Consumer and Demographic Data Archives

The importance of interpretability within machine learning: A case study with Acorn

Posted on: 30 October 2024

In today’s data-driven world, understanding and leveraging data effectively can transform how we approach various challenges. One powerful technique in the realm of data science and machine learning is K-means clustering. This is the algorithm behind Acorn, CACI’s flagship data product which groups together postcodes with similar characteristics into segments. Whilst this process can all be done fairly easily with modern ML techniques, one crucial component is often overlooked: human interpretability.

What is K-means clustering in the context of Acorn?

K-means clustering is an unsupervised machine learning algorithm used to group data points into clusters based on their features. The goal is to minimize the variance within each cluster and maximize the variance between different clusters. To demonstrate how this is done in practice, the following steps are used in the build of Acorn, although the same process will broadly apply to any K-means algorithm:

Initialisation: The number of clusters (K) for initial segments is chosen
Assignment: Every UK postcode is assigned to the nearest cluster segment based on information such as average house price and children per household before the mean distance between postcodes and their segment is calculated
Update: The cluster centroids are repositioned and mean distances for each segment are recalculated
Repeat: The assignment and update steps are repeated until distances are minimised and centroids no longer change significantly

The need for human interpretability

This process looks like it requires very little intervention from a human at any point – and that’s because it doesn’t – in theory. However, in practice as with all unsupervised ML techniques, for Acorn to be of any use as a segmentation tool, information must be scrutinised at every step along the way.

First, the number of clusters must be chosen either randomly or with prior domain knowledge. Acorn has been around for nearly 50 years and the last iteration featured 5 marketable segments, making it a good starting point. However, after stakeholder input from across the business, a conscious choice was made to increase the number of groups to 6 to reflect changes in society since the previous build.

Next, input variables need to be decided on before the clustering process begins. With a wealth of data from the 2021 Census as well as newly available information such as disposable income data, this list of variables needed to be carefully refined for Acorn to be both a mathematically and commercially sound product. As an example, the inclusion of planning extension data was tested as a promising new input to highlight areas undergoing gentrification. However, the results from this didn’t make intuitive sense and so this variable was excluded from the model. Such conscious decisions were also made to ensure that Acorn is fully compliant with the UK Equality Act and exemplify the need for human input before even running a model.

With the input variables decided on, K-means clustering can be applied, but the element of human input does not end here; the segment outputs must be dissected to ensure they are dissimilar from each other and contain an acceptable minimum number of data points. In the context of Acorn, this meant looking at the number of postcodes in a segment and measuring average values for the input driver variables. For example, averages for percentage of houses that are detached and household income were measured for each segment. These figures were found to be highest for groups containing individuals more likely to be in managerial roles, which acted as a useful sense check and allowed such groups to be labelled as more affluent. The number of postcodes within each group also needed to be sufficiently large to allow different marketing strategies to be applied for each segment.

Postcodes were also analysed visually on a 2D scale using a Python package to identify overlaps between segments. The difference between the old and new versions of Acorn can be seen below.

Figure 1: A representation of old (left) and new (right) Acorn postcodes drawn down from a multidimensional space to a 2D visual, with each colour representing an Acorn segment.

The reduced ‘bleed’ of segments into other segments as seen from this visual made it clear to analysts that this newer version of Acorn has much more well defined segments – a result of new data, advanced ML capabilities and of course, stakeholder input.

Finally, and arguably most importantly, the ultimate question must be answered: will the outputs of the model be valuable to the end user? If the answer is not a definite yes, then the process needs to be reviewed and, more often than not, this will involve decisions around the human element of the process rather than consulting the ever-growing list of ML techniques and tweaks. To increase the value of Acorn for clients across sectors including retail, finance, charities and utilities, questions from survey partners were mapped onto Acorn to provide insights such as digital attitudes and channel preference by segment. These questions are updated on an annual basis based on stakeholder feedback to ensure questions are current and relevant to clients.

Conclusion

Ultimately, results need to be useful for individuals or teams and there is currently no way of achieving this without human interpretation and intervention to some degree at every stage of the process. This rings even more true in consumer segmentations, where there is no ground truth or right or wrong to compare to. Lots of packages will allow you to build a model with very little input or intervention, especially with the rise in autoML capabilities, but to build a trustworthy, useful product, humans need to be on hand at every step along the way.

