Google Reviews Analysis AI - complaints, praise & contacts

Google reviews analysis for your places — complaints, praise and who to call, every week.

⚡ First 25 businesses free · 💵 ≈ $0.019 per qualified business all in (the review scan on your account included, $0.001 of it ours) · 🤖 $0.009 + ≈ $0.004 AI tokens with the AI analysis, on plans with model access or your key · ⏱️ demo in 5 seconds

Try it in 30 seconds. Click Try it — a free demo is pre-filled. Then paste your Google Maps places (or type "dentists in Austin, TX" as your goal): your first 25 qualified businesses are free, and businesses that fail your filters cost nothing here (the review scan runs on your Apify account, capped). Bring your own data: review rows in itemsList or a datasetId — no scraper run. → What a run costs · What the AI adds · Run it weekly

⚡ At a glance

What you get	one report per business: repeating complaints, praise, reply rate, risk score, contacts, an opener
You provide	Google Maps place links, a category + city, a scraper dataset — or review rows
Output	a ranked sheet — Excel, CSV or Google Sheets — plus an e-mail digest of what is new each run
Typical run	20 places × 30 reviews → ≈ 12 qualified reports (estimate, 60 % pass the default filters)
Cost of that run	$0.37 rules only · $0.52 with the AI ($0.36 review scan + 12 × $0.001 / 12 × $0.009 here + ≈ $0.05 tokens)
Free tier	first 25 qualified businesses per account
Keys / setup	Plan with Apify model access: no key · Other plans: your first 5 real businesses get live AI verdicts on us, then your own key — free Gemini / Groq = 1–3 AI verdicts per run, paid = every business
Which plan	Free demo, your own rows (template), the Places API with your key: any plan · live scraper search: a plan that runs Store Actors
Agencies	10 client cities × 20 places re-checked monthly with AI ≈ $5 a month, one digest and memory per client
Works with	Schedules, webhooks, Sheets, Slack, Teams, Discord, Telegram, Make / n8n / Zapier, MCP & AI agents

🎯 What this Actor does

It groups the reviews of your places into one row per business; we charge only for businesses that qualify:

• The review picture in numbers — recent average vs the rating, negatives, the unanswered ones, reply rate, a 0–100 risk.
• What customers keep saying — up to 4 repeating complaints, praise, three talking points tied to a review.
• Who to contact — from the business's own site: e-mail, phone, WhatsApp, socials, the owner's name if the site names one.
• An opener for your channel — e-mail with subject, SMS, WhatsApp, DM or a call script, on one observed detail.
• A score and label (hot / warm / cold) against your goal — sorted by who to call first.

Free filters: minRating, maxRating, minReviewsPerBusiness, requireEmail, target findings, an exclude list. Contacts: enrichWebsite reads the business's site, every e-mail MX-checked. Any plan: the Google Places API source with your own key. Deep research (off): the AI may search the web before scoring.

🚀 How to use it

1. Click Try it — the free demo is pre-filled and works as is.
2. Paste your places or describe them ("Google Maps places to analyse", or a category and a city in "What are you looking for?") — the live analysis on your Apify account, capped by "Max source spend". Leave the AI on: your first 5 businesses get live AI verdicts on any plan (which AI your plan gets).
3. Press Start — open the Reputation report or Outreach sheet view; export CSV / Excel, or wire a webhook.
4. Schedule it weekly with your address in notifyEmail — only new businesses are delivered and charged.

📦 Already scraping Google Maps? Analyse your Google reviews scraper data

Paste the dataset ID of your last Google Maps Reviews Scraper run (or Google Maps Scraper with reviews) into Dataset — no second scrape, we charge only for the qualified businesses. After every scrape: the chained-run recipe.

📚 Learn more: Step-by-step tutorial — Google reviews analysis with this Actor

🆚 Why this instead of running the Google Maps Reviews Scraper yourself?

The Google Maps Reviews Scraper (compass, 6,800+ users a month) returns one row per review — this Actor runs it for you and does the work that starts after.

