Loading ParaVida…
Demonstration dataA demonstration portfolio. Every property and every reading was authored to exercise the product. No hotel supplied any of it and no company is described.
ParaVida models the relationship between one person and one place across 11 dimensions. This account holds 12 properties across 10 markets, scored against 5 guest profiles by a deterministic engine. Everything below was computed when this page loaded.
The shape of the evidence, before any conclusion is drawn from it.
Strongest on beach (8.0), thinnest on mountains (2.7).
A portfolio has an environmental character whether or not anyone chose it. This one is computed from verified readings only — unknowns are excluded rather than estimated, so the character described here is the character the evidence supports.
Grade sets the ceiling on what confidence can claim. A reading cannot report more certainty than its source allows, however the intake was written.
The findings that should change a decision, rather than everything that could be counted.
3 different properties lead for 5 guest profiles.
The same buildings, reweighted by who is asking. This is the whole proposition, computed rather than claimed.
Training costs 19% of every weighted point the portfolio gives up.
A portfolio loses most points where it is merely adequate at something guests weight heavily — not where it is weak at something they do not. That the arithmetic improves is certain; that it returns anything commercially is a hypothesis ParaVida has no basis to price.
Things blocking value today, with what would resolve each one.
2 properties return no compatibility at all.
These are not low scores. They are refusals — ParaVida will not estimate a value it has not verified, so the property cannot be recommended to anyone at any price.
3 readings exist only as an AI proposal and cannot count until a human confirms them. Confirming or gathering them is what lifts this property over the coverage floor.
2 readings exist only as an AI proposal and cannot count until a human confirms them; 6 dimensions have no evidence at all. Confirming or gathering them is what lifts this property over the coverage floor.
An AI-proposed reading is retained in full and excluded from every score until a person confirms it. A dimension with no evidence is excluded from both sides of the arithmetic — never counted as zero.
Most repeated gap: Recovery, missing on 4 properties. A gap that repeats is usually one unowned process rather than 4 separate oversights.
Three routes, in the order they pay off. Nothing here is a recommendation to buy anything.
Decide who this portfolio is for.
The diagnostic states the identity, the redundancy and the guest the portfolio separates least, each with the evidence underneath it and the leadership question it raises.
Close the evidence that blocks value.
15 readings stand between this portfolio and full coverage. The Verification Center shows what each one unlocks and what would resolve it.
See it inside your own experience.
ParaVida adds an intelligence layer to a property page you already own. It takes no part in booking, pricing, inventory, loyalty or the customer relationship.
What this account cannot do yet
Every number on this page, and where it came from.
Every figure here was produced by the ParaVida compatibility engine at request time — the same engine, at the same versions, that serves the consumer product. No number on this page was written by hand, cached from a previous run, or produced by a language model.
Produced underscoring 1 · weights 4 · ontology 1 · climate model 1 · engagement 1 · min coverage 0.5
Each guest profile is a real Passport whose weights are derived by the frozen weighting pipeline — never written down. Each property is projected onto the same 11-dimension spine. The score is the weighted share of a perfect environment across the dimensions that carry a counted reading; dimensions without one are excluded from the numerator and the denominator both, which lowers confidence rather than the score. Below a coverage floor of 50%, the engine returns no score at all.