Who is actually shopping in my store?
Measuring retail behaviour from real-time vehicle movement data.
Retailers relied on consumer demographic information for decades even before the “big data” revolution. Census data, repackaged into segments by consumer demographics organisations, give you information from digitality, age and spending profile.
But how do retailers know which of the nearby residents actually get into their car and come and spend in their stores? Retail catchment areas were a start, and loyalty cards the next step, mapping individual baskets to addresses. But increasingly, retailers are recognising that purchasing decisions are driven by very specific contexts, and what I want to buy today and tomorrow might be completely different, depending on where I’ve come from, where I’m going and how I feel.
At Gyana, we saw the potential to use some of our external data sources to provide a high quality picture with real context on consumer purchasing behaviour. This is what we found.
Case Study: Large Supermarkets Around Sussex And South London
We focused on supermarket A and supermarket B, two household grocery chains with multiple stores that support large car parks. Our question: where do people who shop in these supermarkets come from, and where do they go afterwards? What can we learn from this?
We used an origin destination data source from our internal data warehouse. This contains anonymised and aggregated origin and destination for around 10% of all vehicle journeys made around the Sussex / south London for a two month period.
To give you an idea of what this looks like, here is the entire dataset plotted for the journeys that started and ended inside the area:
Many of those journeys are consumers driving from home to a supermarket and back again. We filtered the data to only include consumer vehicles (not taxis, vans) and extracted all the journeys that started at the home of the consumer and ended at the supermarket, and vice versa.
For example, here is the start location of every trip to a supermarket (orange), with the supermarkets in blue.
And finally, we joined that data on consumer demographics information from the CDRC (consumer data research centre). They use the census data to segment the U.K. population into groups like “Rural white-collar workers”.
Forget retail catchment areas: what we’ve produced now is a distribution of who actually shops in each supermarket, and when. And it’s not based on guesswork, surveys or car counting- it’s based on real data.
Here’s a sample of how powerful this is.
Wonder Who Prefers Supermarket A over B?
Supermarkets position their brand to target specific demographic, but time and time again big brands have found that their “brand myths” are just that. What does the real data say?
We’ve plotted the relative fraction of supermarket A customers minus supermarket B customers per segment. A score of +1 means “100% in this segment shops exclusively at supermarket A” and vice versa for supermarket B.
What do we find? The three most popular segments for supermarket A are:
- 6a4: Ageing in suburbia
- 5b1: Delayed retirement
- 4a2: Private renting new arrivals
And the three most popular segments for supermarket B:
- 1b2: Rural white-collar workers
- 2a1: Student communal living
- 6b2: White suburban communities
When do people shop at supermarket A over B?
Not everyone shops at the same time. Knowing when different people are going to come directly impacts advertising both in store and out. We found a compelling statistic when comparing the two supermarkets on the weekend (Supermarket A is blue, B is orange):
Answer: Supermarket A is notably more popular for a midday shop on Saturday, whereas supermarket B is notably more popular early on a Sunday.
How far are people willing to travel?
Intuitively, you’d expect people to travel further to a big supermarket for a weekend shop. We can test this and drill down to see which demographics are more willing to go the distance:
Answer: Thursday is the longest, and Saturday is the shortest.
External data provides context to your sales
There’s a huge potential to go deeper into this data. People drive to many other places before and after the supermarket, and that provides granular context on their motivations and mindsets when they come to your store.
For instance, which demographics do their school post-school supermarket runs on a weekday afternoon, vs a more leisurely run on Saturday afternoon? Maybe it depends on the number of children, or the distance of the supermarket? How can you target in store ads to leverage this opportunity ?
The traditional approach to this problem is to roll out a loyalty card scheme, or iPhone application, or in store beacon technology. These solutions are expensive, lack coverage, don’t apply to competitors and don’t provide any context on what is happening before and after the shop. And why bother, when the data already exists (it’s just not you who created it).
With realtime, ground truth data, these are the kinds of questions you can ask. Gyana has the data and the big data stack to find patterns that might just change the way you do business. We’ll provide an intuitive, SaaS solution for your team. Turnaround time? Less than a week.
Contact us at info@gyana.space and let’s start a conversation.
