Free Content Score Checker — Score Any Page for AI Citations

ContentScale is a free SEO content score scanner that analyses any webpage using the advanced GRAAF model (50pts) + CRAFT (30pts) + Technical SEO (20pts) — returning a deterministic 100-point ContentScore with a prioritised fix list in 30 seconds. It tells you exactly what to fix to produce smarter content that ranks and gets cited. No login required. Built in Amsterdam by Ottmar J.G. Francisca. A free content quality checker, distinct from Content at Scale (now BrandWell), Surfer SEO, and Clearscope — with done-for-you recovery services from €297/month.

This scanner is not the AI Citations Tracker. This ContentScore scan predicts citation-readiness from on-page signals — the things Google's algorithm looks for (E-E-A-T, structure, schema). The separate free AI Citations Tracker verifies, live, whether Google AI Overview, Perplexity, Copilot, Claude and ChatGPT actually cite your page right now. Prediction vs verification — run both for the full picture.

Content at Scale rebranded to BrandWell — the CRAFT methodology and affordable pricing are gone

Content at Scale rebranded to BrandWell and split into two products. The main homepage (brandwell.ai) is now a GTM intent platform — TrafficID, buyer intent, CRM enrichment — targeting enterprise B2B teams at $999–$3,999/month. An AI writing app still exists at app.brandwell.ai (WriteWell + RankWell) but the CRAFT framework as an explicit methodology, the affordable pricing for freelancers and agencies, and the content quality education community are gone.

If you were a Content at Scale user who valued the CRAFT framework, the content quality scoring, or the SEO methodology — ContentScale is where that lives now. Free, no login, 100-point GRAAF + CRAFT score in 30 seconds.

Why AI content detectors are unreliable — and why ContentScale measures what actually matters

AI detectors sound authoritative. They are not. The research is clear:

• Accuracy ranges from 61% to 90% depending on the tool — far below what vendors claim. A Sultan Qaboos University study found performance on hybrid human–AI text dropped to nearly 0%.
• False positives are rampant. University of Maryland researchers found that minimal paraphrasing with GPT-4o defeats detection in 10%–75% of cases. The same human-written text gets different scores on the same tool run twice.
• Non-native English writers are disproportionately flagged. Common Sense Media documented a 20% false positive rate for Black students versus 7% for White students — the detector is measuring language patterns, not AI use.
• Google doesn't use AI detectors. Google has never used AI detection as a ranking signal. They evaluate whether content is helpful, original, and credible — not how it was written.
• Freelance writers are losing work unfairly. Gizmodo reported in 2024 that AI detectors caused job losses among freelance writers whose human-written work was incorrectly flagged.

The right question is never "does this read as AI?" — it is "does this content demonstrate genuine expertise, cite real sources, and give readers something they cannot find anywhere else?" That is what the advanced GRAAF model measures. Robotic, undifferentiated output — regardless of which artificial intelligence tool produced it — fails that test. The GRAAF Framework is the part of your content creation stack that makes the difference: it gives every AI language model a structure to work from, so the result is smarter content that ranks and earns citations, not generic text that disappears into the SERP. That is what Google rewards. That is what gets you cited in AI Overviews.