Duplicate Titles in Your CSV? Check the Export Before Rewriting the Blog

You export a blog's titles, sort the spreadsheet and find the same phrase in dozens of rows. It is tempting to send the list straight to a writer. First check what those rows actually represent: different articles, old URLs that now reach the same article, or failed requests that all returned an error-page title.

A title export is an observation of a response. It becomes an editorial task only after you establish that the response belongs to the page you intended to review.

Keep the original URL beside the final URL

Do not discard redirects during cleanup. If several old addresses redirect to one current article, an extractor may report that article's title against every input address. Those rows do not necessarily represent several live pages with duplicate titles.

Group by final URL before grouping by title. Keep the original addresses for a separate redirect or migration review, and send only the current page to the editor. Avoid assigning five rewrites when there is only one editable article.

If you are still building the export, Guangsuan's guide to extracting page titles into a CSV provides a local example with original URL, final URL, status, title and H1 fields. Those fields are more useful for this task than a single column of headlines.

Separate fetch problems from writing problems

Review status codes and response content before treating a blank or generic title as an editorial omission. “Access denied,” “Sign in” or an identical site-wide error title can indicate that the collector received the wrong document. Even a 200 response deserves a spot check when the title looks suspicious.

Keep blocked, timed-out and non-HTML responses in a technical review queue. Do not overwrite the page's stored title just because the export could not read it. Where the site changes its title with JavaScript, compare the delivered HTML with the rendered page before deciding which layer needs attention.

This does not require a browser crawl of every URL. Start with a few examples from each suspicious group and escalate when their response behavior explains the pattern. If the site blocks automated access, use an authorized CMS export or a permitted method rather than disguising the collector.

Review duplicate text in the context of the article

Once the rows correspond to actual pages, compare the HTML title, visible heading and content promise. Different titles are not automatically better: changing a few synonyms can hide a duplicated topic without making the article more useful.

Consider a teaching example: three interviews are all titled “Inside the Studio.” The series name can remain, but each title should identify the guest or subject that distinguishes that interview. “Inside the Studio: A Ceramicist on Kiln Planning” gives the reader information that “Inside the Studio — Latest Update” does not.

For a tutorial series, a meaningful step or outcome may distinguish the pages. For two genuinely overlapping articles, ask whether their purposes should be clarified before rewriting either title. Merging or redirecting articles is a separate editorial decision; a duplicate-title flag alone is not authorization to do it.

Google's title-link guidance recommends descriptive, concise titles and avoiding repeated boilerplate. It also explains that search-result title links can draw from headings and other sources. Your CSV records the site's title element, not necessarily the title Google displays for a query.

Give each row a decision, not just a red cell

• Keep: the page is distinct and its title describes it accurately; a shared series label is not obscuring the difference.
• Rewrite: the page is valid, but the title is vague, incomplete or promises something the body does not deliver.
• Technical review: the response, template or rendering prevented a reliable reading.
• Content review: two pages appear to answer the same question and need a scope decision.

For a rewrite, include the current title, proposed title, reason, owner and review date. Check that a newly added year, number or claim is supported by the article. A character count can flag unusually long text, but it is not a universal Google display limit; displayed width depends on the device and wording.

After publication, export the affected URLs again and compare them with the approved decisions. Confirm the live title and heading first. Checking whether Google later adopts the wording is a separate follow-up, not a reason to keep changing a correct title every day.