Sentiment Analysis of Energy Drinks on Twitter: A Marketing Case

This case study explores the sentiment surrounding energy drinks on Twitter, analyzing consumer perceptions of the product category and specific brands. The research employs data extraction tools and web scraping methods using the Twitter API to gather and classify tweets, focusing on sentiments related to safety, alternatives, and marketing messages. The analysis reveals predominantly positive sentiments towards energy drinks as a safer alternative to other options, while also highlighting concerns about accessibility and potential negative impacts on youth. The study concludes by suggesting that companies should refine their messaging to address negative connotations and emphasize the benefits of their products, tailoring their approach to align with their desired brand positioning. Desklib provides similar solved assignments and past papers for students.

Student Name
Course Name
University Name
24th-July-2018

Introduction
There are nearly 4 to 5 companies in the space of energy drink who operates in the
region. Other than these there are few who also serve this product from informal sector. It
is believed that informal sector has nothing much to offer and it is the formal sector with
major 4 to 5 brands that are running the show in space of energy drink. In the current
promotion mix, social media is getting a whole new traction. Companies are spending huge
money on this platform as it provides means to reach out to the target segment in the
fastest manner (King, Smith, McNamara, Matthews & Fridberg, 2014). However, there are
certain challenges which come along with this. One of the major challenges is the
cacophony associated with drink; hence any promotion by any means is not desired.
Objectives
The objective of this research is to do adequate research and see what social media
has to say about energy drink. The product itself is emerging energy drink content product
but it does not generate any kind of toxic combustion which the other energy drink content
product generates. This is the prime reason why it is considered friendlier even with the
negative connotations attached. There are no of companies which are actively promoting,
connecting with target segments in twitter
(Bullen, Howe, Laugesen, McRobbie, Parag, Williman & Walker, 2013)
Literature review
Twitter platform is a famous social networking platform and is now a very important
platform for product placement, promotion and other such activities. The platform initially
started with character limitations of 140 and has now removed it

(Vickerman, Carpenter, Altman, Nash & Zbikowski, 2013). The platform has grown by leap
and bounds over the year and has done some amazing connectivity across the world. A
single tweet is a textual expression which and user can write or post in the twitter platform.
Somebody who likes the idea being shared can re-tweet the same a tweet from his/her
profile. There are also ways of striking a textual instant conversation and since the platform
is so popular, it has grown by leap and bounds (Farsalinos, Polosa, 2014).
The world of twitter is amazing and many companies have utilised the platform for
placement of the product, creation of brand identity, communicating features of the
product and above all promotions. However since twitter is an open platform, there are
huge risk involved as the message might be wrongly consumed and can backfire. This is
where one needs to be very content specific while marketing on social media platforms like
twitter. The damage control process is not very good in this kind of platform (McCarthy,
2014)
Methods
The method of this research will involve deep diving into various tweets of the
subject and see what kind of sentiments are rising about the product. This will help us in
doing proper analysis and segmentation which in turn will result in object oriented
marketing. We will use data extractor tools along with web scrapping methods using API
provided by twitter to get such tweets. Our analysis will be limited to analysing these tweets
and see what the belief and outcome is.

The twitter streaming API will help us in getting the tweet content, along with proper
classification under the heads as described in the above picture. We will have the twitter
name of the user, the feature extractor which is combination of MALLET and features. On
this we will run the classification model to get the desired result from the research
(Tremblay, Pluye, Gore, Granikov, Filion & Eisenberg, 2015)
Results and Discussion
We tested the data that we fetched and using weighted model averaging approach,
keeping alpha as 0.85, we initially trained the model so that they can recognise the final
outcome. After training the model and back testing it, we ran the same on the data which
we have kept for analysis. The sampling method applied here was stratified sampling and
the same was put to test.

If we see the results, it is visible that the model used by us generated a very high
precision model and strong recall values. This shows that user mentioning about the same
has generated strong recall values and hence the conversation has been centred. The focus
of second part of analysis was to see what the content actually threw up (Grana, Popova &
LingA, 2014). The conversant highlighted the usage of energy drink and how it is considered
to be safer compared to the other mode like alcoholic drinks.
Conclusion
Our analysis suggests that sentiments are more or less positive in this space. People
realise that although the product is bad for consumption but this approach towards the
product is comparatively safer. The tweets also highlighted that instead of energy drink
content one can also look at flavoured energy drink which are very safe and can help in
curbing consumption of energy drink content to the larger extent. The smoke free aspect
and the fact that the product line is new also came out of the analysis. There were some
negative connotations also linked with it, the easy online availability reduces all kind of
censorship on the product. Hence this is one reason where youth and kids can get exposed
to this form very early and usage won’t be easily traceable. Few tweets also highlighted

outcome of surveys which shows that people want to know why people want to consume it.
Some people want a safer approach, and has recommended usage of it. Companies are
spending huge money on this platform as it provides means to reach out to the target
segment in the fastest manner. However, there are certain challenges which come along
with this. Some have also talked about giving up; quitting and the kick factor. Some have
suggested that this is not the real thing. The competition is with juice category. As a product
company these connotations need to be corrected and a right strategic set of promotion can
go a long way in building the brand. The message conveyed should be a combination of
benefits and also the fact that this is the still the real thing. The message can also differ as
part of the brand positioning the organisation want to achieve.

References:
Aex.Cubana. King, Lorry.Jack. Smith, Paresh.Jack. McNamara, Aanthony.K.ing Matthews, Dre
krod.James. Fridberg, 2014, Passive exposure to electronic drink (e-drink) use increases
desire for combustible and energy drink in young adult smoker, Energy drink content
Control.
C. Bullen, C. Howe, M. Laugesen, H. McRobbie, V. Parag, J. Williman, N. Walker, 2013,
Electronic drinks for drink cessation: a randomised controlled trial, The Lancet, 382 (9905),
pp. 1629-1637.
K.A. Vickerman, K.M. Carpenter, T. Altman, C.M. Nash, S.M. Zbikowski, 2013, Use of
electronic drinks among state energy drink content cessation quitline callers, Nicotine
Energy drink content Res., 15 (10), pp. 1787-1791.
K.E. Farsalinos, R. Polosa, 2014, Safety evaluation and risk assessment of electronic drinks as
energy drink content drink substitutes: a systematic review, Ther. Adv. Drug Saf., 5 (2),
pp. 67-86.
M. McCarthy, 2014, Youth exposure to drink advertising on US television soars, BMJ: Brit.
Med. J., 348.
M.C. Tremblay, P. Pluye, G. Gore, V. Granikov, K.B. Filion, M.J. Eisenberg, 2015, Regulation
profiles of energy drink in the united states: a critical review with qualitative synthesis, BMC
Med., 13 (1), p. 130.
R. Grana, L. Popova, P. Linga, 2014, longitudinal analysis of electronic drink use and drink
cessation, JAMA Internal Med., 174 (5), pp. 812-813.