Challenging areas for the grey belt: London and the South West

Posted on: 28 October 2024

In this final article in our grey belt series combining CACI’s housing demand data with VirginLand’s grey belt site identification, we take a deeper dive into two areas where the impact of the grey belt will be less obvious: London and the South West.

This is not to diminish the overall value of the initiative, but to point out that it is not the single answer to the UK’s housing challenges. To be most effective, grey belt reallocation should be considered alongside other mechanisms to accelerate housing delivery such as brownfield, infill, repurposing and urban regeneration.

How will the grey belt initiative affect London?

Home to over 7 million adults, London is by far the most densely populated region in the UK. As a result, demand for housing is particularly acute in the capital and the conversation is dominated by affordability. It’s easy to see why; house prices are 11 times the average household income, and private rent is 37% of income. Although households in London earn 17% more than the national average, these high prices mean that homes are 59% less affordable to buy and 32% less affordable to rent.

In this context, any initiative to increase the overall supply of housing in this region is welcome; particularly if it’s targeted at the more affordable end of the scale. So, what impact will the grey belt have in London?

Although home to 14% of the population, London can house just 0.4% of all grey belt homes – a total of 1,955 dwellings across 31 sites. This is not for want of green belt (22% of the London region, by area, is currently designated as green belt), but for want of suitable locations that could be re-designated. Analysis by VirginLand shows that just 0.2% of the available green belt land is likely to be reallocated grey belt, with much of the London’s green belt holding additional designations like Designated Open Space, Country Park, Woodland or Nature Reserve and Conservation Area and Grade 1-3b agricultural land grades.

The challenge in London is also compounded by the location of the sites relative to movers. Being on the outskirts of the urban sprawl, just 11% of all London home movers live within the catchment of the identified sites; roughly 192,000 individuals. Although more than enough to absorb any new homes delivered, the scale of movers puts into perspective how limited the impact would be on demand; if all sites were built out to their fullest, there would be 98 movers for each home.

While there is little doubt that the 31 identified locations would be additive to the overall housing stock, the question is over how much of an impact these limited sites can have on a particularly strained market. With the population set to grow by another 6.1% in the next 10 years, London will need other initiatives, alongside the grey belt, to accelerate housing delivery in more urban neighbourhoods.

How will the grey belt initiative affect the South West?

Just 4.5% of the South West is designated as green belt, well below the national average. It therefore follows that grey belt opportunities in the South West will be similarly limited, and just 228 potential sites have been identified with the combined potential to deliver 11,868 new homes.

While this is not an insignificant number of homes, it represents just 2.2% of the total grey belt opportunity spread across 9.4% of the population. The location of grey belt sites also limits the initiative’s regional impact, as just 18% of the 1.6 million potential South West movers live within the grey belt catchment, against a national average of 36%.

Although limited in scope for the region as a whole, there are some pockets where the grey belt will be more impactful, and the characteristics of the catchment movers in these locations point to the type of homes that should be prioritised.

With concentrations of sites close to urban populations in the likes of Bristol, house-to-earning ratios and rent-to-earning within the grey belt catchment are higher than those outside of it (7.1 times income and 27.4% of income respectively). High concentrations of Family Renters, Tenant Living and Cash-Strapped Families within this catchment, and relatively large sites averaging 1.7 hectares, suggest a particular opportunity to deliver larger mixed neighbourhoods with high levels of rental product.

As with London, the grey belt initiative has the potential to support some of the housing needs of the South West, but an overarching housing strategy for the region should also be mindful of the 82% of home movers that live outside of the grey belt catchment.

How CACI can help?

To learn more about how you can ensure that your developments are meeting the demands of local movers, contact CACI.

Missed the previous blogs? Find the links to the series so far below:

How grey belt sites will help tackle the UK housing crisis – CACI – Tolga Necar

Grey belt sites: what they are, locations & impact on housing – CACI – Steve Norman and Sam Bedford, Virgin Land

Assessing the impact of the grey belt initiative on a National scale – CACI – Tolga Necar

How will the grey belt initiative affect North West England & Scotland? – CACI – Tolga Necar

How grey belt sites will help tackle the UK housing crisis

Posted on: 30 September 2024

The UK has not been meeting its house building targets for some time. This is not new news, but it is worth reiterating the scale of shortfall. Over the past five years, we have consistently delivered 20% fewer homes than were targeted: a total miss of nearly 300,000 homes (or put another way, an entire year’s target).

Couple this with projected population growth and we can see why house building has made its way up the political agenda. The population is expected to swell by 3.8 million people over the coming 10 years, and naturally, these people will need somewhere to live.

