Adding Value to Promotional Forecasting through a Decision-Support Dashboard Quantifying Error Impacts
A Case Study at Picnic
F.W.A. Kraanen (TU Delft - Civil Engineering & Geosciences)
A.J. van Binsbergen – Graduation committee member (TU Delft - Civil Engineering & Geosciences)
J.H.R. van Duin – Mentor (TU Delft - Technology, Policy and Management)
G. Bandini – Mentor (Picnic)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Promotional forecasting poses a significant challenge for online supermarkets, as consumer demand tend to be highly volatile. Such volatility often leads to forecasting errors, which in turn have financial, operational, and reputational consequences. This study investigates the financial and operational challenges in the context of Picnic France and addresses the knowledge gap concerning the measurement of the diverse impacts of over- and under-stocking caused by forecasting inaccuracies.
The research objective was to develop and embed an artefact that quantifies the consequences of forecasting errors in financial and operational terms. The Design Science Research Process (DSRP) was applied to structure the study. The phases were structured as follows; first the ’Problem Identification and Motivation’ phase was conducted in the introduction setting out the problem, structure the research and introduce the case study. The next phase ’Objectives of a solution’ is conducted by mapping the complete promotional forecasting process to understand problems and clearly pinpoint where the artefact can address the problem. In the ’Design and Development’ the indicators and their calculations are defined, followed by the ’Demonstration’ phase which shows the design of the artefact, a decision supportive dashboard. Afterwards, the evaluation phase presents the results for two articles and subsequently discusses the feedback received from end-users. The report finalises with the ’Communication’ phase, presenting how the dashboard will be integrated in the existing process. This approach resulted in a dashboard that will be supportive in two key steps of the promotional forecasting process: the first when products are selected for promotion, and the second when Picnic’s analysts review the forecasted quantities and decide whether to adjust them based on the insights provided by the dashboard.
The study offers both academic and practical contributions. Academically, it adds to the limited body of research focusing on the operational impacts of forecasting errors and on understanding how these impacts influence inventory management decisions. By integrating financial and operational cost factors into the assessment of forecasting performance, the study advances the understanding and quantification of over- and under-stocking risks. This claim is supported by a preliminary literature review, which revealed no relevant studies addressing these aspects. Practically, the developed dashboard serves as a valuable decision-support tool within Picnic, aiding in both selecting products for the promotion and reviewing the forecast quantities to be stocked for the promotion.