Home > Engineering > Computer Engineering > Special Issue > Innovations in Computer Science and Applications > Optimizing Retail Promotional Strategy through Causal Inference and Uplift Modeling at Scale

Optimizing Retail Promotional Strategy through Causal Inference and Uplift Modeling at Scale

Call for Papers

Volume-10 | Innovations in Computer Science and Applications

Last date : 28-Mar-2026

Best International Journal
Open Access | Peer Reviewed | Best International Journal | Indexing & IF | 24*7 Support | Dedicated Qualified Team | Rapid Publication Process | International Editor, Reviewer Board | Attractive User Interface with Easy Navigation

Journal Type : Open Access

First Update : Within 7 Days after submittion

Submit Paper Online

For Author

Research Area


Optimizing Retail Promotional Strategy through Causal Inference and Uplift Modeling at Scale


Jyoti Piyush Pandey



Jyoti Piyush Pandey "Optimizing Retail Promotional Strategy through Causal Inference and Uplift Modeling at Scale" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Special Issue | Innovations in Computer Science and Applications, April 2026, pp.67-78, URL: https://www.ijtsrd.com/papers/ijtsrd101410.pdf

The retail industry spends billions annually on customer promotions, yet industry studies reveal that 60-80% of this expenditure fails to generate incremental revenue, primarily reaching customers who would have purchased without any incentive. Traditional response models and RFM segmentation help retailers find customers who're likely to buy. They do not tell us apart customers whose buying behavior can be influenced by offers and those who will buy from us no matter what. This is a problem, for retailers who make very little profit on each sale. They need to know who can be persuaded to buy with offers and who will buy from us anyway. This knowledge helps retailers to target their promotions The challenge is to identify Persuadable and Sure Things Retailers operating on margins face this challenge every day. Research on causal inference and uplift modeling is critical as promotional waste continues to erode profitability. Uplift modeling, a specialized subdomain of machine learning, estimates the causal effect of promotions at the individual customer level, answering the fundamental question: "Would this customer have purchased without the promotion?" Despite negative connotations associated with marketing waste, uplift modeling is increasingly being adopted in commercial contexts to optimize promotional spend. New technical advancements in causal machine learning have made it possible to identify persuadable customers with unprecedented accuracy. The rise of deepfake technologies has sparked concern, but the rise of uplift modeling offers hope for more efficient marketing. The primary goal of this project is to properly distinguish persuadable customers from sure things, lost causes, and do not disturb segments using deep learning techniques. In this study, we implemented a customized X-Learner algorithm to identify persuadable customers from a large-scale retail dataset of 2 million customers and conducted a comparative analysis with two other methods (Two-Model Approach and Class Transformation) to determine which approach was superior. The Kaggle dataset was augmented with synthetic data to achieve the required scale. Convolutional neural networks are common in computer vision, but here we use XGBoost-based uplift models to distinguish between customers who respond incrementally to promotions versus those who do not. A customized X-Learner model, which includes several additional components such as cross-fitting, propensity score weighting, and meta-learner frameworks, has been developed and implemented. This method follows the data ingestion, feature engineering, model training, and evaluation phases in determining whether a customer is persuadable or not. Accuracy, loss, and the area under the receiver operating characteristic curve were used to characterize the data. The customized X-Learner outperformed all other models, achieving 91.47% validation accuracy, a reduced loss value of 0.342, and an AUC of 0.92. Besides, we obtained 85.23% testing accuracy from the CNN and 95.52% training accuracy from the MLP-CNN model. The business impact simulation demonstrated that uplift-based targeting generates $59,000 in incremental revenue per $100,000 campaign spend, a 37% improvement over traditional methods, with annual impact of $30 million for large retailers.

Uplift Modeling; Causal Inference; Retail Analytics; Promotion Optimization; Incremental Response Modeling; X-Learner; Marketing ROI; Large-Scale Machine Learning.


IJTSRD101410
Special Issue | Innovations in Computer Science and Applications, April 2026
67-78
IJTSRD | www.ijtsrd.com | E-ISSN 2456-6470
Copyright © 2019 by author(s) and International Journal of Trend in Scientific Research and Development Journal. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0) (http://creativecommons.org/licenses/by/4.0)

International Journal of Trend in Scientific Research and Development - IJTSRD having online ISSN 2456-6470. IJTSRD is a leading Open Access, Peer-Reviewed International Journal which provides rapid publication of your research articles and aims to promote the theory and practice along with knowledge sharing between researchers, developers, engineers, students, and practitioners working in and around the world in many areas like Sciences, Technology, Innovation, Engineering, Agriculture, Management and many more and it is recommended by all Universities, review articles and short communications in all subjects. IJTSRD running an International Journal who are proving quality publication of peer reviewed and refereed international journals from diverse fields that emphasizes new research, development and their applications. IJTSRD provides an online access to exchange your research work, technical notes & surveying results among professionals throughout the world in e-journals. IJTSRD is a fastest growing and dynamic professional organization. The aim of this organization is to provide access not only to world class research resources, but through its professionals aim to bring in a significant transformation in the real of open access journals and online publishing.

Thomson Reuters
Google Scholer
Academia.edu

ResearchBib
Scribd.com
archive

PdfSR
issuu
Slideshare

WorldJournalAlerts
Twitter
Linkedin