Skillseed: A Marketing Analytics Approach
1 like341 views
This document summarizes a marketing analytics study conducted for the social enterprise Project Skillseed. The study used clustering, conjoint analysis, and logistic regression to identify two target customer segments - "Forced Participants" and "Social Aspirants". Key recommendations include focusing products for each segment, leveraging social networks and schools to reach "Forced Participants", and emphasizing autonomy and fun to appeal to "Social Aspirants". Limitations of the study are also noted.
Skillseed: A Marketing Analytics Approach
- 1. A Marketing Analytics Approach to todays Social Enterprises
- 3. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION COMPANY BACKGROUND Project Skillseed For-profit social enterprise Founded in March 2013 by HJ Phua Connecting students and educators to meaningful service learning experiences around the world. Unique Selling Proposition Top NGO partnerships Expert mentorships, effective application Caters to participants unique interests.
- 4. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION COMPETITORS Educational Institutions Travel Agencies Non- government organizations (NGOs) Competitors as potential partners
- 5. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION PROBLEM RECOGNITION Dont know their current position and competitors Wants to increase reach Need more information on customers and market To see if tertiary students have a demand for Project Skillseeds services. Research ObJective
- 6. Methodol ogyHow we did it
- 7. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Preliminary Review Potential Market Size Focus Group Understanding Overseas Expeditions
- 8. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Preliminary Review Location Price Size FriendMix Degree of Control Project Length
- 9. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Primary Research Consumer Survey MR Techniques 136 Respondents, 119 valid Students in Singapore Awareness of community service Clustering Techniques Means Testing (2, 3 cluster) Cross Tabulation Customer Profiling (Logistic Regression) Conjoint Analysis Linear Regression
- 11. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Clustering Hierarchical Stage 1st Cluster Combined 2nd Cluster Combined Coefficients Coefficient Difference (From Previous Stage) 110 45 82 55.000 0.000 111 1 45 59.567 4.567 112 1 47 60.091 0.524 113 6 91 62.500 2.409 114 1 48 70.295 7.795 115 1 6 77.469 7.174 116 1 71 91.345 13.876
- 12. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION K-means Clustering k Ratio Within/Bet ween Cluster Variance 2 173.19 3 41.15 4 18.56 5 8.13 0.00 50.00 100.00 150.00 200.00 2 3 4 5 Ratio Within/Between Cluster Variance Ratio Within/Between Cluster Variance
- 13. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Snake Plot 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23 Q24 Q25 Q26 Q27 Q28 Q29 Cluster Center Means Snake Plot (3 Clusters) Cluster 1 Cluster 2 Cluster 3
- 14. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Snake Plot 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Cluster Cent
- 15. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION 2 Step CLusterIng
- 16. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION 2 Step CLusterIng Number of Clusters Akaike's Information Criterion (AIC) AIC Changea Ratio of AIC Changesb Ratio of Distance Measuresc 1 2453.317 2 2431.084 -22.233 1.000 1.342 3 2444.068 12.983 -.584 1.063 4 2463.130 19.062 -.857 1.264 5 2502.408 39.279 -1.767 1.262
- 17. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Clustering: Which Solution? Technique 2-cluster solution 3-cluster solution Hierarchical (Dendrogram) INCONCLUSIVE Hierarchical (Agglomeration Schedule) K means Snake Plot 2 Step
- 18. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION CrossTabs Cluster Membership( 1 or 2), Crossed with Intention to join Skillseed CSP Past Behavior Gender Education Marginally Significant Significant Significant Marginally Significant
- 19. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION CrossTabs Cluster Generalization Cluster 1 Cluster 2 Mostly Men High School Students Mostly Women JC/ Tertiary Students
- 20. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Logistic Regression Can Gender Predict Cluster Membership? Demographic Variables Cluster 2 Membership Significant model Only Gender was significant 319.9% odds if customer were female
