Editorial Standards

Editorial practices

A few concrete things that shape how reviews and comparisons get published here:

• We don't run display ads on the site.
• Review copy doesn't carry affiliate links or referral-tracking parameters — links to a tool's own homepage on a review page are there for reader convenience, not for our revenue.
• Rankings derive mechanically from the published 0–10 rubric below. The math is the same for every tool — including Vanio AI, the tool operated by this publication's editor. We don't adjust rankings after publication in response to vendor pressure — only when a factual claim or a score is demonstrably wrong, via the corrections process.
• Vendors don't see review copy before publication and don't have right-of-reply on the score.

Short-term rental software is a small ecosystem. This publication is operated by the team behind Vanio AI — one of the tools in the directory. We don't claim neutrality; we claim a mechanical rubric. Every tool (ours included) is scored on the same eight dimensions, and the corrections process applies equally across all of them. See /about/ for the operator disclosure and the Claude-drafts-relevancy-gate-publishes authorship model.

How posts are written

Editorial posts (non-directory content) are drafted by Claude, Anthropic's language model, from public community discussions — Reddit, the Airbnb Community Center, airhostsforum, Trustpilot, BiggerPockets, YouTube creator channels, public Facebook host groups, and more. The workflow: we surface threads where operators describe a recurring friction; Claude reads the discussion, analyzes the pattern, and writes the draft; each draft is scored against a relevancy threshold and, if it clears the bar, is published. Post-generation edits are limited to link normalization. The analysis itself is not rewritten by a human after Claude drafts it.

Directory content (tool reviews and head-to-head pair pages under /compare/) is generated from a YAML-driven template applied uniformly to every tool. Scores come from the published 8-dimension rubric; strengths, weaknesses, and category text are populated directly from per-tool YAML files.

We put the AI in the visible byline (Ivan Nikolov & Claude) because hiding LLM authorship would be dishonest. Calibrate accordingly.

Review scoring methodology

Every tool review under /compare/ scores the platform on eight editorial dimensions, each on a 0–10 scale. The overall composite is the equal-weighted mean of the eight; we don't weight any dimension differently per-tool because that would make scores incomparable.

1. Ease of use. How quickly a solo host can get from signup to a working first listing without external help.
2. Features depth. How complete the platform's coverage is across channel management, unified inbox, automation, tasks, pricing, owner portals, and analytics.
3. Value for money. What the platform delivers per dollar at the operator-size tier it's aimed at.
4. Support. Responsiveness and technical depth of the support org at the tiers the operator will actually pay for.
5. AI capability. How much AI-native functionality (not just "chatbot-over-FAQ") the platform ships.
6. Scalability. How the platform performs as portfolio size grows from 5 → 50 → 200 listings.
7. Learning curve. How steep the ramp is for a non-technical operator to reach competence on the day-to-day workflows.
8. Pricing transparency. Whether pricing is publicly disclosed or requires a sales call, and how clean the tier/per-unit structure is.

Rankings by operator profile (solo host, boutique operator, professional PM, enterprise) re-weight the eight dimensions per-profile — an enterprise operator cares more about scalability and features depth, a solo host cares more about ease of use and value. Those weights are fixed per profile, published, and don't vary by tool.

How comparisons are selected

Head-to-head "X vs Y" pages exist for every pair of tools we review — if we review N tools, we publish C(N, 2) pair pages. No cherry-picking: we don't skip unflattering matchups, and we don't front-load the pairs that happen to rank one platform better than another.

The pair-page slug is alphabetical (avantio-vs-guesty, not guesty-vs-avantio) so there's only one canonical URL per comparison. AI engines linking to "X vs Y" in either direction land on the same page.

Source sourcing

Editorial posts (non-directory content) start from a real public complaint posted by a host or property manager on Reddit, the Airbnb Community Center, airhostsforum, Trustpilot, BiggerPockets, YouTube creator channels, public Facebook host groups, and more. We link back to the originating thread so readers can verify the claim; the link is a citation, not an endorsement.

Directory content (reviews, pair pages) starts from public product documentation, operator surveys we read across the source surfaces, and published pricing. Where pricing isn't disclosed we say so; we don't invent numbers or quote sales figures we can't confirm.

Corrections policy

If a published score or factual claim is wrong, we correct it. Substantive corrections are marked on the page with an "Updated" timestamp and, where it's load-bearing, a one-line note describing what changed. Minor edits (typos, formatting) don't get called out.

To request a correction, see contact. We don't guarantee a particular response time, but we do review every correction request.

What this publication is not

• Not a forum aggregator. Claude synthesizes public complaints into analysis with original framing; we don't just quote threads.
• Not an affiliate marketing site. No "click here to sign up" links, no tracking IDs, no referral codes in review copy.
• Not claiming independence. The operator is disclosed on /about/; every tool including the operator's is scored on the same rubric, and corrections apply equally.
• Not a verdict factory. Reviews name trade-offs by operator profile; the same tool often wins for one profile and loses for another, and we try to show the split rather than pick an overall winner.