SEO Regression Proof

Audit SEO directives, canonicals and metadata; compare snapshots to find deployment regressions.

Start here

{
  "urls": [
    "https://crawlee.dev"
  ],
  "maxPages": 1,
  "stateKey": "my-monitor"
}

Run once, inspect the dataset and SUMMARY, then schedule repeat checks using the same monitor name.

Supported scope

Supplied pages or same-origin XML sitemap trees (five XML files), up to 100 audited pages per Actor source. Optional canonical destination fetches, HTTP status/redirect evidence and change comparisons. Not rankings, browser speed metrics or a full internal-link crawl.

Every row carries source evidence and observedAt. Publisher evidence is not a guarantee of real-world correctness. Blocked, failed, capped and malformed sources are explicit, not silently fabricated results. HTTP-only: JavaScript-only facts may be absent. Source data and image URLs are not a grant of republication rights.

Monitoring

FIRST_OBSERVATION, NEW_RECORD and UPDATED describe observed records. Changes include before-values and changedFields; SEO includes newly introduced issues. Technology disappearance needs two successful checks and only confirms missing signals, not uninstallation. Event and TED histories preserve up to 5,000 observed identities; disappearance from a finite source window is not deletion. Revision links require the same official notice identifier, not merely a similar title or procedure. Partial/failed/budget-stopped scans never replace the baseline. Don't overlap runs of the same named monitor.

Cost

$0.005 per run start plus $0.005 per observed page ($5/1,000), platform usage included under Apify PPE. Unchanged observed records count even with onlyChanges enabled. Failed sources have no record charge. Readable contacts/technology pages with no matches are still observed records; an event page containing no structured event produces no event charge. Configure max run cost; budget-stopped runs preserve history.

API integration

POST https://api.apify.com/v2/acts/hadi.kanj~seo-regression-proof/runs with your own Authorization: Bearer APIFY_TOKEN and JSON body. Poll /v2/actor-runs/{id}, retrieve /v2/datasets/{defaultDatasetId}/items and /v2/key-value-stores/{defaultKeyValueStoreId}/records/SUMMARY. Use /run-sync-get-dataset-items for short runs; asynchronous runs are safer for bulk inputs.

Export JSON for nested evidence, or CSV/Excel for tables. Use Apify schedules, webhooks, Make or n8n to retrieve completed results. No outbound email or messages are sent by this Actor.

Status and limits

complete describes the requested fetched source scope, not the entire internet or all company pages. Inspect errors, pagesFetched/requests and observedRecords. Public HTTP(S) only, ordinary ports, protected DNS and redirects, robots rules and crawl delay, two MB HTML and eight MB TED response limits. Up to 100 URLs per Actor run, five contact pages/domain and 1,000 events/notices per source.

Support

Open an Actor issue with a public source URL, expected field and run link. Never post API keys or private client data.