1

Let's say I have a website with a couple of hundred pages – some performing well, others not so well. And then there are some pages that are what I would call 'invisible pages' – no visitors ever entering through these pages since they are not ranking well in SERPs.

If I now wanted to focus my SEO efforts on these 'invisible pages' – how do I even find them? In PIWIK and Google Analytics Landing Pages are only listed if they generate 1 entry or more per chosen time period.

I asked this question already over at the PIWIK forums ("How can I track entry pages with 0 (zero) entries?") but to my surprise I never got any answer or suggestion or whatever. So maybe this is not as easy to accomplish as I first thought? Any pointers or suggestions appreciated.

  • 1
    Nice and interesting perspective! Perfectly valid and valuable. I do not have an answer except one. If you spider your site just once, perhaps any page with just one view would stand out. Cheers!! – closetnoc Apr 9 '17 at 19:24
  • Thank you for your comment – and thank you for your suggestion! Yes, I can see the 1 view 'low performers' – and if I go to weekly or monthly stats I can actually find the very low performers – but still those zero-view pages are invisible… :/ So even if I get 99% of all pages listed in the monthly stats, that's still 10 hard to find low performers, when working on a 1000 pages website… – tillinberlin Apr 10 '17 at 8:29
1

Invisible pages, using your definition, will simply not appear in GA (or any other JS based analytical software). GA works only after its JS engine triggers and reports something back.

So, the best approach to this issue would be to run a crawler on your site, starting from home page. Let the crawler discover what you have there, then make a CSV file containing all unique URLs. Then go to GA, all pages report, and export everything that you consider to be "visible" or "performing well" - it is up to you how to define this, some suggestions:

  • all pages that had at least 10 visitors during last 30 days
  • all pages that had at least 50 impressions during last 30 days

After that you will have 2 CSV files - all pages (crawler) and all "visible" pages (GA). Then remove URLs that are in both files, which will give you only these pages that need some work.

I suggested a crawler because it is better than exporting directly from your CMS - crawler, in theory, should have the same access level as your users and search engine bots. It is safe to assume that if a bot cannot find a page, then users will not as well.

  • Thank you for your suggestion! However – I know I could do this outside of Piwik / GA – I already have a WP plugin installed that actually exports ''all'' posts and pages to a CSV file… But that wasn't really my point – I was looking for a way to to this ''inside'' of Piwik / GA – but maybe this is just not possible yet? Hmm… – tillinberlin Apr 11 '17 at 9:03
  • I'd say that yes, it is not possible with GA, unless you apply some sort of data import from somewhere else. As I wrote, GA can't tell you anything about pages it did not process yet - and such processing is possible only when somebody actually visits a given page (with JS enabled). – Bartek Apr 11 '17 at 9:27
  • well… If it wasn't google I would totally agree :) But even in Piwik there could theoretically be an option to either spider all pages, or to remember all pages that at some point were counted at least once (1 initial hit) in the past… I guess this should become a feature request then… – tillinberlin Apr 11 '17 at 12:14
1

It might be possible within GA to get 99% of the pages if not all (do you have search console linked to GA?)

Go to the Behaviour Report -> All Pages.

After the report is loaded for the desired timeframe, select the pivot table view enter image description here

then on the left hand side “Pivot By Medium”, after that sort the table by Organic. Done.

Consider some manual work

Alternatively, you can export the All Pages report and compare this result with the indexed pages in Search Console (aka Google Webmaster tools). There is no point to consider those pages that has not been indexed, they will not be receiving any organic traffic anyway, will They?

  • Thank you for pointing this out. I do have access to Search Console and I could probably list all indexed pages and compare the indexes – similar to what @Bertek suggested… Anyways – I was hoping there would be a way to index all pages from within either Piwik or GA… Still not giving up… :) Maybe somebody should write a plugin for that? …I'll try in the Piwik feature Request forum… – tillinberlin Apr 11 '17 at 8:59
  • 1
    Respectfully, I disagree with titico's comment - pages that haven't been indexed are even more important to work on - precisely to fix their indexation issues. I'd use Google's data only as something to mark pages that are (somewhat) OK. By definition, Google will only provide you with info about what it already found/indexed etc. - OP wants to improve pages that weren't performing at all. – Bartek Apr 11 '17 at 9:34
  • @tillinberlin the problem is that Piwik and GA are not Web crawlers. This answer will be the correct one if we can confirm that the "all pages" report is listing all the website pages where the Google Analytics tracking code has been inserted – Raul Reyes Apr 11 '17 at 22:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.