In this blog series, CACI and Virgin Land will uncover key questions around the future of house building targets and how they can be addressed via grey belt sites, including their locations and desirability, whether they create suitable opportunities for inhabitants and how they vary by region.

Where should new house building targets be focused?

Population growth will concentrate around major towns and cities, especially given that 14 of the largest 15 towns and cities have projected growth rates that outstrip the UK average. However, cities are not always where the market has delivered new homes. In fact, eight of those top 15 towns and cities have housing delivery rates that lag behind the national average. Therefore, new housing targets should be geographically directed to the places that people want to live.

Who should be the focus of house building targets?

Housing growth should be targeted at the people that need it most. Left to its own devices, the market has delivered new housing that concentrates around a few demographic groups. Using CACI’s Acorn segmentation to profile new homes delivered in the last five years, we can see clear trends in the data; Tenant Living (young, urban renters) comprise 18% of new homes but just 12% of the population, Semi-Rural Maturity and Mature Success (two affluent, older groups who are likely to be downsizing) collectively account for 20% of new homes but just 13% of the population. Lower affluent, urban families such as Limited Budgets, Hard-Up Households and Cash Strapped Families, however, have received disproportionately little housing development.

This is not to lay blame on housebuilders; the commercial challenges of development in urban environments are clearly contributing factors, however, the impact is one of acute supply challenges in specific demographic groups who are coincidentally the groups most likely to be living in over-occupied housing. To maximise the impact of housebuilding initiatives, the route forward requires a more collaborative approach, which the newly formed MADE Partnership may well deliver.

How will the grey belt make a difference for house building targets?

Central to the Labour government’s housing policy is the rezoning of poor-quality green belt sites into the “grey belt”, effectively opening swathes of previously unavailable land for development. But how influential could this policy change be? CACI and Virgin Land have partnered to uncover the potential impact that opening up the grey belt can have on housing market dynamics.

How CACI can help?

Stay tuned for the next blog in this series, where we’ll dive deeper into grey belts, their locations and their impact on housing. In the meantime, contact CACI to learn more about how you can ensure that your developments are meeting the demands of local movers.

Why retail destinations should invest in consumer experiences & perceptions

Posted on: 9 July 2024

Want to increase your visitors’ spend by 25%? Invest in your amenities.

Facilities are a vital part of retail and leisure destinations. Despite not directly producing turnover, they play an essential part in driving performance. Through research from our Shoppers Dimensions dataset– our database of over 1 million respondents across 270 UK-wide destinations which enables key performance indicator (KPI) benchmarking of assets against similar locations across the UK to contextualise performance and enhance decision-making– we analysed how various KPIs are impacted by consumers’ experiences and perceptions.

So, how exactly are services and amenities within retail destinations affecting consumers’ behaviours? How can retail destinations leverage these insights to bolster experiences and perceptions?

How do consumers’ overall shopping experiences influence their spending behaviours?

According to our research, retail destinations capable of improving their rating of a person’s overall shopping experience may recognise an increase in their average retail spend by £21. There is an uplift across the board when overall shopping experience is rated 5 out of 5, with average retail spend increasing by 25% and catering spend by 17%.

How do experiences & perceptions of toilets impact retail destinations?

It may not be a glamourous topic, but toilets are often called out by customers as an issue. They are expensive to renovate and maintain, and without a direct revenue stream associated with them, it is easy to think of toilets as a cost. Despite this, our data shows that investing in facilities can actually drive performance.

Firstly, when looking at shopping centre locations of those that rate the toilet facilities 5 out of 5, our data shows that this leads to an uplift in time that a person stays at the destination by 16%, which accounts for 12 additional minutes per customer. But how does this additional dwell time translate into spend? Customers that give toilets a top rating record a 26% uplift on their average retail spend, an increase of approximately £21.34 per customer.

Retail is not the only category affected. In fact, catering conversion experiences an uplift of 5 percentage points and the average spend on catering increases by 19%. There is therefore direct value to unlock by maintaining and improving these facilities, even if that means you have to spend a few pennies to do so.

How to attract more family groups from further afield

Family groups can be a hard group to target, but once at the destination, they are likely to come for ‘Big Day Out’ trips which are associated with a higher average spend. For many destinations, this group tends to live further afield, such as in the suburbs of a city. When family facilities are rated higher, there is an uplift in their drivetime by 23%, an increase in their dwell time by 17% along with an uplift of 25% in associated retail spend. Showing that better family facilities draw in these high-spending visitors from further away

How do car park experiences & perceptions impact interactions with the rest of the shopping centre?

