Meow, Voyager: Descriptive and Correlational Analysis for the Cat Café Business Model
Updated: Mar 31
This was written for CAS 829: Evaluation Techniques in the Michigan State University Strategic Communications MA program.
Entrepreneurs/cat lovers who seek a purr-fect business opportunity may find lucrative opportunities in the “cat café” (CC) market. Originating in Taiwan circa 1998, the CC business model places cat lovers in a cozy, feline-filled environment while enjoying coffee, tea, and sweets (Smith, 2018; Kelly, 2018). CC owners often work with local animal shelters to promote cat adoption, and CC customers can spend time with cats they otherwise cannot adopt (Warren, 2015; Crimaldi, 2017). The idea may elicit warm-and-fuzzy images of cuteness, but the cat café business model is not without challenges. Health codes, zoning, regulations, safety, and sanitation are constant issues that may scratch at profits (Cagan, 2018; Smith, 2018; Warren, 2015). From a marketing perspective, strategic partnerships—bolstering both the restaurant and the animal aspects—are crucial for success. To that end, the purpose of this paper is to analyze and to report the findings of a CC-based consumer study. Areas of study include likelihood to patronize and prospective business names. In a secondary objective, this paper will present strategic recommendations based on insights from this study.
The participants included 78 adults 29 men (M = 5.72, SD = 2.41) and 49 women (M = 25.16, SD = 3.82). These adults completed the survey voluntarily and were not compensated for their time.
How to Interpret Results
Descriptive statistics help researchers study simple summaries and basic features of a study, and correlational research can help predict future events (Trochim, 2020; Graziano & Raulin, 2012). Often, researchers utilize populations or groups as samples of a subset (Graziano & Raulin, 2012). When asking questions of sample populations, one ponders if the relationship supports or refutes the “null hypothesis (NH)”—a statistic used to determine if variables are close enough to infer a significant relationship (Graziano & Raulin, 2012, p. 114). Think of the NH like conjecture or the possibility that a theory is based on chance (Hayes, 2020). Rejecting a NH indicates a significant relationship between variables (Hayes, 2020). When determining the NH, researchers use the p-value to support or to refute the relationship; in short, a p-value is used to measure the level of significance at which one would reject the NH (McLeod, 2019; Beers, 2020).
The alpha value (AV) serves as a regulator for the p-value: an AV offers a threshold (defined by a level of extremity) against which one measures the p-value—ultimately providing a level of confidence (Taylor, 2019). When the p-value is less than the AV, the NH is rejected. When the p-value is greater than the AV, the NH is not rejected (Taylor, 2019). The most common AV is 0.05 (Beers, 2020).
The r-value, or correlation coefficient, measures the relationship alignment between a data set (Graziano & Raulin, 2012; Taylor, 2020). This r-value is measured between +1 and -1; the closer to positive one, the better the data alignment (Crossman, 2019; Taylor, 2020).
Test one: (Appendix A) Hypothesis one states that females are more likely to frequent a CC than males. The results support this hypothesis. In a study rating likelihood based on gender, women are more likely to become a CC customer. The results of a t-test showed that women’s propensity to patronize a CC (M = 6.86, SD = 2.47) was larger than men’s likelihood (M = 5.72, SD = 2.41), t = 1.98, p < .05.
Test two: (Appendix B) In another likelihood study—focusing on age—results show a negative correlational relationship between age and likelihood to patronize a CC, r = -.24, p = < .05.
Test three: (Appendix C) Hypothesis two states that females will respond more favorably to the name “Furry Frappuccino” (Name 1). The results do not support this hypothesis. Researchers measured reaction to Name 1 based on a scale of 1 (hate it) to 10 (love it). This question was administered separately to males (M = 6.59, SD = 2.58) and females (M = 5.3, SD = 2.88), t =1.97, p = .05.
Test four: (Appendix D) Hypothesis three states that males will respond more favorably to the name “Don’t Stress Meow-t” (Name 2). The results do not support this hypothesis. Participants rated their response on a scale of 1 (hate it) to 10 (love it). The results were close between males (M = 6.51, SD = 2.65) and females (M = 6.77, SD = 2.5), t =0.42, p =0.67. However, the higher p-value negates this hypothesis.
Test five: Researchers tested the correlational relationship between Name 1 and age (r = -0.11, p = 0.34) and Name 2 and age (r = -0.30, p = <.05).
Test six: (Appendix F) The final study rated independent variables against each other. Name 1 versus Name 2 scored an r-value of .33. Name 1 versus age score an r-value of -0.11. Name 2 versus age scored an r-value of -0.20.
