Archetypes of Footfall Context: Quantifying Temporal Variations in Retail Footfall in relation to Micro-Location Characteristics

Abstract

The UK retail sector is constantly changing and evolving. The increasing share of online sales and the development of out-of-town retail provision, in conjunction with the 2008–09 economic crisis, have disproportionately impacted high streets and physical retail negatively. Understanding and adapting to these changes is fundamental to the vitality, sustainability and prosperity of businesses, communities and the economy. However, there is a need for better information to support attempts to revitalise UK high streets and retail centres, and advances in sensor technology have made this possible. Footfall provides a commonly used heuristic of retail centre vitality and can be increasingly estimated in automated ways through sensing technology. However, footfall counts are influenced by a range of externalities such as aspects of retail centre function, morphology, connectivity and attractiveness. The key contribution of this paper is to demonstrate how footfall patterns are expressed within the varying context of different retail centre architypes providing both a useful tool for benchmarking and planning; but also making a theoretical contribution to the understanding of retail mobilities. This paper integrates a range of contextual data to develop a classification of footfall sensor locations; producing three representations of sensor micro-locations across Great Britain: chain and comparison retail micro-locations, business and independent micro-locations and value-orientated convenience retail micro-locations. These three groups display distinct daily and weekly footfall magnitudes and distributions, which are attributed to micro-locational differences in their morphology, connectivity and function.

Using unstable data from mobile phone applications to examine recent trajectories of retail centre recovery

“…Whilst there are some important considerations to make about the temporal stability of such data before using it to answer new research questions, correct use of mobility data offers significant advantages over other economic performance measures for retail centres. For example, mobility data does not certain geographic areas or locations within retail centres, as is the case with footfall sensors (Philp et al, 2022), and typically offers a greater temporal resolution than other 'static' measures of economic performance, such as vacancy rates (Dolega & Lord, 2020). However, it would still be more preferable to use actual sales data to monetise the performance of retail centres, as is the case with individual stores (e.g., Rose & Dolega, 2022), but the potential to do so has not yet been realised, given a lack of suitable data.…”

“…Total spend would be of greatest utility, but is difficult to obtain or estimate given the decentralised nature of retail (er) organisation. As such, there are numerous proxy measures that have been used, such as vacancy rates (Dolega & Lord, 2020;Jones et al, 2022), footfall (Philp et al, 2022;Ntounis et al, 2020) or attractiveness and retail mix (Dolega et al, 2016;Jones et al, 2022). However, such measures are subject to limitations, such as overly privileging certain geographic areas or having limited temporal resolution.…”

“…Thus, what remains clear from this modelling exercise is that retail centre recovery (∆ i ) during this time is dependent on the overall structure, function and regional geography of the retail centres, though to varying degrees, with function and regional geography contributing significantly less. It appears that the structural and catchment characteristics of retail centres remain a greater determinant of changes to retail centre activity during this time, thus more research is needed to unpack how at finer geographical resolutions (as opposed to regions), different structural characteristics of retail centres geographies determine such responses (Dolega & Lord, 2020;Philp et al, 2022). However, there are lots of additional unanswered questions that need addressing, such as the role of multidimensional typologies (e.g., Dolega et al, 2021), seasonal and weather effects (e.g., Rose & Dolega, 2022) and the recent cost-of-living crisis, which has exacerbated inequalities between different regions (Wood, 2019).…”

“…Although complex to capture, there is widespread consensus that data-driven empirical measures of performance hold great value for policy and planning of the future of cities and retail (Enoch et al, 2022 ; Philp et al, 2022 ). There are however no uniform indicators for measuring retail centre performance (Dolega & Lord, 2020 ), owing to the complexity of such a measure, and the influences of internal and external factors (Philp et al, 2022 ), as well as demand and supply (Jones et al, 2022 ). Total spend would be of greatest utility, but is difficult to obtain or estimate given the decentralised nature of retail (er) organisation.…”

“…In addition, existing studies have utilised various forms of data to assess the economic performance of consumption spaces, such as vacancies (Frago, 2021;Dolega & Lord, 2020) and footfall (Philp et al, 2022;Ntounis et al, 2020). The utility of mobility data in answering such questions was first identified in Trasberg and Cheshire (2021), providing significant scope for the use of similar data, such as the Geolytix aggregated in-app location dataset (CDRC, 2021a), to unpack how such responses have manifested in later phases of the COVID-19 pandemic once the key limitations of such data have been addressed.…”