Trying to improve SEO ratings

I’ve been writing my photoblog for about 4 years now, initially as a google blogger site, then as a wordpress site, and for the last couple of years as a wordpress site with it’s own domain. In that time I’ve seen the traffic move up steadily from only a few 10’s of visits when I started to about 3,000 visitors a week and about 20,000 views as measured on the wordpress stats tool.

Well, that was true until about 2 weeks ago, but then something happened and for the last two weeks my visitors and views have been dropping to the point that I will be getting about 800 less visitors this week. On top of that change, there seems to have been a change in the way that WordPress count gallery views, so instead of getting an average of 6 to 8 views per visitor that number has dropped to about 2 – 3. Update: An email to the wordpress support people has informed me that the individual views for gallery pictures is removed once 3rd party plugins are installed, so I caused that issue myself when I added Yoast SEO.

As with all people who write for fun rather than profit, this isn’t a serious problem which I will lay awake at night worrying about – I don’t rely on the site for any income – but it is quite discouraging to see the graphs just tail away when they had been improving.

Finding the problem

So, over the last couple of days I’ve been trying to sort out what has been going wrong and what, if anything, I can do about it.

The first thing I did was had a look on my Google Search Console results to see if I could work out what the problem is.

The picture on the right shows the view I set up to do this. This is the Google search console Search Analytics page and shows clicks and impressions for my site, with a comparison set up for the period when the problem stared and the couple of weeks prior to that.

It’s possible to see from this graph that the number of impressions I’ve had has dropped from a maximum of about 7000 a day down to a current level of about 4000 a day and because of this the number of clicks has also dropped from about 400 to about 340.

At this point it might be worth pointing out what these terms mean.

An impression is a result in a search request that someone has performed. So, for example, in the picture above the top query is shown as ‘pentax p30’ – if you type that query into the google search page you will find that one of the results you get will be a link to my post on the pentax p30. In fact I’ve just done that and you can see the page on the left  – in the number 4 slot is my link.

So what I’ve generated there is an impression and it counts because the search result appeared in the user’s page of results. If the result had not been on the first page, but on page 2 it would not have counted as an impression unless the user actually looked at page 2. That is an important point – the impression only counts if it’s on the page the user has seen. It doesn’t have to be ‘above the fold’ ie the user doesn’t have to scroll it into view to see it, but it must be on the page returned from the search.

A click is recorded when someone chooses my link in the results page and visits the article. As I said above, the impression is recorded even if the result isn’t in view, but the click only happens if the user sees the results, so another important part of this is where on the page your impression sits. If you have the number one slot in the results you will get significantly more traffic than if you are number 2, and if you are number 2 you get more than number 3 etc.

So the next thing I looked at was if my average position in the results had changed, or if the click through rate (which is the ratio of number of times clicked to number of impressions gained) had changed and the image to the left shows those numbers.

When I looked at the graph I could see that there had been a marginal change – the Average Click Through Rate had gone from 6.19% to 6.01% and the Average Position changed from 12.7 to 13.1. Unfortunately, all these numbers really seem to confirm was that I’m getting less impressions and that could be because more results are returning on the second page or later. Also, changes in an average which is made up of thousands of urls (and when I count up the number of posts I’ve written and pictures uploaded it is in the thousands), are probably the result of a large number of small changes.

Improve SEO ratings

So the analysis I had carried out led me to think that in order to fix this problem I needed to improve the SEO rating of my posts so that I get more results on the first page of the google search results. And of course, that is the problem that everyone is trying to fix.

My first though was to try to use the Yoast SEO plugin to try to analyse some posts for a particular query and check if the SEO was considered to be good or bad. I tried this on the term above ‘Pentax P30’ and this is shown to the left.

As you can see there are a few suggestions here that I could do to improve this article and hopefully make it more attractive from an SEO point of view and I started to make some changes to a few articles.

After I’d done a few changes however it suddenly occurred to me that the search query I’d chosen ‘Pentax P30’ was probably not the only one which resulted in an impression for my article. So I went back to the search console and did another check.

On this occasion I used the same Search Analytics page but I set up a filter on the url of the page on my site which contains the Pentax P30 review. With that set I then looked again at the queries which have resulted in an impression and click over the last few weeks and compared it to the same period a couple of weeks before.

And this page told me something interesting.

The view is shown above to the left and it shows that for any particular page on my site there are many terms which result in an impression and therefore a possible click. This of course suggests that trying to ‘tune’ the page for one particular query is a waste of time – if you improve it for one you will probably ruin it for another.

In this particular instance there are 85 individual queries, and although the top one is bigger in terms of impressions than all the others, there are still a number of different terms which result in meaningful search result impressions.

As I was looking down the numbers however, I noticed that there seemed to be differences towards the lower numbers. The image above on the right shows what I mean. The earlier period had a number of impressions generated from queries which are only tangentially related to the subject matter of the post. For instance, 4 impressions for ‘pentax mz-m review’ and 12 for ‘pentax 35mm slr camera’. On the newer results these impressions are not recorded at all. In total I counted 67 impressions which appeared in the earlier period but not in the later period.

So I think this has lead me to the reason for my drop in visitors. Google have tightened up their search algorithm and some of the impressions I was getting have been removed. Although I don’t particularly like it, that is actually a good thing I guess because it means that the search results are better targeted at what people are looking for.

At the moment I’m carrying on looking to see if there is anything else I need to do to try to restore my results – probably it means just means editing some of the more sloppily written posts to improve them, but I’d be interested if anyone else who runs a vintage camera site has had similar experience in the last few weeks.