The AI PMS Index is live — how it works

The AI PMS Index is a dated record of what “AI” actually means inside short-term rental property management software, and what it costs. It exists because the category has a name nobody maintains: vendors define “AI PMS” on their own blogs, and the generic phrase is swamped by residential property management results that have nothing to do with nightly bookings.

The editorial position is one sentence, and it is applied mechanically: an AI claim without a ship date, a plan tier and a price is a claim.

• A capability matrix covering 12 platforms against six AI capabilities. Every cell carries a status, the plan tier where one is published, an evidence grade, a link to the vendor’s own source and the date it was read.
• A comparison tablewith entry pricing, transparency grades, the release record for each platform and a verdict block per vendor — the second half of which is always “who should not buy this”.
• A true cost calculator that turns a property count into a real monthly bill for the 8 platforms that publish enough to compute one, minimums and band scales included. Results are ungated and every assumption is printed next to the number.
• A five-part definitionyou can run against any vendor’s own pricing page in ten minutes, and a disambiguation page for everyone who searched the obvious words and got shown apartment-leasing software.
• 12 platform profiles, on the identical template, including one for the operator’s own product.

1. 2 of 12 platforms publish no price at any tier. One of them serves its pricing page as a three-step lead form. Another returns a 404 for /pricing entirely. Quote-only is a finding, and it is recorded as one.
2. The loudest AI claims have the thinnest shipping record.The vendors making the most aggressive “AI-native” and “world’s first” claims are, without exception, the ones publishing no changelog and no release notes — so every capability they claim is graded on marketing copy alone. Meanwhile 8 platforms ship AI guest messaging and only 2 expose an external agentic surface an AI client can actually drive.
3. A minimum commitment beats a headline rate. The lowest advertised per-unit price in this index belongs to a platform whose published $400 per month minimum makes it the most expensive option in the set below roughly 21 units. That is not a gotcha; it is arithmetic anyone could do, and nobody publishes. The calculator does it.

• Nothing is ranked.Every table is alphabetical. There is no “best AI PMS” list, no editor’s choice and no pinned row.
• Nothing is rated.No scores, no stars, no review markup anywhere — including on the operator’s own product.
• Nothing is estimated.A missing price renders as “Not published”. A capability we could not establish renders as “Unclear”, never as “Not shipped”.
• Nothing sits for more than 90 days. Every fact carries the date it was last read at its source. If the dataset date on the masthead is stale, the site has broken its own rule and you can see that it has.
• No analytics. No tag manager, no analytics script, no session recording, no advertising pixel, no tracking cookie. Nothing here is watching you.

The company behind Zugrow, which is one of the 12 platforms in the index. That conflict is real, it is disclosed on every page and in the page markup, and the mitigations are structural rather than promised: alphabetical ordering everywhere, a published rubric, a mandatory “who should not buy this” section that in Zugrow’s case names four segments, and no affiliate links, paid placements or lead selling of any kind. The full statement is on the ownership page.

The most useful thing you can do with this site is try to catch it out. Corrections are dated and published under the corrections policy, and that applies to the operator’s own entry first.