchase price and revenue assumption. None of it is produced by the language model.

What each part of the system does

The pipeline is deliberately built as separate components, each with one job. The separation matters: it is what allows facts, assumptions and estimates to be kept apart rather than blended into a single output.

1. Intake. The submission form on this site collects three things: your name, your email address and the property address. Nothing else is required, because nothing else needs to be. Access is gated behind an account, so reports are traceable to a requester rather than issued anonymously.

2. Dispatch. A small piece of glue on this website hands the submission to a background queue and releases your browser immediately. You are not held on a loading spinner while the analysis runs. The queue also carries a duplicate guard, so a double-click or a refresh cannot generate two reports or two charges against our data allowance.

3. Retrieval. This is the component that makes the whole approach possible. It takes the address and queries a licensed property-records provider for the physical and financial record of that parcel: configuration, interior size, year built, lot size, property tax, assessed value, HOA dues where recorded, an automated valuation and a long-term rent estimate. It separately checks whether street-level imagery exists for the address before requesting it, so the report never embeds a broken or placeholder image. Each data source is independent and optional — if one is unavailable, the others still return, and the report states which sources answered.

4. Scoring. The retrieved facts are passed to our analysis engine, which produces the written assessment, the six scoring components with their justifications, the Guest Magnet narrative, the regulatory review, and a revenue estimate. It is explicitly instructed to produce narrative and judgement only. It is told not to calculate mortgage payments, net operating income, cap rate, debt service coverage or projections — those are not its job, and language models are unreliable at arithmetic that has to reconcile across a dozen tables.

5. Underwriting. A separate calculation engine takes the purchase price and the revenue assumption and computes every financial figure in the report by formula: debt service, the full operating expense schedule, net operating income, NOI margin, levered cash flow, gross yield, cap rate, DSCR, cash-on-cash return, the ADR-by-occupancy sensitivity grid and the five-year projection. Because these are calculated rather than written, they reconcile with each other and with the assumptions printed at the top of the report.

6. Assembly. The narrative, the retrieved records and the computed figures are composed into the report document. Property facts are placed on the page directly from the retrieval layer — they are never re-typed by the analysis engine on their way to you, because anything that passes through a language model can be altered by it.

7. Delivery. The finished document is rendered to PDF and emailed from our authenticated domain. Delivery is verified rather than assumed.

Three categories of number — and how to read them

Everything in an Insim report falls into one of three categories. We label them distinctly because conflating them is how real estate analysis goes wrong.

1. Retrieved facts. Beds, baths, interior size, year built, lot, tax, assessment, valuation. These come from public records. They are as accurate as the county's own data, which is to say: usually correct, occasionally stale, and always worth confirming against a current listing or an inspection.

2. Modelled figures. The financing assumptions — 25% down, 75% loan-to-value, 6.00% fixed, 30-year amortisation — and the operating expense ratios are standard underwriting assumptions applied uniformly to every property. They are transparent, consistent, and stated on the face of the report so that you can substitute your own. They are not predictions about your specific financing.

3. Estimated performance. Average daily rate, occupancy and gross booking revenue are analytical estimates derived from the property's size, location, price and amenities. This is the genuinely uncertain part of any short-term-rental analysis, and we do not present it otherwise. Where a property has reported trailing income, the report says so and instructs the reader to obtain platform statements before relying on it.

Because the underwriting maths is computed rather than generated, the figures reconcile with one another. The DSCR in section 6 is derived from the NOI in section 5, which is derived from the revenue and expense lines above it. A reader can follow the arithmetic from top to bottom and check it.

The two scores

The Insim STR Score™ is a 0–100 rating built from six weighted components: Location Demand (25), Revenue Potential (25), Property Layout (15), Guest Appeal (15), Operations Risk (10) and Investment Efficiency (10). Each component carries a written justification grounded in the retrieved facts, so the score is auditable rather than opaque.

The Guest Magnet Score™ is a separate 0–100 measure of booking appeal only — how compelling the property is likely to be to guests, independent of whether it is a sound acquisition at the asking price. A property can be highly bookable and a poor investment, or vice versa. Separating the two prevents one from disguising the other.

Properties are graded against their own absolute merits, not ranked against competing listings.

What this report is not

This is a preliminary screening document. It is designed to help you decide which properties deserve a full underwrite — not to replace one.

Property facts, taxes, insurance, rental income, operating expenses, permits, zoning, HOA and CCR provisions, septic and well capacity, financing terms and short-term-rental eligibility must all be independently verified before acquisition. Local regulation is the single most common reason an otherwise attractive short-term rental fails, and it cannot be established from a property record; it requires reading the county code and the HOA documents.

Modelled assumptions are not guarantees of performance. Revenue estimates are estimates.

Why we built it this way

It would have been considerably easier to let a language model fill the entire report from an address. It would produce a confident, complete, professional-looking document in seconds — and the tax figure, the HOA dues and the square footage would be fabrications indistinguishable from fact.

For a document that informs a six- or seven-figure acquisition decision, that is not a trade worth making. Retrieving real records costs money per lookup and occasionally returns an incomplete result. We consider both an acceptable price for a report whose numbers mean what they say.

Insim Properties · Insim STR Score™ · Guest Magnet Score™
Preliminary analysis only. Independently verify all figures before acquisition.