The ‘halo effect’ – how can retailers understand and track the impact of their store estates on their digital channels?
As the retail sector continues to shift towards an omni-channel model of customer interaction, so too has the role of stores evolved. With this shift has come an increased emphasis on using analytics to understand the interaction between the physical store channel and other direct channels.
This requirement is particularly pronounced in the UK where retail has moved furthest and fastest towards omni-channel. Where the store plays a central role in the cross-channel customer journey, and the overall shopper experience that the retail proposition aims to deliver, understanding the relationship between physical and digital channels is crucial in informing the right investment strategy – Should we open stores? And/Or continue to support our store channel? Or are we right to re-balance our investment towards digital channels?
This blog explores the data capture and analytic approaches which shed light on the dynamics of the ‘halo effect’ to understand how the changing physical footprint of a retail chain, whether increasing or decreasing store numbers, drives performance measured in other channels.
Retailers will, of course, routinely open stores anticipating increased incremental revenues through the store channel as a result of the opening. Often, this thinking will incorporate some assessment of the likely cannibalisation between stores, whereby a proportion of the new store sales may be a diversion of sales that would previously have gone to other stores in the network rather than stolen from competitors.
But how often do retailers consider the impact on their direct channels, whether positive or negative? And is this dynamic fully reflected in the investment capex appraisal process?
The ‘halo effect’ describes the positive interaction between the physical store channel and online channels for retailers.
Many sectors and B2C businesses (but not all) demonstrate this positive relationship, where the presence of stores in a territory helps drive a higher uptake in direct channels.
Many retailers assume that offline and online channels are competitive – if a customer purchases through my website, then perhaps it follows that they are less likely to purchase in my store – but the evidence suggests that the opposite is more often true, whereby direct and store channels are complementary. Why is this?
There are two principal reasons explaining why online and offline channels can work in tandem across a geography:
- The presence of physical stores can increase brand awareness, with the stores partly operating as a showcase for the brand that can be exploited through both the store channel itself and via direct channels (e.g. mobile, online, contact centre)
- Many customer journeys (such as click & collect) are deliberately aimed at embedding a store interaction as part of the pathway to purchase. In this model, stores are still very important, by either allowing customers to see and touch items before purchasing (in the reserve online model), or providing a service support function that the direct channels alone may fail to replicate.
Understanding the ‘halo effect’ gives retailers a way of measuring the importance of physical stores to direct sales and vice versa. This helps to drive robust ‘shape of chain’ decisions.
If there is a positive relationship between store and online channels within a business it is highly likely this pattern evolves over time. Repeatable and consistent analytics are key to understanding if and how this effect is at play. So, how do you analyse the halo?
There are six primary analytical approaches to measuring the ‘halo effect’ in multi-site, multi-channel businesses. Each approach requires offline and online customer data to different degrees. Ideally, analysis explores the relationships at play through more than one of these lenses.
1. Aggregate catchment penetration analytics
This approach (the simplest of the six proposed) compares evidence of offline and online market share penetration across the defined store catchments (or regions) to establish overall patterns relating to penetration by channel. Customer data can be used to define catchments around stores. A positive correlation in comparing these two datasets suggests that the ‘halo effect’ is probably present in your business.
2. Drivetime penetration analytics
This approach explores how the offline and online market penetration dynamics evolve with increased distance from stores. In our experience, whilst customers living closer to stores will inevitably have a higher propensity to transact with the retailer through the store channel they also, in most sectors, display an increased likelihood of transacting through direct channels too. Understanding the halo effect through this lens provides the first step in building spatial interaction models that can accurately predict the impact on direct sales that will result from changes to the store network.
3. Post-sector scoring analytics
This approach compares online penetration within geographic areas (e.g. a post-sector or SOA) vs. a ‘branch accessibility’ scorecard (which amalgamates the overall access to the store network for residents living in particular areas). The scorecard may use a range of factors such as venue size (e.g. VENUESCORE), catchment population, location type and drivetime or travel time to assess accessibility. For example, this may show a higher online penetration near ‘accessible’ stores which facilitate easy click & collect services etc.
4. Pre and post change analytics
Trend analytics using one or more of the above approaches, exploring time-series data within catchments following changes to the branch network can add further weight to validate the relationships which point in time analytics help to identify.
5. Customer lifetime analytics
A variation on approach 4 is to assess channel interactions over time for individual customers relative to their access to the store network. This approach is the most advanced in that it requires a Single Customer View of data over an appropriate time horizon which may or may not be available.
6. Mobile data analytics
The use of geospatial digital applications connecting customer data to movement data points (e.g. loyalty apps using location services) opens the door to a new generation of analytics that can demonstrate the impact of stores on specific digital activities, in a more granular way than ever possible before.
Although much of the thinking to-date has been the ‘reverse halo’ (i.e. how to use digital to drive in-store actions – e.g. through contextual marketing and promotions), the opportunity to track digital behaviours that correlate to store visits and/or access to a strong store network will add a new dimension to our understanding of how the physical world encroaches and influences consumer behaviour, in the digital realm.
Whether channels operate in a competitive or a complementary way within a retail business, the challenge for an omni-channel retailer is to value the physical store estate accurately.
This insight is vital in informing store property investment decisions; in a retailer promoting customer journeys within and across a range of channels, it is simply not possible to design the right shape of chain for the future if the role the store channels plays, in influencing spending decisions that may transact in other channels, is not fully understood.
For further reading on this topic, read Accenture Strategy’s report – Adaptive Retail: Redefining the role of the stores to improve competitiveness.