When checking website text and links, organize the work around four goals: finding spelling and grammar issues, identifying internal links in the main content, reviewing link path types, and finding links that open in a new tab. Do not combine these into a single list of “website errors.” The following is based on Screaming Frog’s feature overview published in July 2020. Descriptions of the tool’s interface and capabilities refer only to that article and should not be taken as a description of the current version.
Text checks: trace page-level errors back to shared templates
To help improve text quality, the SEO Spider described at the time could check spelling and grammar while crawling a website. Its Content tab listed error counts for each page, along with error types, explanations, suggested corrections, and previews of page text. When reviewing results, look beyond the counts and check the actual text to determine whether each suggestion applies.
- Page-by-page review: identify whether a flagged issue concerns spelling or grammar, then check the context.
- Cross-page summaries: the tool could list the top 100 unique errors and the number of URLs affected by each, helping identify issues in shared templates.
- Scope adjustments: specific content could be included or excluded by HTML element, class, or ID.
Confirm the language before adjusting dictionaries and rules
Text checks need to match the page’s language. The feature described in the article detected the language through the HTML language attribute and also allowed manual selection. However, the supplied excerpts did not explicitly confirm support for Traditional Chinese, so the feature’s suitability for Traditional Chinese websites cannot be guaranteed.
At the time, users could also ignore words, add terms to a dictionary, or disable grammar rules. Terms added to the dictionary were retained for future crawls. After adjusting the dictionary, ignore list, or rules, users could update the results without recrawling the URLs. These settings only change how checks are performed; they do not mean that ignored text has been verified as correct.
Internal link analysis: distinguish main content from navigation first
To understand which pages link to a particular page from their main content, analyze those links separately from links in the main navigation, sidebars, and footers. According to the article, the tool could determine a link’s location using its XPath and semantic markup. When a website used semantic HTML or appropriately named elements, the tool could classify links automatically, though the results might still need adjustment.
For example, the mobile menu in the article sat outside the nav element, so its links were incorrectly classified as main-content links. Adjusting the classification settings using the mobile-menu__dropdown class name in the path allowed the tool to classify those links as site-wide navigation. This is why you should check that links are classified by location correctly before analyzing internal links in the main content.
Paths and new tabs: the type alone does not indicate a problem
At the time, the tool recorded link path types, including absolute, path-relative, protocol-relative, and root-relative paths. The original article only noted that relative links could raise integrity, security, and performance issues in certain circumstances; it did not call for all links to be changed to absolute paths.
You could also inspect the target attribute to find links that used _blank to open in a new tab, then assess their usability, performance, and security. This attribute is only a starting point for investigation; opening in a new tab alone does not establish that a link is problematic.
Checking website text and links: perspectives from across the web
Do spelling or grammar errors directly affect search rankings?
The supplied material does not establish a direct effect on rankings. In its 2020 article, Screaming Frog noted that Google’s Search Quality Evaluator Guidelines repeatedly identified spelling and grammar errors as characteristics of low-quality pages. However, that description cannot be interpreted as a rule in the ranking algorithm. When checking text, improving the user’s reading experience is a reasonable starting point.
Can a list of common errors represent all text issues on a website?
No. The list described in the 2020 feature overview included only the top 100 unique errors and the number of URLs affected by each. It was useful for finding recurring issues across templates, but that does not mean it covered every issue. To review individual pages, consult the page-by-page results and suggested corrections, and assess them in context.
References
- Screaming Frog SEO Spider Update – Version 13.0www.screamingfrog.co.uk · 2020-07-01T13:04:29+00:00 — Discusses: Do spelling or grammar errors directly affect search rankings?; Can a list of common errors represent all text issues on a website?

