New frontiers in forecasting, predicting, and explaining

Journal of Marketing Analytics (2023) 11:559–560
https://doi.org/10.1057/s41270-023-00248-0
EDITORIAL
New frontiers in forecasting, predicting, and explaining:

an introduction to the special issue
Michael A. Levin1,2 · John T. Gironda3
Revised: 11 August 2023 / Accepted: 12 August 2023 / Published online: 15 September 2023
© The Author(s), under exclusive licence to Springer Nature Limited 2023
This Journal of Marketing Analytics special issue on “New that an effective application of quantitative techniques would
Frontiers in Forecasting, Predicting, and Explaining” seeks depend on improved qualitative research. Model develop-
to increase our understanding of how current analytical tech- ment and testing should strongly bridge the gap between
niques can be improved and new analytical techniques can academic inquiry and managerial application to ensure that
be applied. Advancement in these areas is needed because the marketing analytical toolkit remains relevant (Petrescu
while data collection continues to proliferate in the current and Krishen 2019a). Moreover, leading marketing analytics
century, the marketing analytical toolkit associated with scholars have been encouraging more theory-based criteria
forecasting and predicting continues to be rooted in previ- for managers concerning marketing analytics use and inter-
ous centuries. Software titles continued to add functionality pretation (Iacobucci et al. 2019).
for practitioners and academicians alike. These tools can Given all of this, we are pleased to introduce this Journal
help novice and experienced researchers tackle large datasets of Marketing Analytics special issue devoted to description,
with more accessible functionality associated with organ- prediction, and prescription, which has generated eight excit-
izing, analyzing, and visualizing data. However, easier does ing articles that provide researchers and practitioners with
not mean better, and we need more significant insights into novel uses of analysis.
the phenomenon of predictive analytics (Brynjolfsson et al. Two papers explore issues with surveys including issues
2021). with inattentive respondents and survey length. Fullerton
Larger datasets allow researchers to develop models that and McCullough examine results from a national survey to
provide more explanatory power and test models for better compare results from attentive respondents to inattentive
predictiveness (Sarstedt and Danks 2021). Those models respondents. The comparisons yield statistically significant
contain biases or can reinforce biasedness in the absence of differences that suggest inattentive respondents yield not so
ethical or legal considerations (Petrescu and Krishen 2020). much noise but measurement error if included in the analy-
Analytical techniques should also extend into working sis. Riggs et al. tackle the issue of lengthy surveys that lead
with data from structured and unstructured sources, which to diminished response rates and inattentive respondents.
are estimated to account for about 80% of available data The authors introduce a method—randomized, multicenter,
(Petrescu and Krishen 2019b). Hair (2007) estimated that parallel-arm research design—as a way of collecting data
predictive analytics would use more mixed-data models that from long surveys.
examine both structured and unstructured data and estimated Several papers introduce novel approaches to prediction
in differing contexts including business to business, retail-
ing, among individual sellers, and the launching of websites.
* John T. Gironda
Richter et al. develop a decision support system on a small
[email protected]
dataset to forecast sales of new products in a business-to-
Michael A. Levin
business context. Using a sample of 101, the authors apply a
[email protected]
six-step process that incorporates the CRISP-DM procedure
1
Lubbock, USA and extend this process with a self-learning component.
2
B.I. Moody III College of Business Administration, Wu et al. present a model that predicts sales at retail. In
University of Louisiana at Lafayette, Lafayette, LA, USA their effort, Google Trends search data are incorporated into
3
Department of Marketing, Huizenga College of Business a prediction model along with geodemographic and indus-
& Entrepreneurship, Nova Southeastern University, try-level variables. The model is trained on almost four years
Fort Lauderdale, FL, USA
Vol.:(0123456789)
560 M. A. Levin, J. T. Gironda
of data to predict six months of purchases. Search activity References

