As we all know, keyword research is among the most important parts of any SEO process. Numerous tools and methods are available for performing optimal keyword researches. However, most SEO agencies employ their own unique methods to select the best and most bankable keywords for their clients.
Recently, I happened to read an interesting post about the distribution of website visitors according to the number of words in the original search phrase. This article mentioned (as shown in the graph below) that the majority of new visits came from two-word search phrases.
In light of this variance, how can an internet marketing company choose the best keywords and phrases? Single keywords? Two word phrases? Perhaps a three-word phrase is optimal. Maybe more. Will this statistic be the same for brand queries? So many questions need to be answered before we can approach the research tools, begin examining monthly search data, etc.
Since we all know that conversion, and not the number of visits, is what truly matters to our customers (the actual number of visitors who made purchases), we’ve decided to take this research one step further and examine which phrases ultimately yield the highest conversion.
A little background…
There are two possible research methods that can be used to guarantee the credibility of this research and provide reliable conclusions:
- Perform the tests on a large number of websites in numerous fields and in various languages.
- Perform the test on a small number of websites with exceptionally high traffic, preferably those that contain sub-domains in different languages.
Because the first option would take longer, and because we would need access to a number of clients with the required large visitor volume, we decided to go with the second option. Data was drawn from two large websites which contain content in over 10 different languages and reflect visitor-behavior in numerous countries around the world.
The following graphs refer to these websites:
Anyone in the internet industry, and in the SEO industry in particular, knows that longer search phrases are better and more focused. Or are they?
Branded Vs Non Branded
The research we conducted at Dynamic Search showed that the above is true, but only when dealing with branded phrases. When we shift to non-branded phrases, we see a rather different result.
The following charts, screen caps, Google Analytics and Google Webmaster Tools data show the distribution of traffic among the two types of searches – branded and non-branded – while emphasizing factors such as CTR, bounce rate and ultimately, traffic volume.
The graph below shows that there is a significant difference between the number of visits originating in branded and non-branded searches. While for a non-branded search the distribution of searches is similar to that described in other studies – the longer the search query, the more traffic it yields – the result differs for the branded queries. We can see that 2-word long branded queries will yield more traffic.
However, a close inspection of the CTR graph below shows that CTR tends to grow in positive relation to the number of words in a branded search phrase. While the bounce rate for non-branded queries remains almost the same as the number of words in the search query increases, a longer search query in branded queries yields a higher CTR .
The bounce rate graph below demonstrates that the bounce rate significantly decreases as the number of words in a branded search phrase increases. An astonishingly low bounce rate (14%) is one that I am positive any website owner would be happy with.
Effectively, these results illustrate a profound change in the SEO industry. The primary goals of internet marketing and SEO companies used to be to increase the number of unique visitors to a website and to essentially increase its visibility. Added to these was the goal of placing the website on the first page of search results for key search phrases.However, those days are history. Business owners have caught on and are no longer satisfied with impressive monthly reports indicating a 100x increase in website traffic. They realize that, in the past, they were paying large sums for a service that may not have been justified.
Today, few SEO professionals think that a website’s traffic volume or rank in search results are the true measures of successful SEO. They know that the truest measurement of successful internet optimization is an increase in revenue. Business owners and SEO companies understand that, unless the client sees an increase in revenue quickly, he may back out of the contract and cancel the service.
Optimal Search Phrases and Keywords – Length Matters!
How can SEO companies catch up and deliver results so quickly? After all, organic optimization takes time. The data shown above enables us to conclude that, if we opt to promote non-branded phrases, we can safely choose almost any search phrase since there is no significant difference between the search lengths other than traffic (which can change according to the nature of the website). However, we are of the opinion is that it is best to focus on a maximum of two- or three-word search phrases in order to get the best results.
An additional advantage we get when promoting two- or three-word phrases is their relative advantage over ‘heavy’ generic words. These are the most important, highly competitive words that everyone chooses to promote. Promoting these words may take valuable time and may even require a higher fee for the service, while the return may not be satisfactory or of equal value.
When dealing with branded phrases, things are a bit clearer. If we opt to promote branded phrases (naturally, only if our client is solidified as a brand) we start to see things changing. Incoming traffic originating from single-word branded searches is relatively high compared to incoming traffic originating from a branded three-word search. Nevertheless, the CTR from a three-word branded search phrase is extremely high (33%) and the bounce rate is extremely low (14.6%). This yields the conclusion that three-word branded search phrases result in less traffic overall, but are more relevant and of higher quality. To clarify, this may mean that, occasionally, it’s preferable to promote higher quality, more focused phrases with a lower visitor potential than to target popular phrases that may increase irrelevant traffic.
Our natural tendency is to fire in as many directions as possible in order to increase our chances of hitting something. However, in light of the above data, it is clear that, ultimately, it’s better to focus your fire at a well-defined target. When we spread ourselves too wide, we may increase traffic. However, our increasing bounce rates may lead Google’s algorithm to ‘think’ that our website’s relevance to the popular search phrase is low, causing it to decrease our ranking and overturn all of our hard work. To sum up, by promoting highly competitive, popular search phrases, we put ourselves at a double risk. First, we risk a low CTR. Second, we risk a decrease in ranking that will definitely lead to a decrease in hits and to an even lower CTR.
Finally, I’d like to share some non-branded search data. In the following example you’ll see how two different search phrases from a highly competitive field (and for which the website is ranked in the first page of Google search results) yield completely different results than expected. For the first search phrase (composed of one word), the website is ranked first on Google.com. Google Keyword Tool shows that the number of monthly searches for this phrase is 30,400,000. However, Google Analytics shows that the number of monthly visitors is only 9,779. This is a very low CTR, especially since the website is located first in such a sought after and competitive search phrase.
For the second phrase (composed of three words), the website is ranked third on Google.com. We can see that the number of monthly searches is only 1,000,000. However, Google Analytics shows a much higher CTR of 108,524 visitors per month. Needless to say, this higher CTR is potentially more profitable compared to the first generic and highly competitive search phrase. It is clear, however, that more resources and efforts were put into promoting the first search phrase.
Hopefully, after examining the data, you are feeling more certain about the keywords and phrases you’ve chosen to promote. However, we’d like to make an important comment here. CTR is influenced by a number of parameters, and may not be the direct result of the original search phrase. For instance, CTR may be affected by the simple fact that, every now and then, Google changes its SERP.
If you’re interested in learning more about this subject, you’re invited to read a detailed research article written by our friend and colleague Branko Rihtman. You can find the article on his blog at seo-scientist.com
Link to this page
Copy the code below to your website.