	Google Maps Reviews Scraper alone	Google Reviews Analysis AI
Your time per 100 businesses	≈ 17 h reading 3,000 reviews, counting replies, writing openers	0 h — ranked, explained, opener written
Contacts	none	e-mail, phone, WhatsApp, socials, owner name
Repeat runs	the same places again	only new businesses are delivered and charged
Same 100 businesses, all in	$1.80 (100 × 30 reviews × $0.0006)	$1.90 rules only · $3.10 with the AI (the same $1.80 scan + 100 × $0.001 / $0.009 here + ≈ $0.40 tokens)

💵 Pricing — what a run really costs

Pay per result: one event per qualified business, nothing else.

Event	When it is charged	Price
free-tier	your first 25 qualified businesses on this Actor, in any mode	$0.00
qualified-business-basic	AI off — metrics, flags, reply rate, contacts, site signals, rule score, a rule-based opener	$0.001
qualified-business-ai	AI on — plus complaints, praise, talking points, AI risk, fit score with reasons, subject, AI opener, next step	$0.009

One free tier: your first 25 qualified businesses are free, whatever the tier.

This Actor never charges for: businesses below minScore, filtered out, excluded, or already received; compute is included. The review scan is not ours to waive: the scraper bills your account ≈ $0.0006 per review (≈ $0.018 for 30), rejected businesses included — capped by maxDiscoveryChargeUsd.

How that compares — $0.001 is 1.3× the median unit price of the source Actor and its closest competitors ($0.00075), but one qualified business replaces ≈ 30 raw review rows — about $0.02 of review rows at that price.

Worked example — 40 clinics × 30 reviews, every one qualifies (the worst case), AI on: review scan $0.72 on your account + 25 free + 15 × $0.009 ($0.135) + ≈ $0.16 tokens = ≈ $1.02, $0.025 per qualified business.

At volume — Agency, 10 client cities × 20 places re-checked monthly (dedupeTtlDays: 30) with AI: ≈ $5 a month · Franchise, 50 locations re-checked monthly with AI: ≈ $1.55 a month after the free tier.

AI tier, all in: ≈ $0.013 per business here ($0.009 + ≈ $0.004 tokens: 2,025 input + 298 output tokens per business measured 2026-10-16, priced at Claude Haiku 4.5 list rates) — through Apify's model access, or on your own key.

🤖 What the AI tier adds

The same business with AI off (qualified-business-basic, $0.001) — the demo run:

{
  "name": "Bright Smile Dental", "googleRating": 4.1, "reviewsAnalyzed": 6, "negativeReviewCount": 4,
  "unansweredNegativeReviews": 3, "reputationRisk": 78, "ruleScore": 53, "score": 53, "label": "warm",
  "outreachOpener": "Hi Bright Smile Dental team — I noticed 3 of your recent negative Google reviews have no reply from you yet; I help local businesses fix that fast. Worth a short call?"
}

AI on (qualified-business-ai, $0.009) — the same business judged against your goal (own-key run, 2026-10-16):

{
  "name": "Bright Smile Dental", "aiScore": 85, "score": 72, "label": "hot", "assessedBy": "ai",
  "recurringComplaints": ["Front desk never answers the phone", "Double booked appointments causing chaos", "Impossible to reach anyone about billing"],
  "talkingPoints": ["3 of the analysed reviews mention difficulty reaching the clinic by phone or getting a callback."],
  "outreachOpener": "I noticed recent reviews mentioning phone calls going unanswered and long waits at Bright Smile Dental, with no responses from the team. We help local dental practices handle review responses, fix recurring scheduling complaints, and build up positive ratings automatically. Would you be open to a quick chat about this?",
  "aiModel": "google:gemini-3.5-flash-lite"
}

Reading the reviews, spotting the pattern and writing a first line took ≈ 10 minutes each on 3 sample businesses (read by hand vs the AI rows, 2026-10-16). The AI also filters: the same run judged the demo café cold — its listed website is not the café's.

Try the AI on any plan: the demo shows labelled sample verdicts (aiModel: "sample") without model access; your first real search gets 5 live AI verdicts on us, then your own key (llmProvider: byok) — a free Gemini key (aistudio.google.com/apikey) assesses a few businesses per run (the rest are basic rows), a paid key every business.