One of the most interesting findings we came across when looking into the impact of ratings was with overall parking experience. This is another topic that consumers are passionate about; ever hard-to-please, the consumer wants it to be cheaper, with more spaces and of a better quality. But do better perceptions really lead to stronger key performance metrics? In short, the answer is a resounding “yes”. Those who rate the overall parking experience 5 out of 5 see an uplift in dwell, retail and catering average spend and conversion. The greatest uplifts are in dwell time and average retail spend. On average, dwell time will see an uplift of 17% (14 minutes) while average retail spend will see an uplift of 30%, leading to an average increase in spend of just over £25.

Key takeaway: higher perceptions equal higher spend

Overall, our data shows that the higher the perceptions, the more people will spend and the longer they will stay. This is the case when we look at the ratings for overall shopping experience, cleanliness, overall parking experience, family facilities, customer services, signage, architecture and toilet facilities. While it might not be glamourous, strong perceptions of parking experience and toilet facilities do lead to an increase in key performance indicators, proving that there is value to be unlocked by investing in these facilities.

How can CACI help?

At CACI, we understand the impact that driving improved perceptions of facilities within a retail destination can have on consumers’ behaviours, such as which amenities encourage people to visit from further away, stay longer or spend more on their trip. To gain a better understanding of how consumers interact with places, reach out to us to discuss how we can help you measure your performance and identify growth opportunities.

Most substantial challenges for healthcare organisations to address in 2024

Posted on: 25 June 2024

Tackling health inequalities is a tremendous challenge. It requires healthcare organisations to understand the demographics, lifestyles, behaviours, needs, and external pressures that individuals across the country face daily with greater accuracy. Access to accurate and detailed data significantly impacts an organisation’s ability to develop a robust response to inequalities and determine which services will meet local needs.

In our recent webinar for NHS England on “Tackling Health Inequalities with Effective Data & Insight”, we explored the impact of our datasets and insights on NHS England’s ability to tackle current health inequalities and devise strategies to improve future outcomes.

So, what have the findings from our various datasets and our Voice of the Nation (VOTN) Q1 2024 survey shown regarding the behaviours and health concerns of different demographic and affluence groups across the UK? How can healthcare organisations apply these findings to improve outcomes for their local communities?

Half of the survey respondents are concerned about their personal wellbeing and mental health

Personal wellbeing and mental health are incredibly important considerations for the NHS. According to our survey results, these have been hugely concerning for people of various ages across the UK, with 50% of our VOTN Q1 2024 survey respondents claiming to be concerned about both. This is the highest number of respondents for these sentiments that CACI has ever seen in the four years of this survey being conducted, demonstrating the need for healthcare organisations to review their current offering of personal wellbeing and mental health services avoiding a ‘one size fits all’ approach that targets all ages.

Millennials are the most concerned of all age groups about their health

While the traditional assumption may be that younger generations are more carefree and less preoccupied with the concerns of the world, our survey results have shown the opposite. Millennials were the most concerned of all age groups (from Boomers to Gen Z) for their personal wellbeing and mental health, with more than two-thirds feeling this way. This further reiterates the necessity of ensuring that all age groups—particularly Millennials—are offered relevant personal wellbeing and mental health services.

Affluence does not shield from health concerns

Our survey results indicated that personal wellbeing and mental health concerns have been affecting individuals across all affluence levels. While one might assume that higher-affluence individuals experience fewer wellbeing and mental health concerns, our findings revealed that as many as half of the respondents from the higher-affluence Acorn categories of Luxury Lifestyles and Established Affluence expressed concern about these aspects of their lives. Respondents from the Low Income Living Acorn category expressed the highest level of concern for both areas.

These insights provide concrete evidence for healthcare organisations to tailor their services based on the specific needs of different affluence groups, rather than relying on open data or assumptions. These results demonstrate the right healthcare services must be accessible across all affluence levels.

How can CACI help?

CACI can help healthcare organisations tackle health inequalities, supporting a range of clinical areas of health inequalities from severe mental illness (SMI) to maternity and chronic respiratory disease (CPD) to early cancer diagnosis, hypertension case-finding and more. Our partnership with NHS England provides all 42 integrated care boards (ICBs) with free access to a variety of datasets that are being used to tackle health inequalities.