does improve model fit and predictiveness.
Glackin and Adivar test a model consistent with Expec- Brynjolfsson, Erik, Wang Jin, and Kristina McElheran. 2021. The
tancy Theory to predict financial success of individuals power of prediction: Predictive analytics, workplace comple-
ments, and business performance. Business Economics 56 (4):
in a direct sales context. First, the authors apply multiple 217–239.
supervised machine learning approaches to group sales reps Hair, Joe F., Jr. 2007. Knowledge creation in marketing: The role of
based on high performance financial outcomes. Second, the predictive analytics. European Business Review 19 (4): 303–315.
authors classify sellers to predict their respective level of Iacobucci, Dawn, Maria Petrescu, Anjala Krishen, and Michael Ben-
dixen. 2019. The state of marketing analytics in research and prac-
outcomes. tice. Journal of Marketing Analytics 7: 152–181.
Hansen and Wilson examine clickstream data to predict Petrescu, Maria, and Anjala S. Krishen. 2019a. Software and data in
the growth of country-specific websites. Country-specific analytics: Lending theory to practice. Journal of Marketing Ana-
data related to culture, economics, and infrastructure are lytics 7 (3): 125–126.
Petrescu, Maria, and Anjala S. Krishen. 2019b. Strength in diversity:
incorporated to test a diffusion of innovation model. Methods and analytics. Journal of Marketing Analytics 7 (4):
Sterlingsdottier et al. apply various analytical tools to pre- 203–204.
dict performance of cryptocurrency. Despite volatility found Petrescu, Maria, and Anjala S. Krishen. 2020. The dilemma of social
in cryptocurrency trading markets, the authors find a model media algorithms and analytics. Journal of Marketing Analytics
8 (4): 187–188.
similar to predicting stock market performance. The model Sarstedt, Marko, and Nicholas P. Danks. 2021. Prediction in HRM
includes sources of discussion of cryptocurrency such as research–a gap between rhetoric and reality. Human Resource
social media networks to improve predictiveness. Management Journal. https://doi.org/10.1111/1748-8583.12400.
Curiskis et al. present a customer lifetime value model
Publisher's Note Springer Nature remains neutral with regard to
that attempts to meet challenges with more conventional jurisdictional claims in published maps and institutional affiliations.
measures of customer lifetime value including temporal
data, multiple products, and differing buyers. The authors
introduce a hierarchical approach to overcome current limi-
Michael A. Levin joined the University of Louisiana at Lafayette as
tations of customer lifetime value. chair of the Marketing Department and Associate Professor in 2023.
It is our hope that the articles here help illuminate these He previously was on the faculty at Otterbein University in Wester-
new frontiers in forecasting, predicting, and explaining, as ville, OH. Michael’s research focuses on strategy, in particular the
well as blaze a trail for future research and advances in the relationship between competitive advantage and the firm’s financial
performance. He has published in Journal of Business Research, Jour-
application of analytical techniques. Potential fruitful areas nal of Personal Selling & Sales Management, and Journal of Market-
for further research might explore the development of addi- ing Theory & Practice. He has given presentations to and conducted
tional frameworks, theories, and models, along with the workshops for digital analysts and marketing professionals on various
application of new contexts, tools, and software. aspects of analysis. He also coaches students in differing analytics
competitions.
We would like to close by thanking all of the authors that
contributed to this special issue, as well as all of the review- John T. Gironda is an Associate Professor of Marketing at Nova
ers for sharing their insightful and constructive feedback. Southeastern University. His teaching and research interests include
Finally, we would like to thank the Journal of Marketing Marketing Technology, Digital and Social Media Marketing, Brand-
ing, Marketing Strategy, and Consumer Privacy Issues. His research
Analytics editors Maria Petrescu and Anjala S. Krishen for has been published in Industrial Marketing Management, Psychology
entrusting us with putting together this special issue, as well & Marketing, Journal of Retailing and Consumer Services, Journal
as all of their help and support during its development, from of Marketing Theory and Practice, and Journal of Marketing Man-
inception to publication. agement. He has also presented papers at a number of conferences
including the American Marketing Association, Academy of Market-
ing Science, ISMS Marketing Science Conference, and Society for
Marketing Advances.
Declarations
Conflict of interest On behalf of all authors, the corresponding author
states that there is no conflict of interest.

New frontiers in forecasting, predicting, and explaining

Uploaded by

Copyright:

Available Formats

New frontiers in forecasting, predicting, and explaining

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

New frontiers in forecasting, predicting, and explaining

Uploaded by

Copyright:

Available Formats

Journal of Marketing Analytics (2023) 11:559–560

New frontiers in forecasting, predicting, and explaining:

of data to predict six months of purchases. Search activity References

You might also like