What is the best speedometer for keyword research

Subject & keyword research tools put to the test

Anyone who puts keyword research or keyword analysis in the sole focus of their efforts with regard to search engine optimization and tries to pimp their content with WDFxIDF has not heard the shot. For the right content strategy, you need relevant topics that support a user in his decision-making process. One possible way to get there, of course, is to identify relevant search terms and user interests, which ultimately result in helpful content with added value. In order to obtain relevant data, you can use Google's Keyword Planner as well as other useful keyword tools, of which we would like to introduce you to Keywordtool.io and Übersuggest.org today.

Fig.1: Keywordtool.io (left) and Übersuggest.org (right)

For the readers among you in a hurry, you can find the results here (in a short and direct way):

This is how the keyword tools work

At mediaworx, we use Übersuggest as well as the Keywordtool.io and only know the differences based on their type of result. Both tools are free of charge in the basic version. The Keywordtool.io also has an API that can be booked for $ 60 per month and with which semi-automated queries are possible. Unfortunately, with Übersuggest it is not possible to query YouTube, Bing and the AppStore, which is why the following test is limited to Google.

One of the main differences is that Keywordtool.io not only determines suggests whose long-tail extensions are at the end of the entered term ...

Basically, both tools return terms based on the well-known Google Suggest function that are also searched for or associated with the entered term. In addition to the initial term, the tools search for further terms in alphabetical order. The numbers 0-9 are also included. A maximum of 10 results are returned for each initial letter or number. Keywordtool.io also provides data on umlauts.


One of the main differences is that Keywordtool.io not only determines suggestions whose long-tail extensions are at the end of the entered term. For example, a query on “search term” would not only return “search term top 10”, but also “top 10 search terms”. In the case of Übersuggest, on the other hand, you would only get the variant “search terms top 10”.

But two research tools that seem to do the same thing? I took a closer look at that. Because I wanted to know which tool would give me the results from which I could draw the greatest potential in my everyday work. Of course, not only quantitative results of keyword research were on the agenda, but also quality. I fix this, for example, to the search volume (SV) that the determined search terms have.

Übersuggest & keywordtool.io - What was tested?

For the tests, I used initial terms from the fashion sector in order to determine the tool's suggestions based on this. A possible practical case could be the establishment of a fashion niche portal or shop. The following terms were used as an example to compare Übersuggest and keywordtool.io:

  • womenswear
  • menswear
  • shirts
  • Jeans
  • socks
  • soccer shoes
  • evening dress
  • tie

Who delivers the most data

Of course, while mass is not everything, having a large set of identified terms can potentially generate more ideas. Of course, based on the way Keywordtool.io comes up with suggestions, it has an advantage here. This is also shown in the following graphic, according to which Übersuggest provides fewer suggestions for all of the requested terms.

Fig.2: Direct comparison of the number of search terms found

In any case, it is noticeable that the number with Keywordtool.io fluctuates strongly, while with Übersuggest it remains constant.

All eyes on ... longtail

Eric Kubitz, among others, showed that the long tail can be extremely important for the success of a website at the end of last year with his article on the subject of "Linkless Ranking". Both Übersuggest and Keywordtool.io are predestined to identify meaningful long-tail terms. The following graphic shows who can apparently deliver the longest term combinations:

Fig.3: Average number of words in the search terms determined

Anyone who has paid attention will recognize that such a comparison using mean values ​​is only valid if the database is the same size. Unfortunately, since this is not the case, we put the data to a t-test, the results of which indicate how likely it is that one tool will return fewer words than the other. This probability is a maximum of 20.6% and on average less than 3%. This means that the two tools output roughly the same number of words in their suggestions.

Who provides the most relevant data

As already mentioned, not only the number of terms determined plays a role, but also their quality. We used Google's Keyword Planner to determine the search volume of the terms.

Anyone who has used these tools themselves knows that not all terms that are entered are also returned by the Keyword Planner. This happens because no search volume can be returned for some of the terms determined. In the following graphic, this is described as the Keyword Planner Pass Rate; i.e. the proportion of keywords that the planner returns as a result.

Fig.4: Direct comparison of the Keyword Planner pass rate

With approx. 76%, Übersuggest provides more unique search terms. The effectiveness of the data provided by this tool is therefore higher. Based on the sheer number of keywords with search volume, Keywordtool.io is still ahead of the game with almost 2,900 terms against around 2,200.

However, the efficiency of the search terms for the activities is also important. Because only if it is worthwhile to occupy these topics and terms does it make sense from an economic point of view to invest work in a concept, the implementation and ultimately the generation and distribution of content. A relevant metric for this is how many people are interested in the topic each month. The point at which this search volume is defined as high enough depends on the type of topic and of course also on individual goals. For the comparison, I have set a minimum search volume (SV) of 100:

Fig.5: Direct comparison of the number of search terms found with SV greater than 100

In 5 out of 8 cases Keywordtool.io returns more keywords with a search volume> 100. The yield of potentially worthwhile suggestions is therefore higher there.

Übersuggest or Keywordtool.io - which tool has now won?

For a topic research where you want to get many and relevant terms, Keywordtool.io provides the most data because of the higher number of keywords. The number of terms that are frequently searched for is also higher there. However, a distinction has to be made between a mass research and a topic-relevant search. Because for other topics that were not discussed in this test, Keywordtool.io may return irrelevant terms. This happens especially with keywords that cannot be clearly assigned to a topic. Here Übersuggest provides a much cleaner database, which is difficult to express in numbers. For this, a semantic evaluation of the terms would have to be carried out - and no, dear readers, there is no such thing at this point

There is no clear winner of the tools for keyword analysis presented here, but at least the knowledge that the greatest potential can be tapped if the results of the tools are coupled with careful competition and market research.

Ralf Schukay

Ralf is Teamlead Analytics and CRO at mediaworx. He helps companies to make their digital services more successful in terms of sales, of course always on the basis of measurable KPIs and with user needs and UX at the center.
He loves music, films, reading and jogging, plays Ableton and enjoys Berlin and garden life with his family.