Bio · Avatar · Schema · Slug · Multi-Site Sitemap Audit
This tool pulls the author and brand transparency signals across the Kompas portfolio into one place — so we know which sites are sending the right E-E-A-T cues to Google and which need cleanup.
For every site we audit three things: each author's root page (bio, avatar, schema, links, slug), the brand's colofon (who's on the editorial team, are they linked, is the page in the footer), and the brand's about page (does it exist, is it linked, is there real content).
The tool covers three layers of author/brand transparency signals across all Kompas sites: per-author audits (Monitor), brand-level colofon checks, and brand-level about-page checks. Each runs independently; cross-references between them are surfaced where useful.
For each site, fetches /author-sitemap.xml (or /authors-sitemap.xml for Kek Mama) and audits every author's root /author/[slug]/ page — once each, no deep crawling. Five signals scored:
description or theme bio block)@type: Person in Yoast's @graphadmin / redactie / numericScoring: 100 − (P0 × 35) − (P1 × 15) − (P2 × 7), clamped to 0. A ≥ 85, B ≥ 70, C ≥ 50, D ≥ 30, else F.
Cap: 250 authors per site, sorted newest-first by lastmod. Override via ?limit=N on the API (up to 500).
For each configured colofon URL, checks four things:
<strong> tags, image alt, author links)/author/[slug]/ page isn't linked from the colofon. Sorted by most recent lastmod so active missing authors surface first.Sites without a configured colofon URL appear in To do and get probed at /colofon/ on each run. When a probe returns 200, the site auto-promotes into the main list.
For each configured about-us URL, checks two things:
Same auto-discovery pattern as Colofon: sites without a configured URL get probed at /over-ons/ on each run.
For each site, fetches the homepage and extracts every JSON-LD block. The Organization-like entity (Organization, NewsMediaOrganization, or Publisher) gets graded on completeness using the same A–F scale as the author audit.
name, url, logosameAs with ≥2 social profiles, descriptionfoundingDate, founder, contactPoint, address, email, telephoneEach row also lists every JSON-LD entity present (Organization, WebSite, BreadcrumbList, Article, etc.) and exposes the raw blocks for direct inspection.
Different shape from the other tabs: paste any article URL, get its byline + timestamp markup audited. Useful for diagnosing one-off issues or verifying a fix. Works on any site, not just Kompas — and supports both Yoast @graph and standalone JSON-LD.
datePublished, invalid ISO dates, dateModified earlier than datePublished@type, missing author.url, future datePublished, missing dateModifiedjobTitle / image / description, Open Graph metadata mismatchesKeeps its own per-URL history (last 30, deduped). Click any history entry to re-check.
For a single author, fetches their archive page, extracts recent article titles (up to 25), and runs Dutch-aware keyword frequency analysis to surface the topics they're qualified to write about. Cheap (no LLM cost), runs in seconds. Useful for:
knowsAbout values for Person schemaMethod: tokenize titles → strip Dutch + English stopwords → count single words AND bigrams (so "werkende moeders" stays together) → require ≥2 articles or ≥3 mentions → de-dupe (bigrams override their constituent unigrams). Top 12 topics ranked by article-count, then total mentions.
Pick from already-audited authors via the dropdowns (populated from Monitor tab data) or paste any author URL directly.
The Monitor's drill-down shows a colofon pill on each author card — green ✓ if they're listed on the colofon, red ✗ if missing, gray "—" if the colofon hasn't been checked yet. On-colofon authors float to the top of the table.
All check results are cached in this browser's localStorage. History keeps the last 50 audit runs. Re-runs always fetch fresh data from the source — the cache is just for restoring state when you reopen the tab.