Does A Brand’s Popularity Increase After A Celebrity Campaign?

Jenny Force Jenny Force, VP of Marketing

Editor’s note: This post highlights the social media analytics research conducted by Noshin Anwar, a Ryerson University student, as a final project for the Applied Retail Research course (RMG700). This is an example of the work and learnings resulting from the Sysomos in the Classroom program, a partnership with higher education institutions.

Are you a professor looking to use Sysomos in your higher education classroom? Apply today!

Experimental Research: Increase in Popularity of a Brand After a Celebrity Campaign


Utilizing celebrity brand endorsements are a common practice in the retail industry. This research report examines whether there is an increase in retailer mentions (#tweets) and popularity of a brand on social media after a celebrity campaign is endorsed through a social platform.

Using Sysomos, this study will be conducted using the social media behavior of a retailer, PUMA, and their recent celebrity campaign with Kylie Jenner.


Hypothesis: A celebrity campaign (Kylie Jenner’s campaign) promoted on a social platform (Twitter) will lead to an increase in conversation on that social platform regarding the promoting retailer (PUMA).

Methods: The research was conducted through the use of Sysomos’ software and PUMA’s official Twitter page. The official release of Kylie Jenner’s PUMA campaign was determined through the use of PUMA’s official Twitter page.

After noting the campaign release date, March 13, 2016, Sysomos MAP  was used in order to see the amount of tweets that were sent per day 19 days prior (February 24 – March 12) and after the campaign was released (March 13 – March 31).

The search query of  (“PUMA” OR “#puma” OR “@puma” OR “#PumaFierce”) AND NOT “animal” was used to analyze the data in the mentioned time periods above.

The “popularity curve” was used to determine per day numbers, which were used to create a dataset. Then, an independent t-test (statistical examination of two population means) was conducted to see the effectiveness of social media.


Results: This study has found that there is no significant difference in social media conversation prior to and after a campaign release. The t-test revealed no significant differences, thus accepting the null hypothesis (before tweets mean = 15,205, after tweets mean = 14,316). The mean is statistically not different (p > .05), therefore concluding that there is no significant difference in social media conversation before and after a celebrity campaign is released.

Screen Shot 2016-06-23 at 2.50.56 PM

Discussion: The purpose of this research was to see whether there was a difference in social media conversation regarding a particular retailer after they release a celebrity-endorsed campaign. The significant findings in this research are that the amount of social media conversation regarding the retailer does not have a significant influence, after releasing such campaign.

Analyzing the population curve, a burst of high conversation is noted retailer mentions, PUMA, however, such conversation trend slows down by the next day. This shows lack of continuous conversation regarding the retailer. Although, these results might not be generalizable as the research focuses on a particular company (PUMA) with a particular celebrity (Jenner).


Implications: This study concludes that the use of celebrity campaigns on social media platforms does not increase social media conversation or popularity regarding the retailer. The null hypothesis is accepted that there is no significant difference in either of these time periods. Such research can be further confirmed by studying other retailers who have chosen celebrities whose image associate well with the retailer’s’ image for campaign endorsements.

Leave a Reply