Agency Review Intelligence — Pain Points & Email

Agency Review Intelligence turns agency and client-review records you are authorized to process into compact sales, procurement, and positioning intelligence. It ranks recurring client pain points, identifies strengths, flags potential dealbreakers, suggests practical improvements, and can inspect each agency’s own public website for a business contact email. Each run produces one stable JSON row per agency plus a machine-readable summary for API and MCP workflows.

This Actor deliberately does not scrape Clutch, G2, Capterra, Trustpilot, DesignRush, or other review directories. Supply records from your CRM, surveys, first-party testimonials, licensed data feed, research file, or another source whose terms allow your intended processing. Review-directory URLs are blocked as crawl targets.

What data can I extract with Agency Review Intelligence?

Field	Meaning
agencyName	Normalized company or agency name
websiteUrl	Agency-owned website supplied in the input
sourceReference	Your provenance label, CRM ID, or licensed-record reference
rating	Optional source rating; never synthesized or blended
reviewsAnalyzed	Number of usable review texts processed
painPointScore	A 0–100 negative-feedback density signal, not a public rating
painPointThemes[]	Ranked themes with mention count, severity, and optional evidence
strengthThemes[]	Positive themes such as quality, strategy, communication, or support
dealbreakerSignals[]	High-severity excerpts that warrant human review
quickWins[]	Practical recommendations tied to recurring review themes
contactEmail	Best public business email found on the agency-owned website
emailConfidence	Confidence based on role mailbox, source page, syntax, and DNS
emailVerification	Domain-level result; never presented as SMTP mailbox verification
warnings[]	Non-fatal provenance, input, or enrichment warnings

Use this Actor for agency prospecting, vendor evaluation, win/loss research, sales personalization, partnership sourcing, and internal competitive intelligence. Do not use it to scrape a review platform, build an unlicensed review database, send unsolicited bulk messages, or treat algorithmic signals as a substitute for human due diligence.

How much does Agency Review Intelligence cost?

The Actor uses predictable pay-per-event pricing:

Event	Price	Charged when
Actor start	$0.00005	Once per run, scaled by memory
Agency analyzed	$0.10	A schema-valid agency intelligence row is produced
Business email found	$0.05	A public email is found on an agency-owned website

The run status displays the maximum possible cost before processing begins. Five agencies without email hits cost $0.50. Five agencies where every website yields an email cost at most $0.75. Failed inputs, blocked review-directory records, and email misses do not trigger result-specific charges.

How to use Agency Review Intelligence

You can pass records directly or point to an existing Apify dataset. If sourceDatasetId is present, it takes precedence over inline records. Common field names are normalized automatically: name, companyName, or agencyName; website, domain, or websiteUrl; and reviews, feedback, testimonials, or clientReviews.

{
  "agencyRecords": [
    {
      "name": "Northstar Studio",
      "websiteUrl": "https://agency.example",
      "rating": 4.2,
      "sourceReference": "crm-2026-0713",
      "reviews": [
        "Excellent design quality, but weekly communication was slow.",
        "Strong strategy. The final milestone arrived two weeks late."
      ]
    }
  ],
  "maxAgencies": 25,
  "websiteEnrichment": true,
  "maxWebsitePages": 3,
  "responseFormat": "concise"
}

Choose concise for AI-agent workflows. It returns ranked themes without evidence excerpts and keeps each row compact. Choose detailed when a human analyst needs short excerpts supporting each pain-point theme. The Actor never returns full source reviews unless those reviews are already present in your source dataset and you separately retain them there.

