Retailers like to talk about the consumer as if such a person actually exists.
The average shopper. The average household. The average basket. The average spend. It all sounds tidy. It gives boardrooms something to hold onto. But I have never trusted the average consumer as a serious idea, because no one lives in an average market. People live in neighbourhoods, commuter corridors, apartment blocks, suburbs, student districts, industrial towns, tourist zones, and office catchments. They move through space in patterns. They buy according to habit, access, income, convenience, time pressure, weather, culture, transport, and mood.
Consumer demand is not floating in the air. It has a geography.
That is the part many businesses still miss. They look at national sales numbers and think they understand the market. They look at regional revenue and think they understand customer behaviour. They look at online search trends and think they understand intent. But demand does not become real until it lands somewhere. A store. A station concourse. A petrol forecourt. A mall entrance. A delivery zone. A residential cluster. A lunchtime office district that is busy on Tuesdays and dead on Fridays.
This is why mobility data has become so valuable. It does not just show where people live. It shows where they actually go.
And in commercial strategy, that distinction matters enormously.
For much of the twentieth century, retail geography was easier to read. Residential districts, work districts, shopping streets, and transport nodes followed more predictable patterns. The high street had a clear function. The office district had a clear rhythm. The shopping centre pulled from a defined catchment. Suburbs had commuter logic. Demand moved in recognisable loops.
That world has not disappeared, but it has become more unstable.
Hybrid work changed weekday demand. E-commerce changed trip purpose. Delivery apps changed food geography. Rising living costs changed where value-seeking consumers go. Tourism recovery changed central districts. Transport disruption changed movement flows. New housing developments created demand before services caught up. Ageing populations changed local spending patterns. Student districts expanded and contracted with visa rules, rent pressures, and university growth.
A national retail figure may show mild growth, but that says little about the street-level reality. U.S. retail sales rose by around 3.7 percent in 2025, while overall foot traffic increased by about 1.8 percent, but even that headline hides a more cautious pattern in the second half of the year, when volume growth softened. That is the important part. The market looked stable from above, but underneath it was becoming more selective.
I think this is the central commercial lesson of the post-pandemic period. Demand did not vanish. It moved. Then it kept moving.
Some districts recovered. Others did not. Some commuter zones came back three days a week, but not five. Some malls became social anchors again. Others became tired boxes with parking lots. Some neighbourhoods gained daytime spending because people worked from home. Others lost office footfall and never replaced it. The consumer became more mobile in some ways and more local in others.
That contradiction is exactly why location intelligence matters.
Traditional market analysis often begins with residential demographics. How many people live within five kilometres. What is the average income. What is the age profile. What is the household size. These are useful inputs, but they are not enough.
A residential catchment tells you where people sleep. Mobility data tells you where they spend time.
That difference can change everything. A wealthy residential district may look attractive on paper, but if residents commute elsewhere, shop online, and spend weekends outside the area, the local opportunity may be weaker than expected. A transport hub may look crowded, but if people pass through under time pressure, the opportunity may suit convenience retail rather than discretionary shopping. A student district may appear low income, yet support strong food, entertainment, and value retail demand because of high frequency and dense daily movement.
This is where the old idea of catchment analysis needs updating. A catchment is no longer just a circle drawn around a store. It is a living movement pattern.
The best retail sites are not always the ones with the most people nearby. They are the ones where the right people pass with the right intent at the right time. That is a much harder thing to measure, but it is also far more useful.
I think many failed retail expansions come from confusing visibility with demand. A location can be visible and still wrong. It can sit on a busy road but have poor access. It can have footfall but no dwell time. It can sit near competitors but serve a different consumer rhythm. It can be surrounded by people who simply do not need what is being sold.
The map can look promising. The movement pattern can say otherwise.
The phrase “market expansion” often sounds grand. New city. New region. New country. But growth increasingly depends on micro-markets.
