As many of you will know I’m a web programmer and without sounding egotistical or arrogant I consider myself a pretty good one. Google certainly thought so too, since it ranked me on page one for that search term for a good number of years, and my ranking was achieved through hard work and white hat SEO techniques, nothing more, nothing less.
Having a programmer’s mindset though can have a downside, and my “I need to know” mentality one day got the better of me. For reasons I still really don’t quite understand I did something rather stupid. I destroyed my page one ranking with some black hat SEO. Intentionally.
I will hypothesise that my desire to do this was driven by a combination of factors. Reading article upon article of people asking how they can repair their Google rank post-Hummingbird certainly played a part, but for now I’ll state that I’ve narrowed it down to three possible factors, or perhaps a combination of them all:
1. I have an inherent desire to know how everything works and why it works. I’m pretty quirky like that, but it’s served me well over the years and I’ve built up a varied and loyal client base because of it. I like to de-construct things just so I can put them back together and truly understand the importance of each and every component.
2. I enjoy providing solutions to problems. By purposely destroying a perfectly good Google ranking for the term ‘web programmer’ with the intent (now there’s a word I’m hearing lots lately) of creating a blueprint that others could follow to fix it.. well, it provides me with quite a bit of personal satisfaction. Look what I did, Mom!
3. I had some kind of mental breakdown or someone spiked my coffee. Anyway, I digress.
Enough already. Help me fix my Google rankings!
Those of you who follow my articles will remember my original post, detailing the aforementioned experiment. Since that day I’ve performed a number of actions with my ultimate goal of being able to provide some kind of tangible measurement to each action.
This article will serve as an interim report detailing the first three issues I’ve dealt with, since I’m now acutely aware that repairing my search engine rankings is definitely going to take a while. It’s come to my attention that it wasn’t ONLY the black hat techniques I used that cause the issue. They merely compounded it.
The Repair Process: Steps One to Three
My starting point for this repair process wasn’t a good one. I’d managed to drop an incredible 359 places on Google in a matter of weeks, and to be perfectly honest, suddenly the experiment didn’t seem like such a good idea. But I had a plan from the outset, and here’s how it’s panned out so far.
1. Disavow, Disavow, Disavow!
The starting point for me was obviously to disavow all the poison links that I’d created. I purposely left older, pre-existing bad links to my website, because (1) they were there previously and didn’t seem to do that much harm, and (2) I’m going to remove them later and see if they also have a measurable value.
Two weeks after submitting my disavow file to Google, I saw some results. I’d moved up to position 239 in the rankings. Not what I expected at all if I am honest. Is some kind of penalty now being applied to my website? I’m not entirely sure, but I would have hoped to be higher than this.
2. Creating the Correct Balance and Structure
“Articles mean nothing without the correct site structure”
This issue really was an eye opener for me. During the early stages of the repair process I started adding a few new quality articles to my website (my popular 7000 word A to Z of SEO Guide: Post Hummingbird being one of them), and promoting them via social circles through all the authority figures I had access to.
They were shared pretty well and I received some great feedback from those in SEO circles. My intention at this point was to see just how much of a boost I could get from quality articles and some social networking in related circles, but what I uncovered was something much more important.
What I noticed was that Google couldn’t seem to make up its mind which of my articles to rank. It wasn’t ranking them all at once and day after day I was seeing myself in the rankings for a different article. Over the course of ten days I saw a total of six articles ranking for the term ‘web programmer’ and I started to wonder whether I’d over-optimised my site for the term. Was it really possible to confuse Google?
Furthermore, where I positioned in the rankings was directly related to the individual article Google served the user. I’d improve my rankings dramatically one day and see a particular article ranked, only to see myself 100 places lower the following day and another article listed. This told me two things – my site structure was somehow wrong and the weight I attached to each article when compared to my website’s main pages was also likely in need of a rethink.
To fix this I started looking over keyword density and how I’d optimised the individual articles. Thankfully I started to see a pattern and the more heavily optimised the article was, the more frequently it seemed to appear in the rankings. This is still being addressed as you read this, by good old fashioned trial and error, but I’m confident of finding the correct balance. Having a site that’s too keyword heavy is definitely no longer a good thing.
The next thing I looked at, since it seemed a natural place to look for me was my XML sitemap. By looking at the priority attribute in my sitemap I instantly saw that my blog articles had the same priority as my main services pages. The priority attribute in case you weren’t aware indicates the priority value of any given page in relation to others on your site. Valid values range from 0.0 to 1.0, with the default usually set at 0.5, and both my services page and blog articles carried a value of 0.7. By lowering the value for articles and increasing the value slightly on my services pages this should indicate that my services pages are the more important pages of my website.
