Open and shut case for legal firm’s Google optimisation...Continue Reading
As someone who has spent over a decade in SEO, I’ve seen a range of ideas, fads and urban myths come and go, but I have to confess, when someone asked me this last month, it made me stop and think. They had created a piece of content for their website and were concerned that they had overused the keyword for which they wanted to rank. The big issue in their mind was that they were likely to get a Google penalty if it was wrong.
I’ve never seen a customer website with a penalty for keyword stuffing, so I have no first-hand experience in this and perhaps, this lack of experience coloured my initial reaction. So, despite never working on a customer site with a keyword stuffing penalty in the last ten years, rather than go with my gut reaction I set out to do some research and establish whether this could be something real.
The starting point is that there are several issues here, all intertwined, beginning with what is keyword stuffing and ending with can it get you a Google penalty? This is how I started to pick this apart.
Keyword stuffing is, quite simply, the overuse of a keyword (or phrase) on a single web page with the aim of manipulating the search results. Historically, this had been used to help websites rank in search, but the advantage of this pretty much disappeared at the turn of the century.
Google give a pretty good description of this on their developers’ page, where they say;
Examples of keyword stuffing include:
We sell custom cigar humidors. Our custom cigar humidors are handmade. If you’re thinking of buying a custom cigar humidor, please contact our custom cigar humidor specialists at firstname.lastname@example.org.
What isn’t clear though is how much is too much? And does the keyword have to appear on the page, or does it include the text in the source code? Is it based on the way a person is likely to react to it or is there some kind of algorithm behind the scenes monitoring this?
On all these good questions, I came up short. Despite extensive searches, I was unable to find a conclusive answer to any of them. Sure, there is a lot of speculation, but very little in the way of hard evidence.
Partly this is because Google will not publish this type of information. If they told people that seventeen mentions of a keyword on a page were fine but eighteen were too many, then every website owner would put seventeen mentions on the page, irrespective of whether it warranted or needed it.
Some websites have decided that there is a figure, which they express as a percentage typically ranging from 2-5%, but even though they state this is the correct figure, it’s not sourced so we have no way of establishing where this figure comes from nor why it is correct. In truth, trying to nail it down to a single figure is pointless, as it will vary site by site and page by page.
The other issue is that there are so many places you can put keywords that it’s hard to know where to stop counting.
When you create a new page, you can easily insert the keyword you want in the following places;
This isn’t a complete list, as there are far more places keywords can be included, including within url’s (including those in the css).
As an example, I have just looked at the home page of a website I own and run and the word for which I want to rank appears no less than 30 times. More importantly, if I look at the source code, I can find it a staggering 377 times as it’s in every single url.
This website ranks number one for pretty much every variation on this keyword and is an authority website in this field. I have used it as a testbed for many years, to establish what does and doesn’t work in SEO terms and as it ranks top for this keyword, despite 377 mentions on the page, I think that we can discount the notion of it being a simple count.
We have established, using the example above, that it is not just a simple count of the number of times a keyword appears, so the next logical area is context; does the keyword appear contextually correct to a user?
As we can see from the Google example about humidors, stuffing the same word repeatedly into sentences is not good from a user perspective. It was precisely this type of content which persuaded Google to act back in 2011 when the introduced the Panda update. If you want a history lesson on what the Panda update was and is and how it changed search, Search Engine Journal have produced a very good guide that explains everything.
The bottom line however was that poor, low quality content, including content that simply repeated the keyword over and over, was diminished, and their value in ranking for certain keywords disappeared overnight. Quality control meant that sites producing low quality content, including article with high keyword repetition, were devalued, and dropped from search.
A great many of the websites that relied on this sort of poor quality writing, with keyword stuffing, complained bitterly at the time that they had been ‘penalised’ by Google but in reality, they hadn’t. They may have lost much of their traffic but what they were doing was pointless, irrelevant, and diminishing the value of search, rather than helping it. Panda was not about penalties, more about rewarding good quality content over poor.
On this basis we should conclude that Panda was not a driver of keyword stuffing penalties, as though it may have contributed, it was not primarily a penalty mechanism.
We’ve yet to nail what could cause a keyword stuffing penalty, and by this stage I am starting to think that it might not exist. Could this be an urban myth?
Searching Google for answers to this, the top, and most authoritative articles on this, all fail to give a concrete example of a website that has received a specific ‘keyword stuffing’ penalty. The nearest I can see is a list of all the likely Google penalties from Search Engine Journal, but even they don’t give an example of a keyword stuffing penalty. Their example is bundled in with ‘hidden text’, which was an old trick of, for example, placing white text on a white background (or black on black) so that search engines would read it as part of the code but visually, a user would not be able to see it.
Forbidden from the start, this was something Google made clear in early Webmaster Guidelines that sites deliberately using this technique to manipulate search results, would be penalised.
Today, in Webmaster Guidelines, Google list a range of things you should avoid and although not a definitive list, it does include;
Any or all of these can be responsible for triggering a Manual Action, which is Google’s term for penalties, but nowhere does it specifically mention a keyword stuffing penalty.
In the past decade I can honestly say I can only recall seeing a handful of these Manual Actions and none of them through work we had done. Most were on websites whose owners approached us to help them recover from a manual action, usually around backlinks, not content.
If you search Google today for examples of keyword stuffing penalties, the top results that are served are all from 2012 and 2013 and none of them show concrete examples.
In short, it doesn’t look like keyword stuffing penalties actually exist. Could they be just an urban myth?
Whilst it is clear from the research that Google will take action against Irrelevant Keywords, or Keyword stuffing, there don’t appear to be any examples, certainly not recent ones, of anyone actually being penalised.
Does this mean that Google don’t act on this? Unlikely, it is probably more that no one bothers with stuffing keywords onto a page these days, as it won’t help their website or page rank in search.
Certain website tools, like the Yoast SEO plugin for WordPress, include a keyword density checker to let you know, when you are writing an article, just how many times a word or phrase appears on the page. Their upper limit for keyword density is 3% (or 3.5% in the premium version) but again, there is no reference to where this figure comes from.
Tools like this, which offer specific guidance, are helping authors and website owners from overloading their articles with too many mentions of a particular word, but in truth, as Google won’t rank irrelevant content any more, this is largely unnecessary.
Google are clear that you should not overload articles with too many repetitions of a particular keyword. They list it as a clear breach of Webmaster Guidelines, under Irrelevant Keywords.
The reality of search today however is that it doesn’t appear to matter how many times a keyword appears, it’s more about how users read and react with that page. On that basis, virtually no one is keyword stuffing as it serves no purpose.
Looking back, I can find no concrete examples of websites that have received a manual action for keyword stuffing and certainly nothing after 2013.
There are plenty of tools that exist to help guide people on how to write good content, but none of them explicitly tell you where they get their ‘guideline’ numbers from. Articles continue to warn against ‘Google Penalties’ but with so few websites actually getting them these days there is scant evidence that anyone receives a penalty these days for keyword stuffing. That’s not to say they don’t exist, more that they are so rare and so infrequently written about, that there are no examples available to study.
The trick is clearly not about how many times your keyword appears on or behind a page, it’s about building something that works for users. The better your content, the bigger your user satisfaction and focusing on user experience and utility is clearly a better use of time and resources.
So, next time you are writing an article and are working about the number of times a keyword appears, just read it out loud. The acid test will not be around the number of times a word appears, but whether you bore yourself before you get to the conclusion.