Output example

{
  "agencyName": "Northstar Studio",
  "websiteUrl": "https://agency.example/",
  "sourceReference": "crm-2026-0713",
  "rating": 4.2,
  "reviewsAnalyzed": 2,
  "painPointScore": 28,
  "painPointThemes": [
    { "theme": "deadlines", "mentions": 1, "severity": 3, "evidence": [] },
    { "theme": "communication", "mentions": 1, "severity": 3, "evidence": [] }
  ],
  "strengthThemes": ["quality", "strategy"],
  "dealbreakerSignals": [],
  "quickWins": ["Add milestone acceptance criteria and early risk escalation."],
  "contactEmail": "hello@agency.example",
  "emailConfidence": 90,
  "emailVerification": "mx_confirmed",
  "emailSourceUrl": "https://agency.example/contact",
  "analysisMethod": "rule-based-nlp-v1",
  "warnings": [],
  "scrapedAt": "2026-07-13T12:00:00.000Z"
}

The analysis is versioned and deterministic. painPointScore measures the density and severity of negative review language relative to supplied review count. It is not a rating, reputation score, or factual judgment about an agency.

Use Agency Review Intelligence through the API

JavaScript:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('muhammadafzal/analyze-agency-reviews').call({
  sourceDatasetId: 'YOUR_AUTHORIZED_DATASET_ID',
  maxAgencies: 50,
  websiteEnrichment: true,
  responseFormat: 'concise',
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Python:

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ['APIFY_TOKEN'])
run = client.actor('muhammadafzal/analyze-agency-reviews').call(run_input={
    'sourceDatasetId': 'YOUR_AUTHORIZED_DATASET_ID',
    'maxAgencies': 50,
    'websiteEnrichment': True,
    'responseFormat': 'concise',
})
items = client.dataset(run['defaultDatasetId']).list_items().items
print(items)

The Actor is also callable through Apify’s hosted MCP server. Its narrow input, declared dataset schema, compact result mode, provenance fields, actionable warnings, and predictable events are designed for agent selection and downstream automation.

Reliability and security

Website enrichment uses HTTP/HTTPS only, checks DNS before every request, rejects private or local network destinations, validates redirects, limits response size, and visits at most five same-domain pages. It checks email syntax and DNS/MX configuration but does not perform SMTP handshakes or claim mailbox-level verification. Partial website failures remain warnings; review intelligence still produces a useful row when the authorized input is valid.

No proxy setting is exposed because this Actor is not an anti-blocking scraper. Agency websites are fetched politely with bounded concurrency and short timeouts. A terminal RUN_SUMMARY records processed items, compliance skips, warnings, charges, and the dataset ID.

Related lead and intelligence Actors

• AI Lead Enrichment
• AI Web Extractor
• Google Maps Scraper
• Google Maps Restaurant Scraper
• Worldwide Dentist Scraper
• USA Plumbers Scraper
• Domain Scraper
• AI Citation Monitor
• AI Visibility Rank Tracker
• Geo Content Auditor

Frequently asked questions

How many results can I scrape with Agency Review Intelligence?

The Actor analyzes up to 500 authorized agency records per run. It does not scrape a source directory. For large datasets, process deterministic batches and preserve your sourceReference field for deduplication.

Can I integrate Agency Review Intelligence with other apps?

Yes. Use Apify integrations, webhooks, schedules, Make, Zapier, Google Sheets, or your own API client. The default dataset exports to JSON, CSV, Excel, XML, and RSS through standard Apify endpoints.

Can I use Agency Review Intelligence with the Apify API?

Yes. Pass inline agencyRecords or sourceDatasetId, wait for the run, then read the default dataset. The examples above show JavaScript and Python.

Can I use Agency Review Intelligence through an MCP Server?

Yes. Add muhammadafzal/analyze-agency-reviews to Apify’s hosted MCP tools. Use responseFormat: "concise" so sampled records leave enough context for agent reasoning.

Do I need proxies to analyze agency reviews?

No. The review analysis operates on your supplied records. Website enrichment visits only supplied agency-owned domains and does not expose proxy or evasion controls.

Is it legal to analyze agency review data?

That depends on the data source, license, purpose, jurisdiction, and applicable privacy and intellectual-property rules. You are responsible for having a lawful basis and contractual permission to process every supplied record. This Actor does not grant rights to scrape, republish, profile individuals, or process a platform’s data against its terms. Consult qualified counsel for high-risk or regulated uses.

Your feedback

If a legitimate input shape is not normalized correctly, open an issue with a redacted sample record and the run ID. Do not include private review text, API tokens, or personal data in public issues.