BUSI 614 Week 5 DB Forum 2 Example

Discussion Board Forum 2: Big Data and the Future of Marketing Research

BUSI 614 Marketing Research, Discussion Board Forum 2 (thread and 1 reply)

Thread by [Student Name]

The forum prompt and the classmate's thread are composites standing in for the course's own; the research cited is real, and the requirements follow Liberty's public course guide for BUSI 614.

Thread

Big data analytics will change marketing research, but it will not replace surveys, interviews and experiments, and companies that treat it as a replacement will make worse decisions. Behavioral data show what customers do, at a scale and speed no survey can match, but they cannot by themselves show why customers act as they do, what people who are not yet customers want or what would happen if a company changed its offer.

The strengths of analytics are real. Wedel and Kannan (2016) reviewed how marketing analytics has developed as firms gained access to transaction records, clickstreams, mobile and social media data, and they described methods that allow companies to personalize offers, allocate marketing budgets and measure the effects of marketing actions far more precisely than before. They also identified challenges that data-rich environments create, including protecting customer privacy, combining data from many sources and drawing causal conclusions from observational data. Those challenges are where traditional research remains essential. A randomized experiment, for example, can show whether a new price causes a change in purchases, while a pattern in transaction data may reflect seasonality, competitors' moves or the kinds of customers who happened to see the price.

Data alone do not produce insight. Erevelles et al. (2016), using a resource-based view of the firm, argued that big data becomes a source of advantage only when a company has the capabilities to turn it into knowledge and act on it, and that insight comes from combining patterns discovered in the data with ideas tested deliberately. In practice, analytics often finds a pattern, and research is needed to explain it.

Large data sets can also mislead. Lazer et al. (2014) examined Google Flu Trends, which estimated flu levels from search queries, and showed that it substantially overestimated flu prevalence in several seasons, at one point putting the share of physician visits for flu-like symptoms at more than twice the figure the federal government's surveillance system recorded. They attributed the failure partly to "big data hubris," the assumption that big data can substitute for traditional data collection and analysis, and partly to changes in Google's search algorithm that altered the data over time. More data reduces sampling error, but it does nothing about measuring the wrong thing.

Consider a streaming service. Its viewing data show that subscribers who watch fewer than three hours a week are likely to cancel, and analytics can predict who will leave. But the data cannot say why those subscribers watch so little: whether the catalog lacks what they want, the interface makes it hard to find or a competitor has a show they prefer. Depth interviews with low-viewing subscribers, a survey of former subscribers and an experiment testing personalized recommendations would answer those questions. The data also say nothing about people who have never subscribed, a population that only research can reach.

The future of marketing research is therefore a combination: analytics to identify patterns and target questions, and traditional methods to explain, test and reach beyond current customers. Researchers who can do both will be more valuable, not less, because they can move from a pattern in the data to an explanation a manager can act on.

Reply to Marcus

Marcus, your thread argued that companies that invest in analytics outperform those that rely on intuition, and that marketing research departments should shift resources from surveys to data science. I think the evidence supports the first part of your claim, though it also suggests conditions that matter for the second.

Germann et al. (2013) collected survey responses from senior executives at large firms about how extensively their marketing decisions relied on analytics and related the answers to firm performance. They found that deployment of marketing analytics was associated with better performance with a stronger link in industries with more intense competition and where customer preferences change rapidly. Deployment itself depended heavily on advocacy from top management and on a culture that valued analytics, since tools and data alone did not make managers use them.

These findings strengthen your argument in competitive, fast-changing markets, which describes much of retail, technology and media. They also suggest that the payoff depends on how analytics is used in decisions, not merely on acquiring it. A company that hires data scientists but keeps making decisions the old way will see little benefit, which echoes the first forum's discussion of why managers ignore research they have paid for.

I would question the conclusion that resources should move away from surveys, at least as a general rule for every company. The firms in the study were deploying analytics for decisions, and analytics in that sense can include survey data, experiments and models combining several sources. Survey research often supplies the measures, such as satisfaction, brand perceptions and purchase intentions, that behavioral data lack, and it can be analyzed with the same advanced methods. The more useful shift may be from research projects conducted in isolation to an integrated function that combines behavioral data with surveys and experiments.

Would you consider revising your recommendation to say that marketing research departments should build analytics skills and integrate data sources, rather than replace surveys? That version seems more consistent with what Germann and colleagues found, and it avoids the risk of losing the ability to learn why customers behave as they do. It would also protect the company in the situations analytics handles worst, such as launching a product category for which no behavioral data yet exist.

References

Erevelles, S., Fukawa, N., & Swayne, L. (2016). Big data consumer analytics and the transformation of marketing. Journal of Business Research, 69(2), 897-904. https://doi.org/10.1016/j.jbusres.2015.07.001

Germann, F., Lilien, G. L., & Rangaswamy, A. (2013). Performance implications of deploying marketing analytics. International Journal of Research in Marketing, 30(2), 114-128. https://doi.org/10.1016/j.ijresmar.2012.10.001

Lazer, D., Kennedy, R., King, G., & Vespignani, A. (2014). The parable of Google Flu: Traps in big data analysis. Science, 343(6176), 1203-1205. https://doi.org/10.1126/science.1248506

Wedel, M., & Kannan, P. K. (2016). Marketing analytics for data-rich environments. Journal of Marketing, 80(6), 97-121. https://doi.org/10.1509/jm.15.0413