A micro-market is not just a smaller area. It is a distinct pocket of behaviour. A cluster of office workers around a station. A residential zone with young families and high school traffic. A tourist strip that peaks in evenings. A logistics corridor with service demand. A luxury mall catchment with international visitors. A suburban retail park shaped by car access and bulk purchases. A university district where demand changes sharply during holidays.
Two locations only a few kilometres apart can perform in completely different ways. One side of a station may attract commuters. The other may attract nightlife. One mall may draw families. Another may draw young professionals. One neighbourhood may support premium coffee. Another may support discount groceries. One corridor may work for fast casual dining. Another may require drive-through formats.
This is not theory. It is observable in the data.
Mall performance in 2025 showed how uneven retail behaviour had become. Indoor malls led full-year visit growth, outdoor malls performed better during the holiday period, while outlet malls lagged. That tells us something important. Consumers were not simply returning to stores in one uniform pattern. They were choosing formats according to value, convenience, weather, social purpose, and spending intent.
This is where I think many executives go wrong. They ask, “Is retail footfall recovering?” That is the wrong question. The better question is, “Which formats are gaining visits, from whom, at what times, and for what kind of spending?”
The answer will not be national. It will be spatial.
Office attendance is one of the clearest examples of why mobility data matters. In July 2025, U.S. office visits reached their highest level since the pandemic began, with office visits up 10.7 percent compared with July 2024, although still 21.8 percent below July 2019 levels. That sounds like recovery. But again, the word recovery can be misleading.
A partial return to the office does not recreate the old geography of demand.
If workers return three days a week, Monday and Friday remain weaker. Lunch demand becomes concentrated midweek. After-work spending changes. Dry cleaners, sandwich shops, gyms, convenience stores, and transport retail all feel the difference. Some city centres become busy enough to look alive, but not consistent enough to support the same business models as before.
New York office traffic reportedly surpassed pre-pandemic levels for the first time in 2025, while cities such as Miami, Atlanta, and Dallas showed stronger recovery patterns than the national average. San Francisco recorded strong year-on-year growth but from a weaker base. These variations matter because they show that office recovery is not one story. It is a city-by-city, district-by-district, sector-by-sector story.
A financial district with strict return-to-office mandates behaves differently from a technology district where hybrid work remains embedded. A government office cluster behaves differently from a creative industry neighbourhood. A central station catchment behaves differently from a suburban business park.
I think the office return debate has been too ideological. People argue about remote work as if it is a moral issue. For commercial planning, the practical issue is simpler. Where are people on Tuesday at 1 p.m. Where are they on Friday at 5 p.m. Where do they buy lunch. Where do they stop after work. Which routes have recovered, and which have not.
That is the geography of demand now.
The cost-of-living squeeze has also changed consumer movement. Value-seeking is not just a psychological behaviour. It becomes geographic. People travel further for cheaper groceries. They shift from premium centres to discount formats. They consolidate trips. They compare petrol costs, parking costs, delivery fees, and time costs without necessarily describing it that way.
Discount and grocery-led formats benefited from this behaviour in recent years, while some discretionary categories saw softer movement. Planned Black Friday and Cyber Monday spending in 2025 was expected to fall by around 4 percent to roughly $622 per consumer, while participation was still expected to rise. That is a revealing pattern. More people shopping, but with tighter budgets. More activity, but less generosity. More hunting, less indulgence.
That kind of behaviour creates new spatial patterns. Consumers may still visit malls, but spend more carefully. They may visit discount stores more often. They may avoid unnecessary journeys unless multiple needs can be met in one trip. They may favour locations with easy access, free parking, and clear value. They may still desire premium goods, but buy them less often or in different channels.
This is why retail demand cannot be understood only through income. Income matters, but confidence matters. Inflation memory matters. Debt pressure matters. Transport cost matters. Rent pressure matters. The household budget has a geography too.
A consumer living in a high-income area but facing high mortgage costs, school fees, and car dependence may behave more cautiously than a lower-income urban renter with dense access to low-cost services and transport. The old categories are too blunt.