Contact us today to learn more about our partnership with NHS England or to find out how our datasets can improve outcomes for your healthcare organisation.

Most impactful food-to-go transaction trends into 2024

Posted on: 12 June 2024

With the continuing trend of hybrid work within worker hubs, consumers’ food-to-go spending in quick service restaurants (QSRs) remains concentrated on some days and displaced on others. Consumers’ wallets also continue to face an ongoing squeeze, resulting in pressures on day-to-day convenience spend.

So, what transactional trends are being observed across different demographic groups, geographies and price-points as these trends continue? What impact do these trends have on operators’ future openings strategies and overall performance?

Food & beverage have become increasingly prominent on High Streets

Over the course of 2019 to 2023, most retail centres in all asset classes have grown their share of food and beverage (F&B) outlets, noting an increase in over 90% of centres in the top four classes— City Centres, Regional Malls, Major Town Centres and Satellite Centres. Despite F&B having become increasingly prominent in shopping and retail parks, there has been a mixture of increases and decreases observed in towns, transport hubs and leisure parks, raising the question of whether oversaturation has had a role to play in some locations.

Centres are polarising

Over the same time period, city centres, regional malls, major towns centres and satellite centres have dropped in their overall level of consumer attractiveness in line with consumers’ changing behaviours. So much so, that the four largest asset classes have seen declines in over 90% of their centres. The picture is a bit more mixed as the retail hierarchy descends into towns, transport hubs and leisure parks, however, with an average of 40% of centres in these asset classes seeing a decline. The ever-increasing proportion of consumer spend moving online has undoubtedly prompted these downward trends.

Given the vast differences in changes at an asset class level, and with many exceptions at a centre level, having access to detailed data on the changing attractiveness and demographics at centre level is vital.

Customer behaviours towards QSRs continue to change

Many may think that post-Covid QSR demand is just about Tuesday to Thursday, driven by changes in working behaviour, but this is an over-simplification. CACI’s local centre mobile app data analysis within our Location Dynamics suite shows that while areas like Fleet Street/St. Paul’s in the City of London now do have a pronounced Tuesday to Thursday peak, it’s far from the universal norm. As shown by the dark-shaded time segments in the graphs below, places like Barkers Pool in central Sheffield have a very pronounced Friday and Saturday night economy. This further contrasts with central Eastbourne, which has maintained a more traditional Monday to Sunday 9 a.m. to 4 p.m. custom and a strong weekend daytime custom.

Ultimately, locations are different, and successful operators must understand the different ‘missions’ their customers will be on to ensure they meet their customers’ needs and ensure that they staff their outlets to provide the right level of services at times demanded by their customers.

For food-to-go retailers to engage with consumers at the right time and in the right place, it will be critical for them to consider:

The F&B offers in local areas
Changing consumer behaviours as a reflection of new and embedded worker patterns,
Centre attractiveness
Overarching market shifts that impact footfall on specific days and times.

How CACI can help?

With these trends in mind, it is critical for food-to-go retailers to have a detailed understanding of who their customers are, where they are located and what times of the week they are most likely to interact with your chain or restaurant. It is equally important to understand your place in terms of its attractiveness to customers and the effect of its location on driving footfall.

Data is key to maintaining a competitive edge amidst evolving trends, an area where CACI excels in providing support. Find out how we can keep you and your team ahead of the curve by reaching out to us today.

Most impactful holiday and air travel trends for 2024

Posted on: 30 May 2024

If the last few years of pandemic uncertainty and budget constraints amidst the ongoing cost of living crisis have shown us anything, it’s that travellers have become increasingly conscious of the cost of travel. As a result, they’ve placed increased value on having an optimal travel experience to justify its cost.

We examined the current driving factors behind optimised travel experiences in our Voice of the Nation Q1 2024 survey, where we asked 2,000 respondents how they felt about an array of travel changes and how the cost of living, airline loyalty and more have impacted their travel choices into 2024.

So, what shared values and needs do travellers of all ages and affluence levels seem to have in common this year? How have these forthcoming trends been affecting the wider travel industry?

Travel spend will increase in 2024 despite decreases in most other sectors

When asked whether their anticipated spending will decrease, increase or stay the same this year compared to last, holidays actually rank third among areas people expect to increase spend in 2024– with groceries and commuting costs coming in first and second– despite an overall expected decrease in spend in other areas this year.

Plans to holiday abroad skew significantly on affluence lines

From Boomers to Gen Z, more than half of respondents from every age group plan to holiday in some capacity– both in the UK or abroad– in 2024.

