How does Marfeel detect the title of an article?

Marfeel automatically extracts the title of an article by evaluating a waterfall of heuristics in a fixed priority order. The first source that returns a valid value wins, and subsequent sources are skipped. This is part of the broader metadata extraction process that the Marfeel Editorial Crawler performs on every article page.

1. JSON-LD structured data (highest priority)

The crawler first looks for a JSON-LD block containing a headline property. This is the recommended approach because it aligns with Google’s structured data guidelines.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "NewsArticle",
  "headline": "Title of a News Article"
}
</script>

2. Microdata

If no JSON-LD headline is found, the crawler falls back to Microdata markup using the itemprop="headline" attribute.

<div itemprop="headline">Title of News Article</div>

3. HTML <title> tag (lowest priority)

As a last resort, Marfeel reads the HTML <title> tag. This is not an ideal fallback since most sites append the site name to their title tag, which can result in a less accurate article title.

Troubleshooting

If the detected title is incorrect or missing, verify that your structured data is properly configured. You can use the metadata debugging tools to inspect exactly which source the crawler is reading from.