The nature of shopping activity is changing in response to innovation in retailing and the growth in online channels. There is a growing interest from transport researchers, policy makers, marketing and retail businesses in understanding the implications of this change. However, existing tools and techniques developed for analysing behaviour in traditional retail environments do not adequately represent emerging complexities resulting from digital innovation. The overarching goal of this research is to advance the development of new modelling and data collection tools for studying choice behaviour in today's increasingly complex retail environments. While data collection and empirical applications relate to grocery shopping in London, the conceptual discussions and modelling frameworks developed are generalisable to all shopping activity. The contributions of the presented work are at three levels. First, a comprehensive conceptual framework was developed incorporating interactions between multiple agents that drive the transformation of the industry, and individual choice behaviour within this broader perspective. Secondly, it is a significant challenge for researchers to find appropriate data sets, which combine travel and shopping related information and also capture online activity. For empirical applications here, an augmented version of a widely accepted revealed preference consumer panel data was used in together with API based data mining tools that offer great potential for enrichment in discrete choice modelling. Third, discrete choice models were developed using gathered data for the joint choice of channel, shopping destination, and travel mode. This extension to traditional destination and mode choice models is critical as it provides the tools to quantify the effects of increased online shopping on traditional store formats and travel patterns. Results revealed important insights into how shoppers choose from online and in-store alternatives, and how mode choice fits in with these decisions. During our study we also identified and explored substantial limitations in empirical applications of discrete choice models. We analysed issues of identification caused by sample size constraints, potential estimation bias due to potentially restricting choice set generation assumptions, and challenges that arise when newly introduced innovative alternatives show low-adoption rates.