When it comes to taking holidays abroad, 38% of respondents are making plans and budget room to do so this year. Of these respondents, as much as 50% come from the higher affluence Acorn categories of Established Affluence and Thriving Neighbourhoods. Approximately one in three of the lower affluence categories of Steadfast Communities, Stretched Society and Low Income Living share the same sentiment.

A quarter of all respondents have no intention of travelling this year, and 22% plan to visit another part of the UK, which would appear to be in an effort to save on travel spending. In reality, no matter where you go for your next holiday, the same proportion of respondents agree that cost will be the biggest determinant behind their destination. 36% of those staying in the UK say that they will go on holiday within the UK because they prefer it to going abroad, showing that while cutting travel costs is a major driver, it is not necessarily the only one.

Half of respondents claim no loyalty to an airline

When asked what the contributing factors towards airline loyalty are, half responded that they have no loyalty to any airlines.

Roughly one-third (31%) of those who are loyal towards an airline felt that their loyalty is driven by more than one factor, such as convenience, discounts and luggage/check-in benefits. In comparison, 18% felt there was only a singular driving factor behind their airline loyalty, showing that where loyalty is in play, it is usually multi-factorial.

Convenience is the most significant driver behind airline choices

Apart from price, respondents’ most significant contributing factors towards airline choices when booking trips came down to flight times and route, both of which are also the only factors heavily skewed by affluence. Nearly 60% of the Established Affluence and Thriving Neighbourhoods category respondents reported this to be significant, compared to just 35% among Low Income Living. Gen Z, however, scored this even lower, with just 32% finding this to be significant and instead placing more emphasis on the ease of booking at 37%.

Families are much more affected by cost this year

In terms of holiday planning this year, one-third of respondents said that they wanted to keep their holiday costs as low as possible to maximise value for money and felt that costs would be the greatest determinant of where they holiday in 2024. Among those with children, 40% said that cost is the biggest determinant of where they go on holiday.

Sustainable transport options appeal much more to Gen Z

Of all demographics, Gen Z appear to be the most motivated by sustainability when planning their holidays, both in terms of those taking immediate action but also those who would like to travel but feel unable to presently. In fact, 18% of Gen Z respondents said that they will be cutting down on air travel in 2024 due to their growing environmental concerns, compared to just 8% among the rest of the population.

How CACI can help?

As the travel industry evolves with travellers’ changing sentiments, holiday and air travel operators must be equipped with the necessary understanding of who their customers are, what their motivations for travel are, what they seek from their travel experiences and how to deliver optimal experiences that will drive loyalty. Data is integral to this, which is where CACI excels in providing support.

To find out how we can keep you and your team amidst turbulent times, get in touch with us today.

Impact of turnover vs. footfall for shopping destinations in 2024

Posted on: 23 May 2024

Footfall has historically been the go-to method for measuring a shopping destination’s performance, conducted through pressure sensor mats, light sensors tracking shoppers’ entry and exit movements, advanced camera systems and more. Although ubiquitous across the retail industry, only measuring the number of people entering and exiting a store misses important aspects of true store performance. The current pace of change in consumer behaviors demands that commercial landlords and occupiers know more about their performance drivers if they are going to thrive.

So, why is this the case? What do commercial landlords need to know about turnover and footfall to stay afloat?

How consumers’ changing behaviours towards shopping locations affect footfall

Since 2019, vendors across the UK have experienced an overall 11.5% drop in footfall. While this may sound like catastrophic news for retail destinations, the truth behind the headline footfall figures is perhaps surprising– an overall rise in consumer spending. Although a shift in consumers’ shopping behaviours is undeniably present, its impact may not be as profound as it seems.

Frequency has been a major driver of this, dropping by 31% over the last five years, meaning that consumers have been visiting shopping places much less often. However, the amount being spent by consumers when visiting shopping locations has climbed 29% over the last five years, counteracting declining footfall.

This increase in trip spending is not just an inflationary rise – the fundamental reason to visit and our behaviours on visits have changed as a result. Successful locations are those that are adapting to the new shopper landscape.

How consumers’ changing spending habits, values & “missions” affect footfall

What consumers are spending any disposable income on has also been changing. While retail conversion has remained relatively unchanged, there have been evident increases in Catering and Leisure conversions on the same trips, meaning consumers are increasingly combining a shopping trip with food/drink or a leisure activity. It is this combination of shopping, browsing, eating/drinking and leisure that has led to the overall increase in spending per trip.

