Case Study

Leading Japanese Departmental Store Derive Key Actionable Insights via Data Analytics and Visitor Analysis

The Challenge

The client is a leading Japanese departmental store well-known for its excellent service, commitment to quality, and industry-leading assortment, which exemplifies the best of Japanese retail and is recognised as a fashion innovator in the local departmental store market. Due to the onset of the COVID-19 pandemic, the client has seen its local revenues declined by 34% in 2020. Although the Singapore economy is slated for recovery due to local vaccination efforts, factors such as continued social distancing measures, reduced footfall traffic, and low volume of visitor arrivals continue to dampen the recovery of the overall retail industry. While the overall retail segment in Singapore has yet to recover to pre-pandemic levels at the end of 2021, the grocery/ supermarket segment has grown by 27% from SGD 97M in 2019 to SGD 123M in 2021, possibly due to higher in-home consumption. To leverage on the business opportunity within the grocery segment, the client turned to ADA by leveraging XACT, ADA’s propriety data management platform and AI-driven solution, to uncover data-driven insights on customers in their existing store locations, the locations of their primary competitors, as well as the demographics of selected locations with the goal of increasing customer acquisition.

The Strategy

ADA introduced a 5-step approach towards identifying and understanding the behaviour of target demographics in selected location.
  1. Location Selection
    Firstly, we worked closely with our client to identify 10 mall outlets, with competitors as our focal locations to geofence.
  1. Geofencing
    We geofenced the selected locations either by mapping the whole mall, or a section of the mall to best capture the audience for our analysis.
  1. IFA Extraction
    The IFAs of devices seen within the geofenced locations were extracted according to an agreed upon timeline with the client. (i.e., between Jan 2021 to Jan 2022)
  1. Profiling
    The IFAs were then ran through our XACT database for consumer profiling.
  1. Derive Insights
    The identified data sets were analysed, focusing on the following 10 areas for our visitor analysis:
      • Cross-visitation patterns
      • Willingness to travel
      • Footfall traffic to mall
      • Age group, gender, and affluence.
      • Persona
      • Online category and offline behaviour
      • Catchment heatmap

The Execution

Through ADA’s robust database and methodology, analysis of the 10 identified malls were completed in a 120-page insights report with details on:
  1. The target audience for each brand/mall were mapped out to determine the concentration of the target audience based on where they have been seen. Colour gradients were used to showcase least to most concentration catchment areas.
  2. The distance travelled by target audience to arrive at the identified malls is tracked from their home/work locations with a radius of <4KM, <8KM, <12KM and <16KM. Home location is inferred by identifying locations the devices are seen regularly during nighttime and work location is inferred by identifying locations devices are seen regularly during daytime.
  3. The movement of the target audience were tracked by day of the week – where different days of the week are categorised broadly for footfall analysis by:
    • Weekdays
    • Weekends
    TIME OF DAY – (3-hour intervals)
    • 12:00 AM – 2:59 AM
    • 3:00 AM – 5:59 AM
    • 6:00 AM – 8:59 AM
    • 9:00 AM – 11:59 PM
    • 12:00 PM – 2:59 PM
    • 3:00 PM – 5:59 PM
    • 6:00 PM – 8:59 PM
    • 9:00 PM – 11:59 PM
  1. Age group, gender, persona, and offline and online behaviours were derived and supplemented using the XACT.
  2. Affluence is derived from condition-based attributes based on device, home location, and appearances of where an IFA frequents. The device price tiers and home property price tiers are then segmentised using percentiles of:
    • Bottom 30% to determine low affluence
    • Mid 40% to determine mid affluence
    • Top 30% to determine high affluence

The Results

The client was able to have a firm understanding on the type of consumers who are seen in the malls. A comprehensive report on each of the 10 outlets were produced to show the mall profile, demographics, willingness of visitors to travel, footfall density, affluence, and cross-visitation behaviours as well as its peak periods, thus allowing the client to profile potential consumers that they could win over from competitors and gauge the largely contributing tenants in different outlets based on the demographics and psychographics of consumers that visited the various outlets.
[class^="wpforms-"]
[class^="wpforms-"]