Building your communications strategy around key terms used by younger twitters

If your product is targeted to young consumers which use twitter as a main communication tool, then you can learn more about their behavior. Mira Analytics is able to search the most frequent terms used by your customers analyzing thousands of tweets. Additionally, you can understand what are the most associated terms (pairs of key words) by looking ate the strongest relationships In order to do so, you can apply standard taxonomies or build your own.

Challenge
In this particular case, the challenge was identifying which are the most used “creative” terms related to the brand Yatekomo and how these terms are co-mentioned together.

Methodology
In order to do it, we created a list of concepts organized in several category such as: brands, creative terms or concepts in general. We searched tweets about Yatekomo, using the brand name as well as the hash tag #Yatekomo. The analysis is based on 43,000+ tweets collected between November 2015 and May 2017.

Results
The following figure shows the most frequent words mentioned in the tweets as well as the most frequent concurrency of words. The violet color shows the main brands available in the Spanish market and the orange nodes represent the creative words. Edges are colored using a mean value of the node’s color.


The network graph shows that the most frequent words related to the brand Yatekomo are “Chino” and “Fideo”. If you look at the other brands, you may see other type concurrency. On the other hand, if we look specifically at the pairs of creative words, Sevillano-Coria or Chino-Fideo concurs. The graph also provides data about the time (date) of each tweet. If data are filtered by time (using the time line) you can see how these relationships changes during the time. The previous analysis was made on the total tweets collected and thus does not provide an evolutionary analysis. However, with the timeline you can do it.

An additional analysis was made using the Maximum Spanning Tree (MST) algorithm. This algorithm reduces the number of nodes by pruning those that are less relevant.  The following figure shows an example although the MST is most effective when a lot of words with different meaning are analyzed.

Take-away
Understanding which keywords are used and how these are connected together allow you to build actionable target oriented digital marketing strategies. For instance, you could use these words for SEO, persuasive online advertisement, creating claims in the web page or on-line packaging branding. Additionally, you can use these insights into the traditional marketing process, innovation team or support better communication strategy.