It’s been a weird year for writing and content generation on this blog. I saw a huge surge in my YouTube traffic, and I’ve collected over a half a million views this year. People have watched over a million minutes of my content.
At the same time, I’ve seen a huge decrease in traffic from Google searches. I was pulling in 8000+ unique visitors a month at one point, and held the top 3 Google search rankings for a wide number of topics, from my Kel-Tec PF-9 handgun review to SentryPro XFC, a dog-killing topical flea and tick treatment, to several other things.
This year my traffic from Google has dropped by about 50%.
Here’s what I’ve done to try to fix things.
- Checked to see if the drop off was because of the site design from September 10 of 2012. That wasn’t it; I had a 54% drop off from 09/10/2012 to 09/10/2013.
- Checked to see if my top posts were still ranking high on Google; turns out a lot of my product reviews have been overtaken by Amazon-based reviews or high-traffic gun review sites. This is to be expected, but the most odd was my SentryPro review, which went from being the #1 result on Google to barely making the front page. My PF-9 review is buried on the fourth page now, it was also a first-page contender for many years.
- Investigated if the All in One SEO plugin I used to populate post meta data such as the title, keywords, and description of my pages was adversely effecting my ranking. In doing my research and looking at my stats I was concerned that Google’s Penguin or Panda algorithm changes were messing with my rankings, but it was hard to pinpoint any published change with a significant drop in rankings.
- I found that March 20, 2012 was the first day where my results started to stabilize from free-fall. Since then things have been consistent, with blips up thanks to incoming links from Reddit, StumbledUpon, various inkjet printer fix-it-guides, motorcycle and gun forums.
- Checked for server down time that might have kept users from viewing my content. Everything seemed okay.
- I found almost 1000 bad URLs spidered by Google. The vast majority of them leftovers from when I allowed pingbacks and trackbacks on Gibberish. Both of these things were supposed to help connect WordPress-based sites together via “backlinks,” that would theoretically help blogs get noticed and allow readers from one blog to find similar blogs. In reality, pingbacks and trackbacks were abused by spammers who would create a network of Web sites that could be used to promote someone’s real Web site — for a fee. I disabled pingbacks and trackbacks, but did not delete the references to them from Google’s search engine spider.
- I discovered that three sites were scraping my content and reposting the summaries on their own sites, with links back to mine. Bad backlinks strike again, and one of the documented effects of Google’s algorithm updates was to penalize sites that had were backlinked from known bogus Web sites … such as the ones I found linking to mine. I am not sure if this was enough to crush my site traffic by 50%, but I put in “disavow” statements to Google stating that I did not approve of these sites linking to mine. I don’t know if it will be effective or not.
- I looked at the bounce rates and times spent on page to determine if my writing got worse. Unfortunately my writing is just as shitty as it’s always been, and time spent on page, number of pages read by the average reader, etc are all about the same. On a good note, my repeat traffic has increased by about 4% since this time last year.
- Referral traffic was the same. I got the same amount of visits from other sources, so I knew that it was a problem with Google specifically.
- I re-evaluated my titles, just to make sure I wasn’t stuffing them with keywords, especially repeat ones. I was all clear here.
- I made sure canonical URLs were enabled so that I wasn’t getting dinged by Google for having duplicate content.
- I searched for unique phrases in my most popular articles to make sure someone else hadn’t copied them, resulting in a downgrade for duplicate content. No luck, aside from the summary-scrapers I mentioned above.
- I evaluated the load times for my site, since I read that page load time may have an effect on Google rankings. Thanks to moving to a virtual private server and updating my WordPress theme my site is performing faster than ever before.
- I moved all of my images to Amazon’s cloud servers for even faster page performance.
- My site does not have any warnings from Google; sometimes sites use sketchy techniques to increase their Google ranking and get busted by a human. These human-generated warnings are reported in Google’s Webmaster Tools, which gave Gibberish a clean bill of health.
- I generated an XML site map of the 1600+ posts I’ve written throughout the years. This should help Google spider my site more efficiently, but aside from the non-existent page errors from the legacy site I don’t have a lot of bad links or unreachable pages.
- I made sure my robots.txt file allows spidering of the site except for directories needed by the admin part of WordPress, the trackback and pingback subdirectories, and any legacy links from WordPress’s p? and cat? tags instead of using the article’s title as a URL.
- Unlinked about 400 broken links — the vast majority of them coming from the URLs provided by people commenting on my blog.
What if my writing just sucks?
Finally, I gave myself an honest evaluation of the content I was generating and the frequency in which I was creating it. There have been times where I wrote every day during the business week, and times that I didn’t write for almost two weeks. However, I wrote 146 articles in 2011 at the height of my blog’s popularity. I wrote 141 in 2012 and 166 this year.
The biggest content change involved firearms. I wrote 9 in 2011 and 40 this year. The reason for the shift was that we settled Sedagive?’s second child custody battle this year. Her ex-husband reads Gibberish and brought a series of articles I wrote to mediation, citing a risk to his children. Once custody firmed up, I started writing again.
Other topics, such as product reviews, game reviews, and music were about the same. I cut my motorcycle-related posts by about 80% given the shitty weather I live in now versus back home. I was getting some small but consistent traffic from my Motoport and Rev’It reviews, as well as my Givi topcase reviews. Not a 500 – 600 person per day drop, but I’m just trying to put everything out on the table.
So, what now?
I’m not sure what to do now. I have done my best to clean up the dead links spidered by Google, and disavowed association with known “bad backlink” sites. Increasing the performance of the site and remove the dead links in my posts should help everyone over all, but my site was never in danger of poor rankings due to its performance.
We’ll see; the drop in traffic and the associated revenue is significant enough I might hire someone for help.