Mobility data helps reveal the behaviour beneath the label.
The physical store is no longer just a sales point. It is part of a wider spatial system involving online browsing, delivery zones, returns, social media discovery, click-and-collect, and local brand presence.
This changes how demand should be measured. A store may not capture every transaction directly, but it may strengthen online sales in the surrounding area. A showroom may reduce uncertainty before digital purchase. A return point may increase customer confidence. A small urban format may serve convenience and brand visibility rather than full inventory. A suburban big-box store may act as both retail site and fulfilment node.
This is why footfall alone can mislead. High traffic with low conversion may still have strategic value if the site supports brand presence or omnichannel behaviour. Low traffic with high intent may be more profitable than a busy flagship. A store with modest sales may still reduce delivery costs or increase customer retention in a key catchment.
I think this is where modern retail strategy becomes much more interesting. The question is no longer simply, “Where should we open a store?” It is, “What role should this location play in the wider network?”
That requires spatial thinking. You need to understand movement, demand density, logistics coverage, competitor influence, delivery time, return behaviour, and channel interaction. A store is not an island. It is a node.
Poor site selection usually fails long before the doors open. The warning signs are there, but they are ignored.
The demographic profile looks attractive, but the access is poor. The rent is expensive because the address is fashionable, not because the demand is deep. The footfall is high, but it belongs to tourists who do not match the offer. The competitor map looks favourable, but there is a hidden anchor nearby pulling demand elsewhere. The catchment looks wealthy, but the residents already have established shopping routines outside the area.
These mistakes are expensive because property decisions are sticky. Leases are long. Fit-out costs are heavy. Staff are hired. Marketing is committed. Once the wrong location is chosen, every later decision becomes harder.
I think site selection should be treated less like expansion and more like risk management. The upside matters, but the downside is brutal. A weak site does not just underperform. It absorbs management attention, damages brand perception, and ties up capital that could have gone somewhere better.
The best location analysis should be slightly uncomfortable. It should challenge the obvious site. It should expose why a popular location may not work. It should show why a less glamorous site may be stronger. It should separate prestige from performance.
That is what good spatial analysis does. It removes the romance from the map.
The real value of location intelligence is not that it produces better maps. It produces better commercial judgement.
It helps a retailer understand why two similar stores perform differently. It helps a telecom operator decide where network investment will meet real demand. It helps a property developer understand whether a district is becoming more valuable or just more expensive. It helps a restaurant chain understand whether evening movement supports dining or only takeaway. It helps logistics firms understand where delivery demand is forming before competitors see it.
It also helps organisations avoid false confidence. A national market may be growing while your target micro-market is weakening. A city may look attractive while the relevant catchment is saturated. A format may be trending while the local population does not support it. A competitor may appear close but serve a completely different movement pattern.
The strongest businesses will not be those that simply collect more data. Everyone has more data now. The advantage belongs to those that interpret spatial behaviour with discipline.
That means combining mobility data with demographics, transaction patterns, transport access, land use, competitor mapping, and local context. It means asking not only where people are, but why they are there, how long they stay, what they are likely to do, and whether that behaviour supports the business model.
Without that interpretation, mobility data becomes another dashboard. Interesting, but not decisive.
Consumer demand is often described in emotional language. Confidence. Sentiment. Value. Preference. Loyalty. All of that matters. But beneath those words is movement. People move through physical environments. They make choices shaped by distance, friction, time, access, habit, and convenience.
That is the hidden geography of consumer demand.
It explains why broad market averages can mislead. It explains why some stores thrive while others fail in apparently similar locations. It explains why mobility data has become central to commercial strategy. It explains why micro-markets now matter more than regional generalisations.
My own view is simple. Businesses that still think in broad catchments and average consumers will keep making expensive mistakes. The world has become too fragmented for that. Demand is more fluid, more local, more time-sensitive, and more dependent on movement patterns than many traditional models admit.
The consumer has not disappeared from physical space. The consumer has become harder to read.
That is exactly why geography matters.