Quality and
effectiveness in
Big data & AI
Marketing
Andreina Mandelli, PhD
SDA Bocconi school of management Milan
EVERYTHING
STARTED HERE
QUALITY OF DATA:
FROM DATA TO
DECISIONS TO VALUE
AREAS OF MARKETING IMPACTED BY QUALITY OF BDAI
Customer side
• Info search & evaluation of alternatives
• Preference formation
• Buying
• Customer service
• Participation and community
• Smart products and personal informatics
Marketer side
• Market research/learning
• Targeting
• Branding
• Omnichannel experience journey & relationship management
• Lead generatiom & Sales
• Customer service operations
• Reputation & CSR
EFFICIENCY OF PROCESSES
QUALITY OF DATA
QUALITY OF ANALYTICS
QUALITY OF DECISIONS
QUALITY OF TRANSFORMATION
DATA STRATEGY
DATA WHAT TO DO WITH THIS DATA?
BUSINESS & MKTG GOALS SELECTION , ANALYSIS AND
USAGE OF THE RELEVANT DATA
DIFFERENT APPROACHES TO PRODUCTION OF KNOWLEDGE IN MARKETING
DEDUCTIVE
ABDUCTIVE
INDUCTIVE
DATA
MACHINE LEARNING
UNDERSTANDING THE
SPECIFICITIES AND DIFFERENCES
GENERALIZED FORECAST
CONTESTUALISED FORECAST
THEORY
BIG DATA
SMALL DATA
ARTIFICIAL INTELLIGENCE: INVISIBLE AGENT IN THE VALUE CREATION (OR
DESTRUCTION)
DECISIONS
SUGGESTED OR
AUTOMATED BY
ALGORITHMS
THE BLACK BOX
IMPACT ON
CUSTOMERS
EXPERIENCE AND
VALUE GENERATED
OR DESTROYED
EXPLICIT OBJECTIVES OF THE
SYSTEM
INPUT DATA
AVAILABLE ALGORITHMS
MACHINE LEARNING SYSTEM
TRAINING PROCESSES
REDUCTION OF STRESS BY CHOICE
INCREASE IN EFFECTIVENESS
REDUCTION OF THE VARIETY TO
WHICH CUSTOMERS ARE EXPOSED
REDUCTION OF TRANSPARENCY IN
DECISIONS
BIG DATA IN SERVICE-ORIENTED MARKETING
Source: De Luca et al., 2020
HUMANS VS MACHINES OR HUMANS + MACHINES?
ARTIFICIAL INTELLIGENCE
HUMAN INTELLIGENCE
EXPLICIT DATA
(EDITABLE)
OBJECTIVES
EXPLICIT DATA
& UNCONSCIOUS
(EDITABLE)
OBJECTIVES
(DYNAMIC)
INTELLIGENT AGENT’S
BLACK BOX
HUMAN AGENT’S
BLACK BOX
(EDITABLE)
COMPUTATIONAL
DECISION
COMPUTATIONAL
DECISION
CREATIVITY 6
IMAGINATION
LEADERS SUCCESSFULLY INTEGRATE HUMANS & AI
https://www.bcg.com/press/20october2020-study-finds-significant-financial-benefits-with-ai
SOLUTIONS TO QUALITY ISSUES IN DIFFERENT BDAI MKTG AREAS
❑Quality of data
➢ Data strategy linked to business goals
➢ Completeness, accuracy, consistency
➢ Diversity of data
➢ Data architecture competences
➢ Data integration
➢ Big data + small data
❑Quality of analytics
➢ Data science competences
➢ Data integration
➢ Quality of ML training
➢ Quality of ML output validation
➢ Diversity of teams and cross-check controls
➢ Big data + small data
➢ Inductive + abductive + deductive epistemologies
➢ Business translators
❑Quality of decisions
➢ Clear business goals
➢ Business translators
➢ Interfunctional and agile teams
➢ Customer-centric orientation
➢ Degree of interpretability (unboxing the black-box) by policy
➢ AI risk management (risk for company but also for customers)
➢ Hybrid AI + Human processes in marketing automation
➢ Participation of the customer to the process
➢ Use BDAI to support creativity
➢ BDAI ethics
➢ KPIs of customer satisfaction, tracking customer frustrations and painpoints
in the journey
❑Quality of transformation
➢ Clear vision and business transformation goals
➢ Beyond opitimization, strategic business transformation first
➢ Support from the top Management
➢ Customer-centric orientation
➢ Service-dominant logic in marketing (Every business is a service business)
➢ Omnichannel orientation
➢ Platform & eco-system approach to business transformation
➢ Data monetization
➢ Nurturing innovation culture and business imagination
andreina.mandelli@sdabocconi.it

From data quality to Big Data & AI (BDAI) marketing quality

  • 1.
