Searching for your low quality pages
If your website has pages which rarely receive any traffic, then those pages could be holding your website back in terms of Search Engine Optimisation (SEO). Let’s investigate….
In our recent episode of The WordPress Show on the topic of SEO, our guest made a great point about Search Engine Optimisation (SEO) which often gets overlooked, let me summarise that for you:
Google form an impression on what our website is about from all of the pages it can find on our website – and if it finds a large number of poor, ‘low quality’ pages, then these could well hinder our SEO and drag the rest of the website backwards.
But what are ‘low quality’ pages?
Low quality pages could be pages which don’t have many words on them or are related to a topic which is no longer of concern to your business. Such pages could then be dead weight on your website, holding it back in the eyes of the Google search engine.
How do I find ‘low quality’ pages?
You need to head over to ‘All Pages‘ report on Google Analytics under the Behaviour/Site Content menu:
….next you’ll want to adjust your date range so that we are looking at a decent period of time (e.g. 6 months to a year+):
After that, Google should present you with a report view showing the most popular pages during that time.
However, we don’t want the most popular pages, we want to see which pages received the least traffic. To do that, simply click the ‘Pageviews‘ heading to reorder that column:
You’ll now see the same report but with your minimal traffic pages listed first. You can verify that by seeing a low number like 1 in the Pageviews column itself:
By all means, have a quick scan of your report to see what comes up but you’ll probably want to export that data so that you can work through it in your own time.
However, before you export this report, you need to show more data on that page (by default it only shows 10, and ‘export‘ only saves the first page), for that we use ‘Show Rows‘ dropdown at the bottom fo the report:
You probably want to show at least 1,000 rows (as you’ll see later, not everything listed in the report will be an issue you need to deal with so we need to throw a large net here to catch the actual issues).
Next, click the “Export” option near to the top right of the report:
You probably want to choose Excel, unless you are happier to use Google Sheets. Google Analytics will then allow you to save that report locally for you to work through.
Ok, but how do I work through these results?
When looking at the exported spreadsheet, what we are trying to find are actual pages on site which are not being used much by our website visitors. If our website visitors are not using them, then we need to ask why does the page event exist?
One issue with looking at this exported data is that there will probably a lot of noise in there which you need to sift through first before you get to actual cases e.g.
In Fig 7, I’m seeing a lot of ?fbclid entries, these are caused by clicks from Facebook (probably from a paid ad). By default, Google Analytics thinks that these are unique page requests, but they are not, the ‘?’ at the start designates that everything to the right of that is part of the Query String – think of it like a parameter or extra information we are passing in.
You can tell Google Analytics to ignore some of these query strings (I’ll show that in a future article) but the key point here is that you may see quite a lot of natural noise in your results which you may wish to hide or delete from your report as you strive to get to actual meaningful results.
Let’s look at some more examples of noise:
In Fig 8 we see more examples of single pages visits during that time period – again we are trying to pan for gold here amongst all the noise – if I drill into why some of these are being reported (and why that is ‘ok’) then perhaps that will help you pan for gold in your report. Let me just dig into those numbered examples a little:
- This page does not exist; I have a distant memory of testing out a new URL for blog (changing it to /articles/ from /campfire/). So this entry may be an artefact of that test and can be ignored.
- About 10 months ago I changed the URL format on my /campfire/ blog (where you are reading this article now!) so that it no longer included the published date as part of the URL page addresss. I’ll save the reasons for that for a future post but the upshot is that URL /campfire/2014/02/28/whats-wrong-with-a-cheap-website/ was ‘renamed’ (for want of a better phrase) around 10 months ago to /campfire/whats-wrong-with-a-cheap-website/ (note the missing date after the first /campfire/ bit)- but Google Analytics still needs to report on any views of the old name as well. Therefore, I can pretty much ignore anything with the old blog format in the URL.
- At first glance, I thought this was a simple mistake but it’s actually a bit more complex than that so I’ll discuss this one separately after this list.
