Pricing

from $200.00 / 1,000 business revieweds

Multi-Platform Reputation Analyzer

Cross-platform review monitoring, crisis detection, and decision-ready reputation risk scoring across Trustpilot, Google, BBB. Routable verdict, 30-day forecast, risk score, intervention playbook, cross-run memory layer. Deterministic. No LLM. Audit-ready.

Pricing

from $200.00 / 1,000 business revieweds

Rating

0.0

(0)

Developer

Ryan Clinton

Actor stats

Bookmarked

Total users

Monthly active users

8 days ago

Last modified

Categories

Lead generation

Other

Record Type

recordType

Optional

Discriminator for downstream filtering. 'business' for per-business analysis, 'comparison' for the cross-business summary record (only emitted in multi-business mode), 'error' for invalid-input / auth-failure / unhandled-exception records.

Type:string | null

Decision

decision

Optional

Routable verdict: 'act_now' (high-severity priority fired), 'monitor' (medium priority or declining trend), 'ignore' (stable reputation), 'no_data' (no platforms returned data). Branch on this in webhooks/Slack/agents.

Type:string | null

Decision Reason

decisionReason

Optional

One-sentence explanation of why the decision was assigned. Human-readable; downstream automation should branch on verdictReasonCodes instead.

Type:string | null

Verdict Reason Codes

verdictReasonCodes

Optional

Stable enum codes documenting which signals drove the decision. Branch automation on these — never on the prose decisionReason. Codes are additive within a major version; existing codes are never renamed. Values: NEGATIVE_SENTIMENT_FLIP / BBB_COMPLAINT_SURGE / BBB_ACCREDITATION_LOST / BBB_GRADE_DOWNGRADED / CROSS_PLATFORM_DIVERGENCE_SEVERE / CROSS_PLATFORM_CONSENSUS_NEGATIVE / REVIEW_VOLUME_DROPPED / PLATFORM_COVERAGE_LOST / MONTHLY_TREND_DECLINING / CROSS_PLATFORM_CONSENSUS_POSITIVE / STABLE_REPUTATION / NO_PLATFORM_COVERAGE / BUSINESS_NOT_FOUND / FIRST_RUN_NO_BASELINE.

Type:

Decision Readiness

decisionReadiness

Optional

Automation gate derived from confidence + decision: 'actionable' (decision is act_now AND confidence ≥60 AND concrete recommendations exist), 'monitor' (decision exists but insufficient certainty for unattended action), 'insufficient-data' (skip automation; verify against original sources). Filter WHERE decisionReadiness = 'actionable' for production-safe automation.

Type:string | null

Why Now

whyNow

Optional

Single-source-of-truth narrative for what changed since last comparison point. Identical string surfaces in this dataset record, the run status message, and the SUMMARY KV value — prevents reconciling divergent copy across surfaces. On first runs this explains the absent baseline; on subsequent runs it leads with the most material change.

Type:string | null

One-Line Summary

oneLine

Optional

Single shareable string ready for emails, dashboard tiles, or Slack messages — includes business name, decision verb, top reason, consensus rating, and platform coverage.

Type:string | null

Action Required

actionRequired

Optional

True iff decision === 'act_now'. Use as a boolean gate in if/else branching.

Type:boolean | null

Failure Type

failureType

Optional

Failure category when coverage is partial or missing: 'no-data' (business not found), 'all-platforms-blocked' (likely bot protection), 'invalid-input' (input validation failed), 'partial-coverage' (some platforms returned, others did not). Null on full coverage.

Type:string | null

Confidence

confidence

Optional

Confidence in this analysis. Fields: score (0-100 integer), level ('high'/'medium'/'low'/'very_low'), band ('high-confidence'/'medium-confidence'/'low-confidence'/'uncalibrated' — uncalibrated is reserved for first runs with no baseline), explanation (plain-English), factorCodes[] (stable enum: rich-sample / thin-sample / multi-platform-coverage / single-platform-only / rich-metadata / thin-metadata / cold-start-no-history / baseline-established / severe-divergence / cross-platform-consensus / sentiment-clearly-skewed / sentiment-mixed / monthly-trend-rich / monthly-trend-thin), components (sampleAdequacy / platformCoverage / metadataCompleteness / sentimentClarity / crossPlatformAgreement, all 0-1), readiness (actionable/monitor/insufficient-data — also exposed at top level as decisionReadiness). Components are combined via harmonic mean — one weak signal cannot be masked by strong others. First runs are capped at 70/100 because no baseline exists yet.