    Quality and effectiveness in Bigdata & AI Marketing Andreina Mandelli, PhD SDA Bocconi school of management Milan
  • 3.
    EVERYTHING STARTED HERE QUALITY OFDATA: FROM DATA TO DECISIONS TO VALUE
  • 4.
    AREAS OF MARKETINGIMPACTED BY QUALITY OF BDAI Customer side • Info search & evaluation of alternatives • Preference formation • Buying • Customer service • Participation and community • Smart products and personal informatics Marketer side • Market research/learning • Targeting • Branding • Omnichannel experience journey & relationship management • Lead generatiom & Sales • Customer service operations • Reputation & CSR EFFICIENCY OF PROCESSES QUALITY OF DATA QUALITY OF ANALYTICS QUALITY OF DECISIONS QUALITY OF TRANSFORMATION
  • 5.
    DATA STRATEGY DATA WHATTO DO WITH THIS DATA? BUSINESS & MKTG GOALS SELECTION , ANALYSIS AND USAGE OF THE RELEVANT DATA
  • 6.
    DIFFERENT APPROACHES TOPRODUCTION OF KNOWLEDGE IN MARKETING DEDUCTIVE ABDUCTIVE INDUCTIVE DATA MACHINE LEARNING UNDERSTANDING THE SPECIFICITIES AND DIFFERENCES GENERALIZED FORECAST CONTESTUALISED FORECAST THEORY BIG DATA SMALL DATA
  • 7.
    ARTIFICIAL INTELLIGENCE: INVISIBLEAGENT IN THE VALUE CREATION (OR DESTRUCTION) DECISIONS SUGGESTED OR AUTOMATED BY ALGORITHMS THE BLACK BOX IMPACT ON CUSTOMERS EXPERIENCE AND VALUE GENERATED OR DESTROYED EXPLICIT OBJECTIVES OF THE SYSTEM INPUT DATA AVAILABLE ALGORITHMS MACHINE LEARNING SYSTEM TRAINING PROCESSES REDUCTION OF STRESS BY CHOICE INCREASE IN EFFECTIVENESS REDUCTION OF THE VARIETY TO WHICH CUSTOMERS ARE EXPOSED REDUCTION OF TRANSPARENCY IN DECISIONS
  • 8.
    BIG DATA INSERVICE-ORIENTED MARKETING Source: De Luca et al., 2020
  • 9.
    HUMANS VS MACHINESOR HUMANS + MACHINES? ARTIFICIAL INTELLIGENCE HUMAN INTELLIGENCE EXPLICIT DATA (EDITABLE) OBJECTIVES EXPLICIT DATA & UNCONSCIOUS (EDITABLE) OBJECTIVES (DYNAMIC) INTELLIGENT AGENT’S BLACK BOX HUMAN AGENT’S BLACK BOX (EDITABLE) COMPUTATIONAL DECISION COMPUTATIONAL DECISION CREATIVITY 6 IMAGINATION
  • 10.
    LEADERS SUCCESSFULLY INTEGRATEHUMANS & AI https://www.bcg.com/press/20october2020-study-finds-significant-financial-benefits-with-ai
  • 11.
    SOLUTIONS TO QUALITYISSUES IN DIFFERENT BDAI MKTG AREAS ❑Quality of data ➢ Data strategy linked to business goals ➢ Completeness, accuracy, consistency ➢ Diversity of data ➢ Data architecture competences ➢ Data integration ➢ Big data + small data ❑Quality of analytics ➢ Data science competences ➢ Data integration ➢ Quality of ML training ➢ Quality of ML output validation ➢ Diversity of teams and cross-check controls ➢ Big data + small data ➢ Inductive + abductive + deductive epistemologies ➢ Business translators ❑Quality of decisions ➢ Clear business goals ➢ Business translators ➢ Interfunctional and agile teams ➢ Customer-centric orientation ➢ Degree of interpretability (unboxing the black-box) by policy ➢ AI risk management (risk for company but also for customers) ➢ Hybrid AI + Human processes in marketing automation ➢ Participation of the customer to the process ➢ Use BDAI to support creativity ➢ BDAI ethics ➢ KPIs of customer satisfaction, tracking customer frustrations and painpoints in the journey ❑Quality of transformation ➢ Clear vision and business transformation goals ➢ Beyond opitimization, strategic business transformation first ➢ Support from the top Management ➢ Customer-centric orientation ➢ Service-dominant logic in marketing (Every business is a service business) ➢ Omnichannel orientation ➢ Platform & eco-system approach to business transformation ➢ Data monetization ➢ Nurturing innovation culture and business imagination
  • 12.