Stayingapi review intelligence digest

Review-intelligence digest → weekly reputation report

Reputation is spread across platforms and scored on different scales, so problems surface late. This workflow pulls normalized reviews for a property (native rating scales preserved, never silently rescaled), summarizes the average, the recurring themes, and the most recent low ratings, and delivers a weekly reputation digest. On the AI-agent path the agent reads the reviews and writes the sentiment summary itself; on the n8n path the aggregation is plain JavaScript, so no LLM key is required — add an LLM node only if you want richer narrative.

• Base URL: https://api.stayingapi.com/v1 (REST) · MCP: https://mcp.stayingapi.com/mcp
• Auth: Authorization: Bearer stay_live_… (free sandbox: stay_test_…).
• Get a free key (no card): https://stayingapi.com/signup · Full machine contract: https://api.stayingapi.com/openapi.json

Steps

1. Choose the property to monitor — Give the workflow a platform and the platform-native listing id (from a prior /v1/search), how many reviews to pull, and a sort order (recent is the default for a reputation watch).
2. Pull the normalized reviews — It calls GET /v1/reviews, which returns reviews in one schema with each native rating scale preserved and echoed (5 for Airbnb/Vrbo/TripAdvisor, 10 for Booking.com/Expedia/Hotels.com).
3. Summarize sentiment and themes — The agent reads the reviews and writes the summary; the n8n transform aggregates an average on the native scale, a theme-frequency table from the text, and the most recent low ratings — no LLM key needed.
4. Deliver the weekly digest — A reputation digest (average, top themes, recent complaints) is posted to Slack or email on your cadence — swap the last node or add an LLM node for a more narrative write-up.

When to use this

The user is a host or property manager who wants to monitor reputation for a property (or a comp-set) across platforms in one schema. Reach for get_reviews / GET /v1/reviews: normalized, paginated reviews with each native rating scale preserved and echoed.

How to do it

1. Call GET /v1/reviews with platform + listingId (or url), limit, sort=recent. Paginate with meta.pagination.nextCursor if you need more than one page.
2. Read ratingScale on each review and normalize before averaging (5 vs 10 scales differ).
3. Summarize: the average on the native scale, recurring themes/complaints in text, and the most recent low ratings (e.g. rating ≤ 60% of scale). YOU are the summarizer on the agent path.

Delivering a weekly digest

Run on a weekly cadence and post the summary to Slack/email/Notion. Lead with the average and its trend, then the top themes, then the specific recent complaints worth a reply.

Comp-set version

To benchmark against competitors, repeat the reviews call for each listing id in the comp-set and compare normalized averages; attach names via GET /v1/listing/{platform}/{id}.

Example — REST

curl -sS "https://api.stayingapi.com/v1/reviews?platform=booking&listingId=abramovic2&limit=40&sort=recent" \
  -H "Authorization: Bearer $STAYINGAPI_KEY"

Example — @stayingapi/sdk

import { StayingApiClient } from '@stayingapi/sdk';

const scout = new StayingApiClient({ apiKey: process.env.STAYINGAPI_KEY });

const { data } = await scout.reviews({
  platform: 'booking',
  listingId: 'abramovic2',
  limit: 40,
  sort: 'recent',
});

const scale = data[0]?.ratingScale ?? 5;
const avg = data.reduce((s, r) => s + r.rating, 0) / data.length;
console.log(data.length, 'reviews · avg', avg.toFixed(2), '/', scale);

MCP (no key pasted into the agent)

On an MCP-capable runtime, connect the StayingAPI server at https://mcp.stayingapi.com/mcp (OAuth 2.1 + PKCE) and use:

• get_reviews — normalized, paginated reviews for one listing.

Async & partial failures

A live call that has to scrape returns 202 + a jobId; poll GET /v1/jobs/{jobId} (free) until data.status is completed, then read data.result. The @stayingapi/sdk auto-polls. On a fan-out, check meta.partial and meta.platformResults[] — report what succeeded and note any meta.warnings[].

Credit awareness

Costs are per-endpoint and metered by result (v3). Failed, empty and blocked calls are never billed, and sandbox (stay_test_) calls are always free. The exact, current costs live only in https://stayingapi.com/pricing and the machine-readable https://api.stayingapi.com/openapi.json — read them there, don't assume.

---

Get your free key → https://stayingapi.com/signup · Docs: https://stayingapi.com/docs · Workflow: https://stayingapi.com/workflows/review-intelligence-digest