⚙️ Input

{
  "placeUrls": ["https://www.google.com/maps/place/?q=place_id:ChIJN1t_tDeuEmsRUsoyG83frY4"],
  "goalDescription": "Clinics whose reviews complain about waits the owner never answers",
  "maxDiscoveryChargeUsd": 0.5,
  "maxQualified": 25
}

Field	Default	What it does
placeUrls	—	Google Maps place links or ids — they start the live analysis
goalDescription	—	What the AI scores against; "dentists in Austin, TX" with no places searches 10 places
preset	custom	A ready campaign; the 0.1 name campaignPreset still works
minScore	50	Decides what you pay: below this = not delivered, not charged
maxQualified	100	Hard cap on output (and on what this Actor charges)
maxDiscoveryChargeUsd	0.5	Stops the review scan at this spend
enableAi	true	AI analysis on/off
notifyEmail	—	E-mails the digest of new businesses
maxRating	0	Keep businesses rated at or below this (4.3; 0 = off)
targetFlags	—	Only businesses with one of these findings
sourceMode	actor	actor, placesApi, dataset, list; {} runs the free demo

Every other option is in the Input tab.

📤 Output

One dataset item per qualified business, best first — JSON, CSV or Excel, or streamed to a webhook.

Which score to use: sort by score; ruleScore is the rule part and aiScore the AI's — trend across runs on ruleScore. Upsert key for a CRM or Sheet: dedupeKey — the Google place id (else the Maps URL), the key the cross-run memory uses. A DIGEST record (what is new and what changed since the last run) and a client-ready REPORT page sit next to OUTPUT after every run that delivered — forward OUTPUT.reportUrl to your client. OUTPUT holds the funnel, charged events, aiCoverage ("3 AI-assessed (model), 0 rule-scored") and timing.

⭐ Found it useful? A review on the Store helps others find it — it takes a minute on the Actor's page.

🔁 Run it weekly

1. Open the saved task and click Schedule after your first real run (OUTPUT.scheduleTaskUrl), with your address in notifyEmail — the digest of new businesses and the trend vs the last run ("avg risk 62 → 58") lands in your inbox.
2. Leave dedupeAcrossRuns: true — only new businesses are delivered and charged; dedupeTtlDays (e.g. 30) re-checks a business with its newest reviews.
3. Paste a Slack, Teams, Google Chat, Discord or Telegram link into Send alerts to (sendAlertsTo), or import rows into the ready tracker sheet (dedupeKey first to upsert on).
4. OUTPUT.history and the REPORT show every run of the search (qualified, hot, average risk, unanswered negatives); historyDatasetName keeps a dated log across months.

Weekly cost is the new businesses only: 10 new per week ≈ $0.31 with AI (scan + $0.09 + tokens).

🎯 Try it for your niche

Pick a campaign in Start from a ready campaign (preset) and press Start — it fills what you left empty; each caps the scan at $0.50.

Campaign	What it looks for	Input
Reputation agency	dentists in Austin, TX rated ≤ 4.3, unanswered complaints	preset
Local SEO agency	restaurants in Denver, CO with few reviews, no replies	preset
CX / booking software	clinics in Manchester, UK with wait / phone complaints	preset
Business owner	your own places — the three things to fix	preset

🔌 Integrations, automation and API

After every compass/Google-Maps-Reviews-Scraper run: in that Actor's Integrations tab add "Run succeeded → start this Actor" with sourceMode: dataset, datasetId = the finished run's resource.defaultDatasetId and enableAi: false — every scrape you already pay for is analysed, rules only until you switch AI on.

Run outcomes — what your integration sees

• Webhook — webhookUrl gets {"event": "business.qualified", "business": {…}, "runId": "…"}; a failed POST is retried once, then listed in OUTPUT.webhook.failedKeys (not re-sent next run); webhookSecret signs every POST (X-Signature: sha256=<HMAC-SHA256 of the raw body>).
• Google Sheets / Slack — one click in the Integrations tab.
• HubSpot / Pipedrive — a Make / Zapier "create or update" step per row (name, bestEmail, outreachOpener), upsert on dedupeKey. Cold e-mail — the Outreach sheet maps onto Smartlead / Instantly / lemlist.

MCP clients (Claude, ChatGPT): add {"mcpServers": {"apify": {"url": "https://mcp.apify.com/?actors=rich_minds/google-reviews-insights"}}}. Minimal agent input (memory off, so a repeated question gets the same answer):

{"placeUrls": ["<Google Maps place link>"], "goalDescription": "…", "maxQualified": 10, "maxDiscoveryChargeUsd": 0.5, "dedupeAcrossRuns": false}

List-mode rows — what a pasted row needs

sourceMode: "list" runs on any plan: paste rows you already have into itemsList — start from the ready template (no scraper run, only our per-business price).

