The evolution of marketing Analytics
The historical evolution of marketing data and analytics has progressed through three distinct stages: the description of observable market conditions using simple statistical approaches; the development of diagnostic models rooted in economic and psychological theories; and the evaluation of marketing policies using statistical, econometric, and operations research (OR) methods to support decision-making. This evolution has been fundamentally driven by technological milestones that expanded the volume, variety, and velocity of available data.
1. The Universal Product Code (UPC) and POS Scanning
The introduction of the Universal Product Code (UPC) and IBM’s point-of-sales scanning devices in 1972 marked the first automated capture of data by retailers. This technology allowed retailers to replace manual, bimonthly store audits with granular, real-time scanner data. The subsequent use of loyalty cards allowed firms to link these purchases to individual customers, creating "scanner panel data". The availability of this granular data catalysed the development of market share and demand models. It specifically led to the creation of scanner panel-based multinomial logit models, which used econometric methods to capture hierarchical consumer decision-making.
2. The World Wide Web and Clickstream Data
The arrival of the World Wide Web in 1995 and the subsequent founding of Google (1998) and Facebook (2004) transformed data from periodic snapshots into massive, continuous streams. Marketers began capturing clickstream data from server logs to track page views and clicks using cookies. The rise of user-generated content (reviews, blogs, and video) and social networks introduced a high variety of unstructured data. This era saw the rise of search marketing and attribution modelling, which assigned credit to different online touchpoints for customer conversions. It also propelled the use of Bayesian approaches to integrate models of varied consumer behaviours, such as web browsing and social network interactions. Furthermore, the complexity of this data led practitioners to adopt machine learning methods (like deep neural networks) from computer science, which offered high predictive power even without traditional causal theory.
3. Smartphones and GPS Capabilities
While mobile devices existed earlier, the introduction of the Apple iPhone in 2007 served as a catalyst for the large-scale capture of location-based information. GPS capabilities allowed for the unprecedented scale of consumer location data collection. This added a situational context to the data, allowing firms to see not just what was bought, but exactly where and when the consumer was at the time of an offer. This spurred the development of geo-fencing analytics. Methods were developed to test the efficacy of targeting customers based on crowdedness (e.g., in subways) and proximity to a firm’s or a competitor's location.
4. The Internet of Things (IoT)
The current evolution is moving toward the Internet of Things (IoT), where billions of devices with sensors connect and transfer data without human interaction. This generates massive, real-time data from household and industrial objects, potentially becoming a primary source for new product development. This is expected to usher in automated attention analysis and highly advanced adaptive personalization, where artificial intelligence (AI) and cognitive systems continuously learn and adapt offerings to changing user preferences in an automated, closed-loop cycle.
Marketing analytics chain of effects
Marketing analytics chain of effects explore how internal firm characteristics lead to the deployment of analytics and how that deployment subsequently translates into superior financial performance. Four critical internal drivers that must be in place for marketing analytics to be deployed effectively:
- Top Management Team (TMT) Advocacy: This is the primary driver. TMT members must champion analytics to ensure they become an integral part of business routines.
- Supportive Analytics Culture: Advocacy from the top nurtures a culture of evidence-based decision-making. Because culture is "sticky" and difficult for rivals to imitate, it can turn analytics into a sustainable competitive advantage.
- Requisite Analytics Skills: The firm must have access to personnel with both technical knowledge (models/concepts) and tacit knowledge (real-world application).
- Data and IT Infrastructure: A firm requires sophisticated physical IT resources to obtain, store, and distribute the vast amounts of data needed for analysis.
The external environment influences the "payoff" a firm receives from its analytics investments:
- Industry Competition: The positive impact of analytics on performance is significantly greater in highly competitive industries. In these environments, firms must achieve higher levels of customer satisfaction to survive, making the insights from analytics more valuable.
- Changing Customer Preferences: Analytics provides an "early warning" system in markets where customer needs (e.g., features, price points) change rapidly. The more customers fluctuate, the more critical environmental scanning becomes.
Marketing analytics applications
Four critical domains for marketing analytics applications: Customer Relationship Management (CRM),the Marketing Mix, Personalisation, and Privacy and Security. Each domain has been fundamentally transformed by the advent of big data offering unique action possibilities known as "affordances" while presenting new organisational and technical hurdles.