These comparisons can be illustrated through what we at CACI call “missions” from our Shopper Dimensions dataset, which illustrate the trip someone is on at a given time, and attribute “missions” to the tangible actions someone takes once at the shopping destination, such as browsing, spending, time spent, etc., to assign a “mission” to each trip.

According to our findings, consumers are relinquishing their less engaged “missions” but concentrating trips around the “Big Day Out” trip. This is illustrated in the shifting profile of the top three missions in Shopping Destinations, which explains why a decline in footfall does not necessarily equate to declining spend. At a glance:

“Big day out” missions are our more engaged trips. They may be less frequent, but they are ones where multiple retail stores are often combined with Catering and Leisure, resulting in a trip spend 2.4x the average mission. Since 2019, these missions have grown to 23% of all shopping missions.
37% of “spending time” missions have no purchasing associated with them. While they may contribute to footfall figures, they do not directly contribute to sales-through-tills. Having dropped off post-Covid-19, these trips are now holding flat at a lower shelf.
“Routine top-up” trips are quick, functional and emotionally disengaged trips that a spend of just 47% of an average trip. These trips are dropping out of our repertoire and can be substituted online.

We can therefore see that looking in greater detail at the changing nature of the trips made provides a clearer understanding of commercial asset performance than simply tracking the overall volume of trips.

Key levers to conclude turnover & application methods to target growth outcomes

To make a meaningful impact in asset performance, commercial landlords must move beyond measuring just the number of visits and start reporting the different levers of shopping location spend.

While there are nuances behind the headlines that apply individually to each location, all spend at a shopping location can ultimately be boiled down to three key levers:

The volume (number) of unique shoppers they have
The frequency of consumers’ visits to a shopping destination
The value that each shopper spends per trip.

Commercial landlords should consider applying the following methods to each lever to effectively target growth outcomes:

Volume: Convert footfall (visits) into ‘spenders’ and target engagement strategies at driving scheme trial; measured by the percentage of the catchment population currently shopping with you (penetration).
Frequency: Embrace the different role that your asset plays for different cohorts, diversifying the occupier offering to give shoppers more reasons to return on different missions.
Value: Determine the highest spending shopper groups to target, segment customers and tailor offers to them to increase cross-shopping opportunities and drive value.

What does good look like?

Now is the time for commercial landlords to leave pre-pandemic comparisons behind. Footfall may be down overall, but the evolution of consumers’ shopping destination behaviour serves as a reminder that relying on the past as an indication of how assets should behave will not lead to longer-term success. If anything, these behaviours have demonstrated that the types of trips people continue to use shopping locations for are more engaged and valuable than ever before.

Our unique view into how and where consumers are spending has been made possible with the help of datasets like Shopper Dimensions, which enable KPI benchmarking of assets against similar locations across the UK and leverage transactional and data spend insights to enhance decision-making. We can help you calculate the impact of each shopper metric and the headroom compared to peers and catchment.

To find out more about what Shopper Dimensions can do for you and your business, speak to one of our experts today.

How River Island use ResolvID to effectively perform identity resolution on customer data

Posted on: 14 May 2024

Background

River Island is a beloved high street retailer that has brought leading fashion trends to UK shoppers for over sixty years, with both a digital and in-store presence.

When the brand began building a marketing and analytics data technology environment with only a Single Customer View (SCV)— a single record that merges all customer data– available, they recognised the need for a SaaS solution that would be able to perform real-time identity resolution on customer data.

The Challenge

Bringing the entire SCV in-house posed a significant challenge to River Island, having to terminate many data feeds and re-evaluate incoming and outbound data that lacked clarity. The original data feeds were also set up by employees who had since left the business, resulting in a trial by fire with their SCV.

The Solution

CACI configured ResolvID, a cloud native solution hosted on Amazon Web Services (AWS) Cloud infrastructure, to supply River Island with data cleansing, standardisation, identity resolution and deduplication. Developed with a Microservices architecture, the bespoke platform offers significant advantages through its scaling, resilience and flexibility when rapid changes and improvements are required.

ResolvID comprises horizontally and vertically scalable Microservices that perform different functions with a seamless interface to enhance River Island’s accessibility. The solution leverages advanced deterministic name and address matching techniques in conjunction with digital and non-digital identifiers specific to River Island customers and their data. As part of this initiative, CACI took a three-step approach to effectively perform identity resolution on River Island’s customer data.

