Upgrade to Pro — share decks privately, control downloads, hide ads and more …

How to use server logs to plan & surface content

Avatar for Yvie Ansari Yvie Ansari
April 30, 2026
280

How to use server logs to plan & surface content

Avatar for Yvie Ansari

Yvie Ansari

April 30, 2026

Transcript

  1. How to use server logs to plan & surface content

    Yvie Ansari Snaptrip Group Linkedin.com/in/yvonneansari | Speakerdeck.com/yvieansari
  2. I am not saying that you should purely look at

    your server logs to build a content strategy – this would not work.
  3. Your content strategy should always focus on the user first

    and include data points based on user behaviours.
  4. YOU CAN ALSO DETERMINE: If there are parts of your

    site that LLMs / Google just can’t get to
  5. Content duplication and similar content is relatively common for large

    ecommerce or travel websites. Finding and resolving this can be done more easily with access log data. SIMILAR CONTENT & DUPLICATION
  6. IF CONTENT IS ACCESSED OFTEN... …but not driving traffic, you

    could have issues with: • EEAT • Opportunity
  7. Property pages for this website are accessed a lot and

    account for a decent percentage of site traffic: HIGH ACCESS & TRAFFIC CONTENT
  8. It’s no secret that bots can hallucinate information including URLs.

    This might sometimes highlight hidden opportunity for pages to be created. URL HALLUCINATIONS MAY SAY…
  9. URL HALLUCINATIONS MAY SAY… A bot may hallucinate something within

    this structure that doesn’t in fact exist. You can validate if it should exist.
  10. However, these hallucinations may also simply highlight an internal linking

    issue that needs to be resolved. URL HALLUCINATIONS MAY SAY…
  11. Using server logs is a great way of finding content

    or pages that Google should not be crawling, which requires a review of your robots.txt file. URLS THAT SHOULD NOT BE CRAWLED
  12. Blank pages due to temporary content forms in the checkout

    flow URLS THAT SHOULD NOT BE CRAWLED
  13. Site sections that are only meant for the user and

    not for indexing URLS THAT SHOULD NOT BE CRAWLED
  14. Categorising URLs into topic clusters will make your analysis and

    planning much simpler. HOPEFULLY YOU’VE LEARNED…
  15. Finding important content with high or low access and low

    performance can be a good place to start in terms of identifying low hanging fruit for improvements. HOPEFULLY YOU’VE LEARNED…