OnCrawl

The Skeptic

This is where the vendor spends its evidence: JavaScript rendering with Core Web Vitals collected as lab data down to individual URLs, cross-referenced against log files and analytics, and impact reports cut by organic visits and bot crawl frequency rather than raw issue counts — Fresh Rank even ties first Googlebot crawl to first SEO visit from the customer's own logs. What I cannot hold: no stated page limits or crawl caps in the captured pages, and I found no public information on between-crawl regression detection, structured-data validation beyond an export dataset, or remediation guidance a developer could file as a ticket. 1 2 6

Report an error

The Skeptic

Keyword research as this category means it is barely present: the ranking dashboards are repackaged Search Console queries, and the keyword "intelligence" in the audience-first article is something you bring from elsewhere while the named audience technologies are third parties. The real content work here is near-duplicate and similarity analysis with boilerplate exclusion, which is genuine but narrow; I found no public information on volume, difficulty, intent, SERP feature data, or German-language keyword depth. 4 5 7

Report an error

The Skeptic

To their credit, the vendor stakes no index claims I would have to take on faith: the ranking data is Search Console data imported daily with a 16-month backfill, country and device breakdowns, and an unusually honest sampled-versus-unsampled display. The log processing claim of +500M lines daily is unverifiable from outside, and I found no public information on a backlink index, local or city-level tracking, or any ranking methodology of the vendor's own. 1 6 7

Report an error

The Skeptic

The API is the best-documented thing in the capture: token and OAuth methods, granular permissions, reference docs with code examples, every chart exposing its own API call and response format, and crawl scheduling driven programmatically — plus warehouse-style export to S3 or Google Cloud Storage in json, csv or parquet and a Looker Studio connector with templates and a published field list. Held back because the captured pages state no rate limits or per-endpoint credit costs, the Looker Studio connector is gated by plan behind a separate key with quota warnings, and I found no public information on white-label or multi-client reporting. 1 7 8 9 10

Report an error

The Skeptic

Exit is treated as a feature: page, link, raw HTML, cluster, keyword and structured-data datasets exportable in open formats to the customer's own S3 or GCS bucket, with export status tracked and failure reasons emailed — and the API exposes the same data the interface shows. But export requires an active subscription and an API token, and I found no public information on retention or a grace period after cancellation, so the archive looks retrievable only while the invoice is being paid. 1 9 10 11

Report an error

The Skeptic

The sovereignty record is empty: I found no public information on hosting location, subprocessors, certification, or a data processing agreement — and for a product that wants your raw server logs and analytics credentials, that silence is itself the finding. The contracting entity's French corporate form is the only European signal in the capture. 1 10

Report an error

The Skeptic

I found no public prices at all: no tiers, no keyword or crawl limits, no seat or credit costs, no VAT treatment on any captured page. The only pricing-adjacent signal is that features like the Looker Studio connector are gated by plan, which tells a buyer the real invoice depends on a sales conversation. 1 8

Report an error