The Results

Leveraging ResolvID has resulted in many tangible benefits for River Island, including the creation of various customer dashboards to monitor more targeted figures and generate better, more timely data that bolsters targeted customer campaigns. There have also been noticeable improvements in workload efficiencies, such as cutting down the time required to action workloads to increase the team’s focus on refining their future strategy of doing more with their data to retain oversight on customer performance.

Once we swapped to ResolvID, the numbers we got were close enough to give us confidence that the deduplication received from ResolvID worked better than our previous managed service.

Ben Anderton, Technical Lead at River Island, shared how this real-time capability now enables the confident and immediate actioning of data and customer signups to produce effective campaigns based on genuine buying behaviours and generate accurate results.

Read the case study

You can access and download the full case study here.

If you have any questions or want to learn more, please get in touch with us to discuss what strategies and solutions that our team of experts can help you deliver.

How Earls Court Development Company use data to help inform a new neighbourhood

Posted on: 2 May 2024

Background

The Earls Court Development Company (ECDC) has a vision to bring the wonder back to Earls Court. Their latest proposals demonstrate how Earls Court will be put back on the map, re-emerging as a destination to discover wonder, an ecosystem for creative talent and a showcase for one of the fastest growing industries in the world – clean and climate tech. The masterplan includes 4,000 new homes, 12,000 jobs, culture, community, retail, dining and leisure. 60% of the land is unbuilt, maximising open spaces and opportunities for nature to thrive. The site will have a series of cultural venues, alongside a commercial campus creating a global destination for clean and climate tech research and skills. Sustainability will be the green thread, with one of the largest zero-carbon energy loops in the UK powering the site. A hybrid planning application will be submitted this summer and the first phase will commence in 2026.

The Challenge

Understanding how current plans would impact the local market, what retail opportunities should be created and how to create a robust masterplan that would address these factors, despite London’s complex market and a high amount of local competition.
Gauging customers and audience — who is already here, what they do, what they need and where they go — in relation to other large-scale central London developments and regeneration master plans in King’s Cross and Battersea.
Prior to partnering with CACI, the company solely relied on qualitative data to understand peoples’ perceptions and inform their decision making, such as speaking to people within the community and stakeholders.

The Solution

ECDC was keen to ensure that an optimised neighbourhood would be created for residents both within and outside of the development along with workers and users of the space. To achieve this, CACI interpreted and analysed raw data and numbers for the company, bringing them to life and narrating the results through comparable’s and benchmarks.

It’s very clear in the presentations that we’re given — whether it’s for local authorities or internally — that the evidence base is robust and ultimately indisputable. That was helpful in providing that context and equips us with a robust way to create and define the master plan moving forwards.

Tom Branton, Development Director at ECDC

The CACI data sources included as part of this study were:

Mobile App Data: Mobile location data generated a precise view into the location’s catchment and visitor profiles, ensuring ECDC would innately understand how visitor profiles and their respective behaviours varied over time. This helped the company assess who users are along with their demographic and spending power, along with insights into how visits changed over a day and week.
Acorn: CACI’s consumer segmentation model, Acorn, enabled ECDC’s understanding of who new residents would be and their needs, and who would shop at the development.
Location Dynamics: Location Dynamics is CACI’s spatial interaction model, creating a digital mirror of the UK retail landscape replicating consumer flows. The engine of the model is a machine learning algorithm that provides future forecast catchments. For ECDC, CACI used Location Dynamics to understand the expected current catchment and spend, as well as leakage to nearby destinations, to provide a detailed understanding of the local retail landscape.

The Results

Newfound understanding of the ‘size of the prize’ of wider London and tourist demographics and audiences. ECDC historically relied on gut instinct when it came to decision making, but working with CACI ensured they were backed with concrete evidence. For example, CACI’s data showed that one-third of the total potential spend in the development area could come from out of catchment.
Enhanced decision-making through evidence-based data on the community. With the development situated across both the London Borough of Hammersmith and Fulham and the Royal Borough of Kensington and Chelsea, their perceptions of the surrounding community to inform decision-making — while strong — are now rooted in evidential data. This has served to alter their perceptions to ensure that a comprehensive understanding of residents and borough dwellers can be met and their audience narrative can be shaped accordingly.

The Future

In the coming years, CACI will continue to support ECDC in the data-backed planning and construction of residential units, retail landscape and office space development.

Read the case study:

You can access and download the full case study here. If you have any questions or want to learn more about CACI’s solutions, please get in touch with us.