Type:object | null

Trust Summary

trustSummary

Optional

Executive-tier trust assessment, distinct from technical confidence. Fields: level ('high'/'medium'/'low'), reason (plain-language paste-ready string for exec emails), alignedSignals (count of independent signals supporting the verdict). Use trustSummary.level in non-engineering reports; use confidence.score in dashboards; use decisionReadiness for automation gates.

Type:object | null

Run Context

context

Optional

Workflow context for this run. Fields: analysisProfile, analysisProfileReason, autoResolved, modeHeadline, hasPriorRun, priorRunCount, outputProfile (compact/standard/full/alert), industry (passthrough tag), multiBusiness ({totalBusinesses, rank} when in multi-business mode, null otherwise).

Type:object | null

Entity Resolution

entityResolution

Optional

Confidence that the actor matched the right business profiles. Fields: confidence (0-1), risk (low/medium/high/unknown), signals[] (stable enum: domain-match-exact / domain-match-platform-url / name-similarity-high/medium/low / platform-url-alignment / multiple-platforms-found / no-domain-input / name-conflict-detected / short-name-ambiguous), explanation (plain-English), matchedNames[] (per-platform observed name + similarity 0-1). Filter WHERE entityResolution.risk = 'high' to flag possible mis-matches before acting on a verdict.

Type:object | null

Review Velocity

velocity

Optional

Review-rate intelligence. Fields: reviewsPerDay (last 30-day window), baselineReviewsPerDay (full history), trend (accelerating/steady/decelerating/insufficient-data), spikeDetected (z-score ≥2.0 vs prior buckets), spikeMagnitude (current rate / prior mean), classification (burst-event / slow-burn / recovery / steady / declining / sparse / unknown), explanation (plain-English), perPeriodRate[] (last 12 buckets). Burst events typically signal PR moments or crises; slow-burn signals gradual reputation movement; recovery signals post-spike normalisation.

Type:object | null

Customer Journey Analysis

journey

Optional

Maps reviews to 7 journey stages (discovery / onboarding / activation / support / billing / product / churn) using deterministic keyword patterns. Fields: stages[] (per-stage mentionCount + weight + sentiment + sentimentScore + topThemes + sampleReviewIndexes), weakestStage (lowest sentiment with ≥10% weight), strongestStage (highest weight + positive sentiment), summary (plain-English). Identifies WHERE the business is failing, not just WHAT customers say.

Type:object | null

Root Causes

rootCauses

Optional

Rule-based co-occurrence inference: when sentiment, theme keywords, and journey stages converge, the actor names a likely root cause. Empty array when no causes inferred. Stable enum of types: refund-policy-issue / customer-service-breakdown / product-quality-decline / billing-or-pricing-disputes / shipping-or-delivery-failures / onboarding-friction / pricing-perception-issue / outage-or-reliability-event / crisis-event-spike. Each: {type, headline, confidence (0-1), evidence[], journeyStage, suggestedFocus}. Arbitration: when 'crisis-event-spike' fires, related symptom causes are demoted. No LLM, no external API.

Type:

Root Cause Summary

rootCauseSummary

Optional

Arbitrated summary of root causes: primary (highest-confidence cause type), secondary[] (next 2-3 with confidence ≥0.5), confidence (of the primary), explanation (plain-English summary). Decouples the dominant verdict from the long causes list — automation should branch on rootCauseSummary.primary, not iterate rootCauses[].

Type:object | null

Signal Integrity

signalIntegrity

Optional

Cleanliness of the underlying data — distinct from confidence (which measures trust in the verdict). Fields: score (0-1), classification (high/moderate/low/very-low), explanation, components (sampleAdequacy / platformBreadth / metadataCompleteness / entityMatchClarity / historyDepth, each 0-1), issues[] (alias of dataGaps). A run can have low integrity but high confidence (the limited data we have agrees) or vice versa. Filter WHERE signalIntegrity.classification IN ('high', 'moderate') for production-safe automation.