But that wasn’t the only way I needed to address priorities and structure. Next on my hit list was to investigate dofollow and nofollow links, since I had to ensure the balance was correct. Again what I found was that there seemed to be no real structure for Google to follow. How could I be so sloppy?
The following Moz guide should point you in the right direction should you have confusion concerning internal link structures:
Once again this is something I am still working on, using nothing more than good old fashioned trial and error. There are easier ways to do this but as I mentioned previously, I want to be able to quantify my results. Results that can’t be measured just aren’t worth having in my opinion.
That said, even while I’m still re-optimising I noticed a leap in the SERPs from around the 230 mark to 140 or lower. Definitely on the right track.
3. A new semantic based approach
The third part of my process was to re-write some content, with a much more semantic based approach, moving away from the keyword heavy content I had previously adopted.
To do this I firstly went through page by page using keyword density tools to not only gauge whether my search term (web programmer) was prominent, but also to ensure it didn’t appear too heavily.
As a sidenote, if you don’t have the tools to do this, there are plenty of keyword density checkers online. Here’s one to get you started:
Once I was satisfied I had a healthy balance, it was time to re-word the copy itself. This is something you have to be very careful with since Google doesn’t respond well to large wholesale changes. Simple and well written changes are the order of the day here.
But what should you change?
To ascertain which blocks of text needed changing I used perhaps the simplest tool of all – Google itself.
I simply typed in a few semantic sounding phrases into the Google search box and listed the suggestions I was receiving from the search auto-complete. Those phrases were then integrated into my text and one week later I started to see results. Improved results.
I’m not saying results are spectacular, because as of the time of writing I am still seeing many of the aforementioned issues present. I’m still seeing different articles every day or so, which means there is still work to be done re-structuring, but my ranking is now hovering anywhere between 35 and 80 which isn’t bad when put into perspective – I’m competing with over 47 million pages after all.
Semantic searching of course is much more involved than that, and I’ll go into greater detail of my thinking and research later, but this was my starting point and I’ve kept it simple since tangible results followed, even using such a basic and primitive method.
Heads up! For a little more on semantics, I recently came across this great article illustrating entities and relationships.
So what for the future?
If I am honest, on a personal level I’m finding this whole experiment quite fun. It’s mentally stimulating to say the least and the more knowledge I gain the more I want. I’m also getting to interact with some very knowledgeable and interesting people along the way and I’m happy so many of you have accompanied me on this journey.
However, mental stimulation and new acquaintances aren’t enough for me alone, and I will not stop until I am back on page one. It’s simply not an option to me and next up is what could become an extremely time consuming stage of my recovery – I intend to dramatically improve my authority levels in the industry by covering all the bases. Not just one or two.. ALL OF THEM.
What I have noticed through research is that Google in its early stages post-Hummingbird is getting things very confused. Intent was a buzz word they banded around frequently whilst explaining the new algorithm, but they don’t seem to know MY intent.
I’m being served up web programmer jobs, web programmer courses, web programmer salaries, and a plethora of articles detailing the difference between web programmers and web developers, and I don’t want any of those things. I want to see a web programmer on the list and I want that programmer to be me!
The only real solution to this as far as I can see if to diversify. I’m going to have to produce semantically written articles on the aforementioned subjects, and they’re going to have to be better than the ones that exist there at present.
It’s not ideal and it’s going to take a while, but until Google knows what I am searching for I don’t really see any other way.
Final point of interest
Something else I have found along the way that may be of interest is that Hummingbird only seems to apply to the very first tab on Google’s website – the ‘Web’ one.
I’m actually still ranking pretty well on the other tabs – I see my site’s logo for example in Google images along with some images from my site.
I also appear to be the very first result on Google Maps, have a video on page one of the Videos tab, and I’m top of the pile where Blogs are concerned too.
That’s something at least!
And my thanks to you
Finally, I’d just like to say to those who helped out with the social sharing, I sincerely thank you for your shares, likes, +1′s, tweets, and pins, but more importantly I really valued your feedback and comments.
This is only the beginning of my recovery, but through hard work and your support I’m sure I’ll not only get to where I want to be, but also uncover some valuable nuggets of information along the way for you all to use.
Thank you all. ( Now please be a gent and click that share button )