- 21. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Binary logistic regression How do the variables predict our decision variable: To Join Skillseed or Not? Behavioral Intention Demographic Attitude
- 22. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Binary logistic regression Behavioral Intention Demographic Attitude Have you ever done CSP project in the past 5 years? Do you intend to do CSP in the next 3 years? Are social enterprises your top 3 places to approach and pay for such projects? Age, Gender, Education Level, Major, Income Level 29 Psychographic/Attitudinal variables
- 23. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Binary logistic regression Cluster 1 P-value: 0.018, Chi-Square 13.604 Joining Odds: decreased by 85.6% if never done a CSP before Joining Odds: increased by 69.8% if agreeing to join CSPs only when mandatory increases by 1 Probability of customer joining SkillSeed: 62.2%
- 24. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Binary logistic regression Cluster 2 P-value: 0.073, Chi-Square 8.546 Joining Odds: increased by 93.3% if preferring to take the lead than be guided increases by 1 Joining Odds: decreased by 45.7% if agreeing to join CSPs only when mandatory increases by 1 Marginally Significant: CSP is for fun and travelling with friends 61.6% Probability of customer joining SkillSeed: 76.38%
- 25. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Cluster Names Segment 1 Segment 2 Forced Participants Social Aspirants
- 26. Findings | Analyses Marketing Mix
- 27. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Conjoint Analysis Location SEA, North Asia, Africa, America Price 2000 -6000, SGD Size 18-25 pax FriendMix With Friends, Strangers, Both Degree of Control None, Partial, Full Project Length 2- 12 weeks 20 Profiles
- 28. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Regression : Forced Participants Significant Model, p = 0.012 Variables with Impact: Price Friend Mix 2000-4000 SGD: - 9.611 4000-6000 SGD: - 21.341 >6000 SGD : - 22.3 Strangers to Friends : -8.921 Prefers Friends to Strangers!
- 29. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Regression : Social Aspirants Significant Model, p = 0.001 Variables with Impact: Location Friend Mix North Asia over SEA- : -3.858 Prefers SouthEast Asia Strangers to Friends : -6.646 Prefers Friends to Strangers!
- 30. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Regression : Social Aspirants Significant Model, p = 0.001 Variables with Impact: Price 2000-4000 SGD: - 12.798 4000-6000 SGD: - 26.443 >6000 SGD : - 31.276 18 - 25 : -5.063 Prefers Smaller GroupsSize 2-4 weeks: + 5.154 4-12 weeks: + 7.345Length
- 31. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Utility: Forced Participants Maximum Possible Utility: 5.316 Skillseed can do better!
- 32. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Utility: Social Aspirants Maximum Possible Utility: 9.961 Skillseed can do better!
- 34. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Product Focus Forced Participants Social Aspirants <$2000 Showcase Activity for Friends South East Asia < $2000 <18 pax 4-12 weeks
- 35. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Marketing Channel Social Aspirants Source of Information as predictors to join: 1. Reliance on Social Network: 111.6% more likely 2. Reliance on NGOs for information: 71% less likely
- 36. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Marketing Channel Forced Participants Source of Information as predictors to join: 1. Reliance on Social Network: 381.7% more likely 2. Reliance on NGOs for information: 308.9% more likely
- 37. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Product Focus Forced Participants Social Aspirants Social Networks Reach out to Schools Social Networks Cooperate with NGOs
- 38. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Marketing Message Forced Participants Social Aspirants Not feasible to reach and acquire Address school interests Sell autonomy as a premium An outlet of fun and recreation
- 39. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Conclusion Both clusters must be addressed. Focus on: Cluster 1. Price Sensitivity School Partnerships Requires less modifications to current offerings. Social Media Presence.
- 40. INTRODUCTIO METHODOLOG FINDINGS RECOMMENDA LIMITATION Limitations of study Insufficient data from JC students Respondents limited to friends and acquaintances Selected IVs not exhaustive Attitudes are proven as weak indicators.