Determining reasons for the SEMrush Keyword Magic Tool’s most peculiar results
Founded in August 2008, SEMrush is now established as one of the most popular SEO tools globally, and it holds more than just one use. Below, Rob Lambert highlights some of the more unusual features and explores some of the quirky keyword searches made by users…
From the multitude of features on SEMrush, you may have come across the Keyword Magic Tool, which allows users to search and examine the search volume and other relevant factors for chosen key terms. You’re also given the option to narrow down your results to solely see questions, effectively singling out the phrases that start with the five Ws of who,what,where, when and why, as well as other alternatives such as am, are, can, do, does, is, were and will.
When it comes to user focused content, targeting search queries that are relevant to certain businesses, their website and their audience couldn’t be any more important. This is down to the nature of this form of content, as it is aimed at picking up the traffic of search engine users through blogs by including appropriate questions with adequate search volume.
A key part of conducting user focused keyword research involves wading through the masses of insignificant results with a view to finding the types of questions with sufficient volume that you were looking for. Although many of these results are only left out due to being completely irrelevant to the research you’re carrying out, some are funny, strange and even disturbing.
These types of results can appear at any time during user focused keyword research regardless of the topic or business sector, and in some cases, they may even be worth considering if they work with the style of piece you’re creating or if the search volume is far too substantial to ignore.
Where does SEMrush data come from?
During the course of past keyword research projects, we’ve found a number of these bizarre searches in SEMrush’s results, but what is it that could be causing them?
For example, if you come across ‘how to talk to your dog about gun safety’ with an average of 10 searches a month, is this an inside joke by SEMrush, a total misunderstanding, or are 10 people genuinely looking for the answer to this question?
On the SEMrush website, an FAQ section details how data is retrieved for many of the website’s features. This information explains that SEMrush uses a machine learning algorithm and trusted data providers in any of the data presented in the databases. It goes on to say that different techniques are used to match the varying information provided on the SEMrush website and that their data is regularly updated and regulated, with the guarantee that the same level of quality is prevalent in all of their databases.
How did these weird questions appear in your search results?
While this level of reasoning straight from the mouth of SEMrush helps to find a basis behind how the website collates data, it doesn’t completely clear up some of the more peculiar results.
To find an answer to why strange questions appear in the SEMrush Keyword Magic Tool, the most crucial factor we looked at was logic in the way that Google works, as well as the way in which the human mind works.
We split these weird questions into three categories:
- questions that are either easy to answer or answer themselves
- questions that immediately raise alarm bells through being controversial or underhand
- questions that are almost strange beyond understanding.
What they are:
In its simplest terms, this type of question is a form of guidance requested by the user, but the answer is either blatant or already answered within the question. There are different variations of this type of query, but the most common is asking how to spell a question while ironically featuring the answer to the question in the question itself.
Now, if you haven’t seen Bruce Almighty, you may be unaware of how to spell beautiful, but a useful tip would be to read how you spelt it when you first questioned Google.
Why they appear:
Not only are these types of questions extremely common, but the likely reason is suitably straightforward. When you start typing out this question, Google will predict the correct spelling of the word you’re looking for even if you’ve spelt it wrong during the search. If you don’t click on the corrected version at this point, it will also appear in your search results.
As the Keyword Surfer plug-in on Google Chrome clarifies, ‘how do you spell beautiful’ has 1,600 searches a month, backing up SEMrush’s data, but ‘how do you spell beautigul’ has 0 volume, indicating that it is searched for, but maybe not recently.
What they are:
Questions within this category tend to only raise further questions. Many examples of this will hint at the concept of breaking the law or doing something dubious, and whether it’s possible for the person asking the question to receive guidance or advice on how to carry out the task.
A common example is queries that appear to indicate that the searcher is looking to fake an injury in order to claim compensation.
Why they appear:
Although assuming a dodgy compensation claim is likely to be the first conclusion you’d come to after finding this type of query, it could also be the case that it has another origin, with websites using the keyword as a guide for learning the trade as a professional stuntman.
The heavily specific nature of these types of keywords only makes them more dubious. You may have found a strange keyword that hints at faking an accident, but a more extreme and questionable example would be ‘how to remove blood stains from a mattress’, which returns an estimated 480 searches a month.
A mixture between the hefty search volume and the extremely specific query make this result unsettling, especially as you wonder how around 480 people are struggling with removing blood from a mattress on a monthly basis. However, you may be surprised to hear that many companies that specialise in stain removal use blood as the perfect example substance due to how difficult it can be to remove. The result is admittedly worrying, but there are other reasons to remove blood aside from murder.
What they are:
On the rough scale of weird search results in the Keyword Magic Tool, these are the results that defy logic and require an extensive explanation to suitably reassure you. Many questionable questions fit into this category, but with the exception of being beyond understanding to a point where the reality of the question is enough to keep you up at night.
Why they appear:
If you’re unfortunate enough to come across a query as utterly terrifying as ‘how to summon a demon’, you’re well within your right to be unnerved by it. However, after looking into this specific keyword, there is a perfectly simple – and relatively disappointing – reason why keywords like this appear.
From the 1,000 people looking for the answer to this query, some may be genuinely looking for instructions on summoning demons, but for the most part, many of the searches are likely to be related to Japanese anime novel series ‘How Not to Summon a Demon Lord’. More often than not, odd questions that are heavily specific or downright odd are actually the title of a humorous novel or another form of media. Not only does this apply to ‘how to summon a demon’, but it also applies to the aforementioned ‘how to talk to your dog about gun safety’ – a reference to satirical novel How to Talk to Your Cat About Gun Safety by Zachary Auburn.
Have you found any peculiar results using the SEMrush Keyword Magic Tool?
Despite many results being without explanation or reason, there are a lot of searches that can be justified solely using logic. You wouldn’t expect to find anything particularly entertaining from conducting simple keyword research, but it can be a weird, confusing or funny perk of it.
We’re interested to hear about any odd keywords you’ve come across during your own keyword research. If you find any, send them to us on Twitter at @ClickConsultLtd including the hashtag #KeyweirdResearch, and it may be chosen for our weekly round-up of the best results found during keyword research.