Type:object | null

30-Day Forecast

forecast

Optional

Directional projection — explicitly NOT a prediction. Fields: direction (improving/stable/declining/unknown), confidence (0-1), timeHorizon (always '30d'), expectedImpact (consensusScoreDelta + complaintCountDelta), riskLevel (low/moderate/elevated/high/unknown), explanation, methodology (linear-regression / velocity-extrapolation / directional-heuristic / insufficient-data), caveats[] (always populated, names limits of the projection). Use for early-warning / monitoring; do NOT use for financial decisions or regulatory filings.

Type:object | null

Anomaly Report

anomalies

Optional

Z-score based outlier detection vs rolling baseline. Fields: detected (boolean), events[] (per-anomaly type / confidence / description / direction / zScore), summary. Anomaly types: consensus-rating-anomaly / review-volume-anomaly / platform-coverage-anomaly / sentiment-trajectory-anomaly / velocity-anomaly. Use to filter for runs that materially deviate from the business's own historical pattern.

Type:object | null

Relative Benchmark

relativeBenchmark

Optional

Two-axis benchmark. vsHistory: position (above-average/at-average/below-average/no-history) + currentScore + historicalMean + deltaPct + explanation — answers 'is this run unusual for THIS business?'. vsPeers: rank + peerCount + currentScore + peerMean + deltaPct + position (best-in-set/above-peer-mean/at-peer-mean/below-peer-mean/worst-in-set) — populated only in multi-business mode. Skips external industry data per design (no fabricated benchmarks).

Type:object | null

Pattern Match

pattern

Optional

Single named situation pattern that best matches the run. Type enum: support_collapse / billing_dispute_wave / viral_negative_event / review_bombing / platform_divergence_anomaly / cross_platform_consensus_positive / cross_platform_consensus_negative / recovery_in_progress / steady_state / insufficient_data. Fields: type, confidence (0-1), explanation, matchingSignals[]. Pure rule-based — no LLM.

Type:object | null

Reputation DNA

reputationDNA

Optional

Long-term signature: baselineRating (historical mean), volatility (low/moderate/high/unknown), dominantWeakness (consistent journey weak point), dominantStrength (consistent journey strong point), stabilityScore (0-1, higher = more predictable), historicalDepth, summary. Answers 'what kind of company is this over time?' rather than 'what is happening now?'. Stabilises over multiple runs.

Type:object | null

Platform Influence

platformInfluence

Optional

Per-platform contribution to the verdict. Fields: primaryDriver (trustpilot/google/bbb/null), secondaryDriver, explanation, perPlatformContribution[] (per-platform 0-1 contribution + observation string). Tells you WHICH source is driving the verdict — when verdict is negative, focus remediation on the highest-contributing platform first.

Type:object | null

Risk Score

riskScore

Optional

3-component composite for due-diligence consumers. Fields: overall (0-100), classification (low/moderate/elevated/high/critical), components (operationalRisk + reputationRisk + trendRisk, each 0-100), explanation, drivers[] (per-component plain-language reasons). Distinct from confidence (verdict trust) and signalIntegrity (data cleanliness). Use for procurement, vendor scoring, and investor due-diligence pipelines.

Type:object | null

Decision Trace

decisionTrace

Optional

Auditable contribution of each priority to the verdict. Each step: type (priority type), impact (0-1, normalised so steps sum to ≤1), description, sourceType. Use for compliance / audit trails — answers 'why did the engine decide what it decided?'.

Type:

Interventions

interventions

Optional

Typed playbook items derived from priorities + root causes. Each: type (11-token enum: support_scaling / refund_policy_review / crisis_communication / platform_response_program / onboarding_redesign / billing_transparency_audit / shipping_carrier_review / pricing_repositioning / reliability_engineering / identity_verification / monitoring_setup), priority (urgent/high/medium/low), headline, description, expectedImpact, timeToEffect (days/weeks/months), suggestedOwner, acceptanceCriteria[]. Drop-in for backlog tools, ticket creation, or playbook automation. Sorted by priority.

Type:

Memory Graph

memoryGraph

Optional

Repeated event chains across history — the actor's 'memory'. Each chain: id (8-char hash) / sequence[] (3-step signal sequence) / occurrences / typicalOutcome (next-run pattern type) / confidence / firstSeenAt / lastSeenAt. Activates after ≥3 historical snapshots. Useful for identifying business-specific recurring patterns ('this exact sequence has happened 4 times before — last 3 led to viral_negative_event').