- WordPress automatically makes available to Google all the CATEGORIES (and TAGS) which we organise our blog posts by. I don’t see any harm in this, it feels worthwhile to allow Google & people to find all of our content organised by whatever criteria we set (the only caveat is that we probably want to do some housekeeping on our CATEGORIES & TAGS to ensure they are still valid as, as you can see, that does impact Google).
Tip: power users here may well be thinking “can I go back and exclude obvious noise from the Google Analytics report?” yes you can, but I’ll avoid getting too into the weeds at the moment!
So let’s return to point 3, as there is a broader learning opportunity there: here we are talking about this URL /campfire/asp-products/45-min-1-2-1/.
I recognised the page right away as it is related to the simple website review service we offer. However, the URL looked odd – it begins with /campfire/ (i.e. part of the blog) when it should not be.
The first thing I did next was to copy and paste the URL and see how the website responds.
I was expecting the website to say 404 page not found (which would confirm to me that the above was just noise) but no, that URL does work.
I’ll now flag that URL to be investigated but the key point I want you to takeaway here is that, for any URL you are not sure of in your exported report, you want to simply paste it into Google (with your website address on the front) to see if it is an actual page or not. If your website comes back with a 404 error, then you may well be able to just ignore it.
Note: some of you may be thinking “hey Joel, am I not compounding the problem here by searching on these things again which will then also be recorded by Google Analytics?”. In which case, good point! Yes, you could be. In my case, I am logged in to the Glass Mountains website (whilst I’m writing this blog post) and my Google Analytics is set to exclude my traffic – so, in this case, I can safely search knowing I will not impact my own Google Analytics. I’ll make a note to blog about this in future but, for now, perhaps look at this article on that subject.
Auditing the results
Now that you’re able to filter through the noise on your report, you should start to see genuine pages on your site which are simply receiving very little traffic.
For these cases, you want to ask yourself:
Is this page still representative of what I want to be found for online? Is it useful to my audience?
If it is, great, can you update the page to improve it?
If not, what should you do? Well, it might make sense to delete the page as, if the page is really no longer valid, then we may as well remove it from Google and allow search engines to focus on our core content itself.
Note: I would never normally advise you just delete a page because ordinarily, you want to put a redirect in place so that Google know that the page URL has changed etc. However, in this case we do need to bear in mind in the effort involved – after all, these pages have hardly ever had any traffic! So if you deleted them and did NOT put a redirect in place, yes a future request would 404, but Google would eventually learn that the page has gone (especially if it vanishes from your sitemap). I’ll write an article on how to redirect in WordPress as the very next post (so feel free to subscribe!), if you need to know urgently, just hope over to the Facebook Group and ask me.
Google Search Console can also help with removing outdated content from the Google index – you just put the link in you want to be removed, and Google will handle the rest.
The Process
Your process should be:
- Find the genuine low traffic pages
- Decide what you are going to do with them
- Take action
If your report is long, then it will take you a while to go through it, however, this task should be something you dip into each week rather than thinking you need to get the whole thing done in one ago. We are chipping away at our website, nurturing it and maintaining it to help keep it healthy. I blogged about treating your website like an allotment previously and this very much carries on in that vein – we are ‘pruning‘ our website – removing dead leaves & flowers to allow the plant to thrive with its strength & resources focused on new, healthy growth.
Final Thoughts
One thing that occurred to me was that the report we are looking at here is showing pages which only had 1 page view – as that is the lowest number for the report to trigger. However, it is not showing which pages did not receive any traffic (which is not a fault of the report, it just can’t report on that) so, if you put your mind to it, you can do some data crunching and perhaps uncover even lower traffic pages (if I come up with anything, I’ll blog on that in future).
If you did not get a chance to watch the WordPress Show episode on SEO, I would give that a go. Nick will come back on the show in the autumn so please do keep an eye on our upcoming events.
Thanks
Joel
p.s. as ever, if you have any comments or questions, a great place to ask is in the Facebook Group.
Tags: Google Analyics, SEO
No Comments