Competitor Change Intelligence Monitor

Monitor authorized public competitor pages for meaningful pricing, product, availability, plan, policy, and changelog changes. The Actor compares each page with a trusted baseline and returns concise evidence, confidence, severity, and a recommended human review action.

Use it for competitor pricing intelligence, product launch monitoring, SaaS plan tracking, public terms updates, and release-note alerts without building a custom monitoring pipeline.

What You Get

• Structured before-and-after changes instead of raw page diffs
• Price deltas and percentage changes when values can be normalized safely
• Categories for price, availability, product features, pricing plans, terms/policies, launches/changelogs, and general content
• Confidence scores, severity, bounded evidence, and suggested review actions
• Stable change IDs for deduplication and downstream automation
• JSON Dataset output plus JSON and Markdown digests
• Optional immediate or weekly HTTPS webhook delivery
• Persistent trusted baselines with explicit candidate review and promotion

Quick Start

1. Initialize a trusted baseline

Run once with a page you are authorized to monitor:

{
  "targets": [
    {
      "name": "Acme pricing",
      "url": "https://example.org/pricing",
      "changeTypes": ["price", "pricing_plan", "product_feature"]
    }
  ],
  "confirmAuthorizedUse": true,
  "baselineAction": "initialize_trusted",
  "notificationMode": "none",
  "dryRun": false
}

2. Compare later runs

Keep the target name, URL, selected change types, selectors, and product identifiers consistent, then change baselineAction to compare_only. Schedule the Actor in Apify Console for recurring checks.

{
  "targets": [
    {
      "name": "Acme pricing",
      "url": "https://example.org/pricing",
      "changeTypes": ["price", "pricing_plan", "product_feature"]
    }
  ],
  "confirmAuthorizedUse": true,
  "baselineAction": "compare_only",
  "minimumMateriality": "low",
  "notificationMode": "none",
  "dryRun": false
}

The prefilled example.com input is a safe dry run. It does not mutate a baseline, send a webhook, or charge a page-checked event.

Output

Each target produces one Dataset report with fields such as:

{
  "targetName": "Acme pricing",
  "status": "success_changed",
  "baselineFound": true,
  "changeDetected": true,
  "materialChangeCount": 2,
  "overallSeverity": "high",
  "overallConfidence": 91,
  "changes": [
    {
      "category": "price",
      "field": "product.pro.price",
      "previousValue": 29,
      "currentValue": 35,
      "delta": 6,
      "deltaPercent": 20.6897,
      "severity": "medium",
      "confidence": "high",
      "recommendedAction": "Review pricing, margin, and positioning before changing your offer."
    }
  ]
}

The default Key-Value Store also exposes:

• RUN_SUMMARY: counts, statuses, delivery result, and page-event assessment
• DIGEST_JSON: bounded machine-readable changes
• DIGEST_MARKDOWN: concise human-readable digest

Baseline Actions

• compare_only: compare with the trusted baseline without changing it
• initialize_trusted: create the first trusted baseline; refuses overwrite
• store_candidate: save a candidate snapshot for review
• promote_candidate: promote only an exact reviewed candidate and parent hash

Use candidate promotion when a detected page revision should become the new trusted reference. This prevents an unexpected change from silently replacing the baseline.

Better Signal With Selectors

For noisy pages, add includeSelectors for the sections you care about and excludeSelectors for banners, timestamps, rotating recommendations, or other volatile content. ignoreTextPatterns accepts only bounded literal or simple glob patterns, not arbitrary regular expressions.

Known SKU, GTIN, MPN, brand, model, variant, and pack quantity values can improve product-change confidence. The Actor does not discover competitor products automatically or claim uncertain products are equivalent.

Webhook Alerts

Set notificationMode to immediate or weekly_digest, provide a secret public HTTPS alertWebhookUrl, and choose minimumAlertSeverity. Delivery happens only after the Dataset report is persisted. Payloads are deduplicated, redacted, bounded, and contain at most ten material changes per delivery.

The webhook URL is secret input and is never logged, returned, or stored in reports or digests.

Pricing

This Actor uses pay per event:

• page-checked: $0.003 for each successfully inspected page report persisted to the Dataset
• Actor start: Apify's small synthetic start event

Dry runs, invalid input, unauthorized targets, blocked or robots-disallowed pages, failed inspections, and failed Dataset persistence do not charge page-checked.

Safety And Limits

• One to 25 explicitly configured public HTTPS pages per run
• One document per target; no recursive crawling
• Public-IP DNS validation and rebinding protection
• Same-origin redirects unless another public HTTPS origin is explicitly authorized
• Robots.txt enforcement, bounded responses, timeouts, and retries
• No login, cookie acceptance, CAPTCHA solving, access-control bypass, or repricing
• Secret, credential, email, and phone-number redaction from retained evidence

confirmAuthorizedUse records that you already have permission; it does not create permission. Use the Actor only on pages that belong to you, intentionally expose content for this use, or that you are otherwise permitted to monitor. Follow source terms and applicable laws.

Common Statuses

• success_no_change: the trusted baseline matches
• success_changed: material changes were detected
• baseline_missing: initialize or review a candidate first
• robots_disallowed: source policy blocked inspection
• blocked_target: URL or network safety validation stopped the request
• http_error, unsupported_content, or internal_error: inspect the report

Support

When reporting an issue, include the public target type, selected change types, run ID, and redacted log excerpt. Never share credentials, private URLs, or webhook secrets.

Source and issue tracker: GitHub repository