Type:object | null

Intervention Outcomes

interventionOutcomes

Optional

Closed-loop INFERENCE — priorities that fired in the prior run and have disappeared in this run, linked to the intervention type that would have addressed them. observedEffect: signal-resolved / signal-stabilised / signal-deteriorated / no-change. Methodology: 'inferred-from-signal-disappearance' (the actor does NOT observe interventions being applied). Caveats[] always populated. Use as directional feedback, not measurement.

Type:object | null

Pattern Evolution

patternEvolution

Optional

Pattern transition between previous and current run. transition enum: unchanged / escalating / de-escalating / lateral-shift / first-classification. transitionConfidence (0-1). Tells you whether the situation is getting worse, better, or different.

Type:object | null

Trajectory

trajectory

Optional

Direction (escalating/stabilising/recovering/stable/unknown) + stage (early-crisis/mid-crisis/peak-crisis/post-crisis/stable/unknown) + daysInCurrentStage + progressionConfidence + explanation. The single most-valuable field for monitoring use cases.

Type:object | null

Decision Stability

decisionStability

Optional

Last 5 decisions + stability tier (stable / oscillating / rapid-deterioration / rapid-improvement / insufficient-data) + interpretation. Distinct from confidence — answers 'is the verdict consistent over time?'.

Type:object | null

Business Impact

businessImpact

Optional

Business-area mapping for executive readers. areas[]: each entry has area (customer_retention / brand_perception / revenue_at_risk / operational_efficiency / regulatory_compliance) + risk (low/medium/high/critical) + driver + explanation. summary aggregates the count by tier. Translates technical signals into business-stake language.

Type:object | null

Confidence Evolution

confidenceEvolution

Optional

Meta-intelligence: how confidence is changing across runs. direction (increasing/decreasing/stable/insufficient-data) + slopePerRun + reason + recentScores[]. Answers 'is the system becoming more or less certain?' — surfaces drift in data quality before it shows up in primary signals.

Type:object | null

Data Gaps

dataGaps

Optional

Plain-language enum codes surfacing missing or thin data: missing-google-coverage / low-sample-google / missing-trustpilot-coverage / low-sample-trustpilot / missing-bbb-coverage / few-dated-reviews / insufficient-trend-history / no-cross-run-baseline / low-entity-resolution-confidence. Used to filter low-quality records before automation.

Type:

Sentiment Engine Metadata

sentimentEngine

Optional

Metadata for the v2 hybrid-rule sentiment engine. Fields: version ('2.0'), method ('hybrid-rule'), meanConfidence (0-1 — average per-review confidence). v2 features: negation handling ('not good' → negative), intensifier weighting ('very terrible' > 'terrible'), contrast detection ('great but...'), sarcasm guard. Per-review fields sentimentConfidence and sentimentSignals[] are also available inside reviews[].

Type:object | null

Business Name

businessName

Optional

Business analysed.

Type:string | null

Business Domain

businessDomain

Optional

Business domain (when provided).

Type:string | null

Summary

summary

Optional

Aggregated metrics: totalReviews, averageRating, sentimentBreakdown {positive,neutral,negative}, overallSentiment, ratingTrend, platformsCovered.

Type:object | null

Platforms

platforms

Optional

Per-platform breakdown: trustpilot {rating, totalReviews, ratingDistribution}, google {mentionCount, averageRating}, bbb {grade, accredited, complaintCount, url}.

Type:object | null

Cross-Platform Divergence

divergence

Optional

Cross-platform divergence analytics: scoresByPlatform[] (normalised 0-5 per platform), spread (max minus min), severity ('none'/'mild'/'moderate'/'severe'/'no_data'), divergenceFlags[] (stable enum tokens), explanation, consensusScore (weighted mean), complaintToReviewRatio, crossPlatformInsights[].

Type:object | null

Priorities

priorities

Optional

Ranked decision actions sorted by severity (high → low). Each entry: type (cross-platform-divergence-severe / cross-platform-consensus-negative / sentiment-flip-negative / complaint-surge / review-volume-cliff / bbb-grade-downgrade / accreditation-loss / declining-trend / no-coverage / platform-coverage-lost), severity, title, recommendedAction, timeToAct, successMetric, expectedDirection. Use priorities[0].type for routing.