👥 Who is it for?

You are…	You run it to…	Start with
A reputation agency	find businesses losing customers to unanswered bad reviews — the report is the pitch	Reputation agency
A local SEO agency	find good businesses with a weak review engine	Local SEO agency
A CX / booking SaaS	reach businesses whose reviews describe the problem you fix	CX / feedback software
An owner or franchise manager	audit your own locations every month	Business owner
A lead-gen freelancer	sell a weekly reputation-leads feed	any preset + a schedule

🧠 How the AI works

• Grounded. One business at a time — Maps facts, the review sample, site signals. A complaint is recurring only when two reviews say it; owner names not on the crawled site are dropped; cold businesses get no opener.
• Your goal, your offer. The fit score measures the business against goalDescription and offerDescription; reputationRisk stays separate.
• Tokens where they matter. Only the best rule-scored businesses get the AI pass, up to aiCandidateMultiplier × maxQualified.
• No keys required where your Apify plan includes model access — otherwise your own key. Default anthropic/claude-haiku-4.5; with your own key the model follows the key (google:gemini-3.5-flash-lite for Gemini).
• Graceful fallback. Without the model, businesses keep their rule metrics — free while your free ones last, then basic.

🔒 Data, compliance and limits

• Only public data, no login; reviewer personal data is never stored. Follow the outreach laws that apply to you.
• Contacts need a website in the source — the Google Maps Scraper brings websites, the reviews scraper does not.
• A sample of the newest maxReviewsPerBusiness reviews (the Places API: Google's 5 most relevant).

❓ FAQ

How much will one run cost me? First 25 qualified businesses free; then $0.001 each ($0.009 with AI), plus the review scan on your account (≈ $0.018 for 30 reviews) and tokens (≈ $0.004) — worked example.

Is this a Google reviews scraper? It runs one — the Google Maps Reviews Scraper, on your account — and turns its rows into one report per business; dataset mode reads a run you have.

Is this review sentiment analysis? More: it names the complaints that repeat, the praise and three talking points — per business, not per review.

How does it help with reputation management? It finds unanswered negative reviews and falling ratings and hands you the evidence plus a first line.

Can I use it to find local SEO leads? Yes — the Local SEO agency preset keeps good businesses with few reviews, no replies and a weak site.

What will my first real search cost? 20 places × 30 reviews ≈ $0.36 of review scan on your account (capped at $0.50); the first 25 qualified businesses here are free.

How long does a run take? The demo 5 s; your own key 6.2 s per AI-assessed business — the table per mode:

Which Apify plan do I need? Any plan runs the free demo, list / dataset mode and placesApi with your own Google Maps key; the live scraper search needs a plan that can run Store Actors. Which AI your plan gets: At a glance → Keys / setup.

What if nothing matches? The run succeeds with 0 rows and charges nothing here; OUTPUT.stats shows what each filter dropped.

🧩 More Actors from the same developer

• AI Local Business Leads — more local businesses to pitch, with a website audit and a verified e-mail.
• Link Building Prospects — sites that will link to your local-SEO clients.
• Email List Cleaner — check the e-mails you collected before you send.

🆘 Support

Something missing or wrong? Open an issue on the Issues tab — requests from real campaigns are shipped first.

📝 Changelog

• 0.2.3 (2026-10-16) — Paste your Google Maps place links into one plain field, or just describe "dentists in Austin, TX" as your goal, and the live analysis starts; the four ready campaigns now run their own city search in one click; Every run now ends with an e-mail / Slack / Teams digest and a client-ready report showing what changed since the last run; the search can be saved as a weekly task straight from the status, and hot or warm businesses get an opener even with the AI off; The AI tier is easier to try and safer to pay for: sample verdicts in the free demo, 5 live verdicts on any plan, your own Gemini key works out of the box, a rate-limited AI stops waiting instead of running up the bill, and a business you already paid for is never charged again under another search.
• 0.2.2 (2026-10-04) — Bring your own data: rows in itemsList or a dataset ID in datasetId pick their own mode, start no source Actor and are charged only for the analysis.
• 0.2.1 (2026-09-28) — New source placesApi: the official Google Places API with your own Google Maps key, so live analysis works on every Apify plan.