Skip to content

Case study · Publishers

A publisher that pruned its way back to traffic

Removed 70 percent of low-value pages from the index and recovered the traffic lost in the previous year.

Who: An independent publisher with an eighteen-year archive, a small editorial team and revenue that depends on search traffic.

Organic sessions a month
520,000from 410,000
Indexed pages that earn a visit
71%from 22%
Evergreen guides refreshed
40from 0
Archive pages in the index
92,000from 310,000
Average search position of the top 40 guides
8.9from 14.2
Guides with article and author markup
40from 0

What changed

Eighteen years of publishing had left hundreds of thousands of thin archive pages: tags, dates, authors and page two of everything. They outnumbered the guides that earned the traffic, and the guides themselves had not been updated in years.

We went through the archive page by page with the editors, kept what earned visits, merged what overlapped and removed the rest with a redirect plan, then refreshed the forty guides that earn the most.

Six months on, pages in the index fell by 70 percent while the share of them that earn a visit tripled. Organic sessions rose from 410,000 to 520,000 a month, past where they were a year earlier.

Organic sessions a month, thousand
  • Start410
  • Month 2395
  • Month 3430
  • Month 5480
  • Month 6520

The situation

A publisher with eighteen years of archive saw a slow slide in search traffic as tag, date and author archives crowded out the guides that earned it.

What they wanted

  • Win back the traffic lost over the previous year
  • Make the best guides the pages search engines show
  • Decide what to remove with evidence, not nerve

What we found

  • Tag and date archives made up most of the indexed pages
  • Top evergreen guides had not been updated in three years
  • Author pages had no description or markup

What we did

  • Consolidated and removed archive pages with a clear redirect plan
  • Refreshed the 40 evergreen guides that earn the most
  • Added article and author structured data

How the engagement ran

  1. Weeks 1 to 4

    1. Score every page

    Every indexed page was scored on visits, links and overlap with other pages, so each removal had a reason written next to it.

  2. Weeks 5 to 10

    2. Prune in stages

    Archives were merged or removed in batches with redirects, and traffic was watched after each batch before the next went out.

  3. Weeks 11 to 20

    3. Refresh the guides

    The forty evergreen guides were rewritten with the editors: current facts, clearer structure, named authors and a visible updated date.

  4. Weeks 21 to 26

    4. Mark up and track

    Article and author markup was added, and we began tracking which guides assistants cite when asked the topic's main questions.

What surprised us

  • Traffic dipped slightly in month two while the first batches were removed, as we had warned it might
  • Four of the forty refreshed guides now earn more than the rest of the archive combined

What they keep

  • A page-by-page decision record for the whole archive
  • A redirect map kept in the publisher's own system
  • Forty refreshed guides with article and author markup

How we measured it

Sessions come from the publisher's own analytics, compared with the same months a year earlier. Indexed pages come from the search engine's index report. 'Earns a visit' means the page received at least one organic visit in the last ninety days.

We were afraid to delete anything. Seeing the evidence page by page is what gave us the nerve.
Managing editor, publisher

What is next: Citation tracking across AI assistants for the main guides.

Ready to be found by Google and named by AI assistants?

Book a 30-minute call. We prepare a free audit of your site first, so you leave with a written read whether or not you hire us.