Type:

Change Since Last Run

changeSinceLastRun

Optional

Cross-run change report when monitorStateKey input is set. Fields: isFirstRun, changeFlags[] (stable enum), changeDetails[] (plain English), priorRunAt, snapshotCount, deltas (consensusScoreDelta, complaintDelta, reviewVolumeDelta, sentimentDelta, bbbGradeDelta). Null when monitorStateKey not provided.

Type:object | null

Reviews

reviews

Optional

Individual reviews collected across all platforms with platform, author, rating, date, title, text, sentiment label, sentiment score.

Type:

Themes

themes

Optional

Theme detection: legacy positive[] / negative[] keyword lists for backward-compat, plus detected[] with bigram theme phrases, polarity, count, platforms[], sampleReviewIndexes, exampleSnippet. Includes sampleSize and insufficientSample flag.

Type:object | null

Monthly Trend

monthlyTrend

Optional

Reviews grouped by YYYY-MM month with averageRating and reviewCount, sorted chronologically. Drives the ratingTrend ('improving'/'stable'/'declining'/'insufficient_data') field.

Type:

Analyzed At

analyzedAt

Optional

ISO 8601 analysis timestamp.

Type:string | null

Schema Version

schemaVersion

Optional

Output JSON shape version. Increments on additive changes; major version bumps signal breaking changes. Currently 5.0.0.

Type:string | null

Error

error

Optional

True when analysis could not complete (missing input, auth failure, or unhandled exception). Present only on records with recordType='error'.

Type:boolean | null

Error Type

errorType

Optional

Error category: 'auth' (Serper API key rejected), 'invalid-input' (missing businessName), 'unhandled' (unexpected exception).

Type:string | null

Error Message

message

Optional

Human-readable error explanation (only present on error records).

Type:string | null

Error Hint

hint

Optional

Suggested fix or remediation (only present on error records when applicable).

Type:string | null

Other properties may be included.

Cross-Platform Social Media Trend Scraper

jeweled_jockstrap/my-actor

Scrape trending from Reddit HackerNews ProductHunt GitHub Google in ONE feed. The only cross-platform trend aggregator for marketers and SEO professionals. Free.

Juan Triviño

Multi-Review Scraper — Trustpilot & BBB in One Run

ryanclinton/multi-review-scraper

Multi Review Scraper. Available on the Apify Store with pay-per-event pricing.

Ryan Clinton

AI Review Analyzer - Trustpilot & Google Maps Review Scraper

webdatalabs/ai-review-analyzer

Scrape reviews from Trustpilot and Google Maps and get AI-powered analysis: sentiment breakdown, complaint themes, fake review detection, and actionable recommendations. Perfect for reputation monitoring, competitor analysis, and customer feedback insights.

WebDataLabs

Health Risk Score API

vivid_astronaut/health-risk-score

Fabio Suizu

Trustpilot Review Scraper

cloud9_ai/trustpilot-scraper

Scrape Trustpilot business reviews and ratings. Extract reviewer names, ratings, review text, dates, and company response data for reputation monitoring and competitive analysis.

cloud9

Local Service Finder — Multi-Platform Provider Search

sovereigntaylor/local-service-finder

Find local service providers across Google Maps, Yelp, Thumbtack, YellowPages & BBB. Aggregates ratings, reviews, contact info, pricing, and services into unified profiles with cross-platform deduplication and analytics.

Ricardo Akiyoshi

Content Performance Tracker — Cross-Platform Analytics & ROI

apricot_blackberry/content-performance-tracker

Cross-platform content analytics: engagement, reach, shares, and conversion signals. See what drives results and why.

Creator Fusion

Email Address Risk Scoring

greip/email-scoring

Validate and score email addresses. Detects disposable emails, free email providers, role-based addresses, and provides risk scoring.

Greip

Fan Engagement Playbook (AI) - Community Strategy

macheta/fan-engagement-playbook

Generate a community engagement playbook: audience segments, content cadence, reply templates, moderation rules, and a crisis playbook for spikes and drama.

Anass

Trustpilot Review Scraper

zerobreak/trustpilot-review-scraper

Trustpilot review scraper that collects star ratings, review text, and reviewer details from any company page, so marketers and analysts can monitor brand reputation and run sentiment analysis without manual data collection.