1. Customer Relationship Management (CRM)
- Opportunities: The era of big data allows firms to move beyond traditional "recency, frequency, monetary" (RFM) metrics to a deeper understanding of Customer Lifetime Value (CLV). Firms can now integrate multiple data sources, such as social media and web crawling, to identify high-quality leads and predict "churn hazard" with greater precision.
- Challenges: A key challenge is the "silo effect," where different teams manage their own budgets and data in isolation, preventing a global view of customer effectiveness.
2. The Marketing Mix
- Opportunities: Big data provides the breadth needed to disentangle firm-controlled drivers (like price) from external factors (like competition). It enables attribution modelling, which assigns credit to various online and offline touchpoints along the entire "path to purchase".
- Challenges: Marketers face the issue of endogeneity, where predictions are biased because management actions are often dependent on prior sales outcomes. Furthermore, firms must navigate the complexity of cross-media effects, such as how television ads trigger online branded searches.
3. Personalisation
- Opportunities: Firms can transition from mass marketing to individual-level personalisation, adapting offerings to specific tastes in real-time. Big data enables "adaptive personalisation," which automates the entire feedback loop of learning, adapting, and evaluating.
- Challenges: There is a risk of "analysis paralysis" if firms cannot process high-velocity data fast enough to act. Additionally, over-personalisation can lead to consumer backlash if it is perceived as too intrusive.
4. Privacy and Security
- Opportunities: Respecting privacy is increasingly viewed as a competitive advantage; giving customers more control over their data has been shown to double click-through rates on personalised ads.
- Challenges: The "mosaic effect" allows firms to reveal private information by fusing multiple non-private datasets. Firms also face the constant threat of security breaches.
Marketing analytics and dashboard visualisation
The quality of information presented on a dashboard does not simply improve decisions directly; rather, it functions through a sophisticated chain of cognitive and psychological effects involving information satisfaction and perceived task complexity.
1. Dimensions of Information Quality (IQ)
- Format: The clarity and understandability of how information is visually presented (e.g., color palettes, chart types).
- Currency: The degree to which the information is up-to-date and reflects the current state of the business environment.
- Completeness: The availability of all necessary data points with sufficient depth and width to address the specific decision task.
- Accuracy: The perceived correctness and believability of the data.
2. The Role of Information Satisfaction
Information satisfaction serves as the first mediating mechanism. It represents the manager’s attitude or feelings regarding the adequacy of the information provided for a specific task. Format, currency, and completeness all significantly and positively contribute to information satisfaction,. When a dashboard is well-formatted and provides a complete, timely picture, managers feel satisfied with their informational resource. Interestingly, accuracy was found to have a non-significant effect on satisfaction in previous experimental setting. This is attributed to the difficulty of discerning absolute accuracy in fictitious or unfamiliar scenarios without external data to cross-validate.
3. The Mechanism of Perceived Task Complexity
Perceived task complexity is the degree of mental effort or cognitive load required to solve a problem. It serves as a vital bridge between information intake and the final decision. High-quality format, currency, and completeness directly reduce a manager's perception of how "hard" a task is. Proper formatting, for example, allows for faster pattern recognition, which eases the cognitive burden. Information satisfaction also has a strong inverse effect on complexity; when managers are satisfied with their information, they perceive the decision task as less complex.
4. Impact on Decision-Making Quality
The ultimate goal, decision-making quality (defined as the accuracy, correctness, and perceived confidence in a choice), is influenced by these internal states in the following ways:
- Information satisfaction was found to be the most significant driver of decision quality. Satisfied managers are more committed and confident, which leads to superior outcomes.
- Perceived task complexity has a negative impact on decision quality. High cognitive load can overwhelm working memory, leading to "analysis paralysis" or suboptimal shortcuts in reasoning.
References
- Germann, F., Lilien, G.L. and Rangaswamy, A., 2013. Performance implications of deploying marketing analytics. International Journal of Research in Marketing, 30(2), pp.114-128.
- Hjelle, S., Mikalef, P., Altwaijry, N. and Parida, V., 2024. Organizational decision making and analytics: An experimental study on dashboard visualizations. Information & Management, 61(6), p.104011.
- Wedel, M. and Kannan, P.K.,2016. Marketing analytics for data-rich environments. Journal of marketing,80(6), pp.97-121