The overall results of this study indicate the target market for this burgeoning CC are young females. Starting with likelihood to patronize a CC, the research shows that females are nearly two times more likely than males to become a CC customer (with scores of 49-29 respectively). The large t-value (1.98) indicates a large difference between groups, and the small p-value supports (<.05) a strong relationship. This paper concludes that the CC target audience is predominantly female.
Regarding age, the negative r-value indicates that as one increases in value, the other decreases; thus, the variables move in opposite directions on a scatter plot (Investopedia, 2019; Graziano & Raulin, 2012). In this study, -0.24 indicates a weak negative correlation. From this result, one can posit that as age increases, likelihood of patronage decreases. Similarly, as age decreases, likelihood of patronage increases. From this insight, one can state that the target market skews young.
Recommendations include marketing to tween, teen, and college females who are “cat people.” Marketing and social media campaigns should target these segments in dynamic, interactive ways that engage these age groups appropriately. Ideas include:
• Hunt statements: Develop hunt statements and distill insights from segments to develop compelling campaigns. Leadership should “begin backwards” with global statements and core objectives based on this now-established target market of tweens, teens, and college females (Legorburu & McColl, 2014, loc.1862).
• Marketing personas: Create a marketing persona for a tween, a teen, and a college female that indicates preferences, brand affiliations, social media platform preferences, music/movie/pop culture archetypes, and disposable income.
• A “cat diary”: Tween girls can post the cute, fuzzy, and cuddly moments they have with their pets. This could be an app (with potential for in-app advertising) or a hashtag (#<businessname>CatDiary) that is easily shared across all platforms.
• Collateral: Digital and print collateral should feature tween, teen, and college females hanging out in the CC wearing branded shirts or backpacks with the company’s logo. The cats could wear branded collars or cat clothes, and subtle branding could be staged strategically in the background. For patrons who have cats at home, the CC could create and market a line of “cat fashion.” The latter could be a mix-and-match collection that solicits multiple retail opportunities—where girls seek to complete a “fashion collection” that is considered both a status and a retail accomplishment.
• “Twinning”: Invite cool cats of the target market to post pictures side-by-side with the CC “residents” and post them with a branded hashtag. Ideally, the “like,” shares, retweets, and comments would amplify the brand’s reach. With a larger number of females in our study results, female customers also could be encouraged to “twin” with their male friends, boyfriends, co-workers, etc. to encourage a male patronage.
• Guest blog: The CC could create a blog on the company website with short- and long-form testimonials, similar to a Yelp review, and invite customers to talk about their visit. This technique ideally would create enthusiasm by using social proof and liking as “weapons of influence” (Cialdini, 2014, p.116). Social media actors are more likely to follow-up on brand loyalty themselves once they have written statements of support—through the self-imposed constructs of commitment and of consistency (Cialdini, 2014).
• Social media influencers: These endorsements convey authority among peer groups: when one is comfortable with someone, they tend to say yes and are more likely to comply with requests (Cialdini, 2014; Ramsland, 2016; Jacobs, 2017). Thus, we can utilize influencers as “message shapers” who create a buzz and who endorse the CC model and mission (McAlister, 2016a, p. 5). Ideally, the CC’s target market would become co-creators of the CC’s persuasive marketing messages (McAlister, 2016a).
• Tri-component model (TCM): The TCM is a persuasion model that combines cognition (what we believe we have learned about a topic), conation (the behaviors we display towards a topic), and “affect” (the way we feel about a topic) (McAlister, 2016b, p. 5). This model could bolster CC marketing in two ways: it can play into and amplify the natural love for cats in animal lovers and it could be used to change negative attitudes (cat sickness, dander allergies, aggressive behaviors, misperceptions, and abuse). Cats are known to have attitude, but in marketing, consumer attitudes are a learned predisposition with consistency—patterns that can persuade current and would-be consumers.
• Artistic proofs: These persuasive concepts include ethos (image/appearance/reputation), pathos (emotions/word choice), and logos (logic/ intelligence) (McAlister, 2016b). They can factor into all messaging and branding as well. Leadership should blend these tools into a powerful ball of marketing twine that entices the customer base through personality, advertising, and evidence-based reasoning (McAlister, 2016c). Cat pictures rule the Internet, and the artistic proofs can tap into this evergreen trend with viral images, and heartwarming stories. Cross-promotion blends reputation, and logic perfectly; thus, leadership could cross-promote with adoption agencies and non-profits. Finally, leadership should explore ways to host animal therapy events for patients, survivors, and family members afflicted with diseases that have efficacy in patient recovery and well-being (cancer, Alzheimer’s, dementia).
