Weighted TURF analysis for product line optimisation as a maximum coverage problem with a customisable definition of reach: Methodology and open source implementation

Abstract

This paper advances total unduplicated reach and frequency (TURF) analysis by introducing a customisable reach definition. Traditional TURF methodologies, often reliant on complete enumeration, are computationally intensive and lack flexibility. The current authors’ binary linear programming model defines reach as customers selecting at least θ items, where θ may be any integer ≥1, enabling a more nuanced engagement analysis. The authors’ implementation can incorporate survey weights and produce multiple solutions despite using a binary linear programming framework that typically defaults to a single solution. The paper is complemented by a publicly accessible R Shiny web app, providing a practical, computationally feasible and adaptable TURF analysis tool that aligns with contemporary consumer research needs. Using a dataset from 200 female multivitamin/mineral gummy consumers, we demonstrate the model’s efficacy, revealing a product line that consistently ranks among the best flavour portfolios under varying assumptions. This article is also included in The Business & Management Collection which can be accessed at https://hstalks.com/business/.