Both Name 1 and Name 2 scored higher with men than with women. However, the p-values for each test did not suggest a significant gender relationship. Defaulting to the predominately female base, the female mean scored higher (6.77 versus 5.30) for Name 2. Among age, Name 1 scored a higher r-value, but this relationship was negated because the NH was accepted. The NH for Name 2 was rejected, indicating a strong relationship (r = -0.29). This plays into the premise that the CC target market skews young and female: as age decreases, the support for Name 2 increases. Thus, the recommendation is Name 2: “Don’t Stress Meow-t—as the CC’s name.
Consumer insights, like scientific research, stems from a curiosity to understand something (Legorburu & McColl, 2014; Graziano & Raulin, 2012). A study of consumer behaviors and interactions with services, transactions, customer support, and media content leads to relevant, emotional connections (McColl, 2015). Leadership can transform disciplined curiosity into a positive brand purpose and experience that connects on an emotional, strategic level (Graziano & Raulin, 2012; Legorburu & McColl, 2014). To purr-loin from Walt Whitman: Meow, voyager, sail thou forth to seek (cat lovers) and to find (a sustainable business model).
Beers, B. (2020, February 19). P-value definition. Investopedia. https://www.investopedia.com/terms/p/p-value.asp
Cagan, S. (2019, April 22). Navigate the combo of pets and human refreshments. Pets Plus Magazine. https://petsplusmag.com/start-a-cat-cafe-four-things-to-know/
Cialdini, R.B. (2006). Influence: The Psychology of Persuasion. Harper Business.
Crimaldi, L. (2017, December). Cat café faces challenges: The owner of Boston’s first feline-themed coffee shop has run into some obstacles. Gale OneFile. https://go-gale-com-proxy1-cl-msu-edu.proxy2.cl.msu.edu/ps/i.do?p=STND&u=msu_main&id=GALE%7CA425455029&v=2.1&it=r&sid=summon
Crossman, A. (2019, July 15). Correlation analysis in Research. ThoughtCo. https://www.thoughtco.com/what-is-correlation-analysis-3026696
Graziano, A.M. & Raulin, M.L. (2012). Research Methods—A Process of Inquiry (8th ed.) Pearson.
Hayes, A. (2020, February 19). Null hypothesis. Investopedia. https://www.investopedia.com/terms/n/null_hypothesis.asp
Investopedia. (2019, April 22). What does a negative correlation coefficient mean?
Jacobs, C.S. (2017, August 22). The art of persuasion. Psychology Today. https://www.psychologytoday.com/us/blog/management-rewired/201708/the-art-persusasion
Kelly, D. (2018, April 19). The dirty truth about cat cafes. Mashed. https://www.mashed.com/118733/dirty-truth-cat-cafes/
Legorburu, G., & McColl, D. (2014). Storyscaping: Stop Creating Ads, Start Creating Worlds. Wiley & Sons.
McAlister, A.R. (2016a). Chapter 1-1—Persuasion [PowerPoint]. Retrieved from Michigan State University D2L site.
McAlister, A.R. (2016b). Attitude formation and change. [PowerPoint]. Retrieved from Michigan State University D2L site.
McAlister, A.R. (2016c). Chapter 3—Persuasion from Aristotle to today. [PowerPoint]. Retrieved from Michigan State University D2L site.
McCleod, S. (2019). What a p-value tells you about statistical significance. Simple Psychology. https://www.simplypsychology.org/p-value.html
McColl, D. (2015, May 26). Insight on insights. Storyscaping. https://www.storyscaping.com/insight-on-insights/
Ramsland, K. (2016, September 16). 6 principles that predators will use against you.
Psychology Today. https://www.psychologytoday
Smith, L. (2019, March 9). Frisky business: Cat cafes pounce on new markets. The Politic. https://thepolitic.org/frisky-business-cat-cafes-pounce-on-new-markets/
Taylor, C. (2019, April 30). What is the difference between alpha and p-values? ThoughtCo. https://www.thoughtco.com/the-difference-between-alpha-and-p-values-3126420
Taylor C. (2020, January 28). Calculating the correlation coefficient. ThoughtCo. https://www.thoughtco.com/how-to-calculate-the-correlation-coefficient-3126228
Trochim, W. (2020, January 22). Descriptive statistics. Social Research Methods. https://socialresearchmethods.net/kb/descriptive-statistics/
Warren, M. (2015, August 15). Cat café craze clawing its way into Guelph: New café hopes cool cats will head downtown. Gale OneFile. https://go-gale-com-proxy1-cl-msu-edu.proxy2.cl.msu.edu/ps/i.do?p=STND&u=msu_main&id=GALE%7CA425455029&v=2.1&it=r&sid=summon