Why do Indian malls see
1000s of visitors a day
but struggle to keep retailers solvent?
India has over 750 operational shopping malls. A significant number, 74 as of 2025, are classified as "ghost malls": high vacancy, declining footfall, tenants exiting. Most weren't killed by bad locations. They were killed by a slow, structural mismatch between who walks through the door and what the business model requires of those people.
Superior-grade malls show under 3% vacancy and strong rent growth since 2019. Mid-tier malls, particularly in Tier-2 cities, face pressure from both directions: e-commerce absorbing commodity categories and consumer expectations rising faster than the mall product is evolving. Tier-2 cities are precisely where this plays out most visibly, with retail infrastructure in place but consumer habits still being formed.
82 people, stopped mid-visit.
Real context, real responses.
Mall Intercept Survey
We physically stationed ourselves at Magneto Mall and approached shoppers directly on the floor, in the food court, and near the exit. They filled a structured 16-question questionnaire on the spot. Capturing people inside the actual environment produces more accurate responses than asking them to reconstruct the experience later at home.
Quota-Based Sampling
We maintained a strict gender quota (50/50, achieved exactly) and deliberately sought variation in age and occupation. This prevents the sample from being dominated by whoever happens to be most present at a given time, typically students on weekday afternoons.
Age Distribution
Occupation
Monthly Household Income
Visit Frequency
8 interviews. Four themes that explain the numbers.
Semi-structured interviews · n=8 · Mix of visitors (5) and mall staff/retailers (3) · Magneto Mall, Raipur · Inductive thematic analysis (Braun & Clarke, 2006)
Footfall is not the problem. What happens inside is.
Self-reported spend. Funnel stages are inferred, not directly observed.
Three visitors walk into a mall. Only one opens their wallet.
Segmented by self-reported spend per visit. Mutually exclusive. n=82.
This is the segment the entire tenant revenue model is built on. Retailers price their rent assumptions around conversion from this group. The problem: they represent less than 4 in 10 people through the door.
They are not a problem to be solved. Many are low-income visitors who simply cannot afford the mall's price points, and their presence is socially legitimate. But they represent the gap between the headline footfall number and the commercial reality that number is being used to imply.
The data contradicts the conventional narrative.
Cross-tabulations from primary survey data. Spend scored 0-4 (Nothing=0, up to ₹500=1, ₹501-1500=2, ₹1501-3000=3, ₹3001-5000=4). Scroll each card to see how scores were derived.
The browser outspends the buyer. Average spend score for window shoppers: 2.38. For those who came to shop for specific items: 2.12. Intentional buyers likely pre-researched online, arrived with a price ceiling the mall can't meet. Window shoppers arrive open.
Window shoppers: n=8, sum of scores=19, avg=2.38
Shopping w/ intent: n=25, sum of scores=53, avg=2.12
All other purpose groups fell between 2.07 and 2.30.
Respondents shopping online 2-3 times per month have the highest average mall spend: 2.42. Those who rarely or never shop online score just 1.56, the lowest of any group. Online and physical retail are complements here, not competitors.
Online 2-3x/mo: n=24, avg spend score=2.42
Once a week online: n=4, avg=3.00 (small n, treat with caution)
Once a month online: n=29, avg=2.03
Multiple times/week: n=9, avg=1.89
Rarely / never: n=16, avg=1.56
Visitors coming more than once a week average just 1.50 on the spend scale, the lowest of any frequency cohort. Once-a-week visitors score 2.27. The most habitual visitors have stopped needing a reason to be there, which means they have also stopped needing to buy anything.
More than once/week: n=6, avg=1.50
Once a week: n=11, avg=2.27
2-3 times/month: n=27, avg=2.11
Once a month or less: n=36, avg=2.14
First time today: n=2, avg=1.50
Four hypotheses tested. The results reframe the problem.
Spearman · Cochran's Q · One-way ANOVA · Chi-square + Cramer's V · Logistic Regression
Primary data, n=82, Magneto Mall Raipur.
Electronics, Groceries, Footwear: p > .20, not significant.
| Predictor | OR | p |
|---|---|---|
| Online frequency | 0.93 | .785 |
| Age group | 1.38 | .461 |
| Leisure purpose | 2.99 | .097 † |
| Dwell time | 1.26 | .531 |
The mall hasn't lost everything. But it's losing the wrong things first.
Respondents stated a preference for each category. n=82 per category.
| Category | Mall | Online | No Pref. | Verdict |
|---|---|---|---|---|
| Clothing & Fashion | 48 | 30 | 4 | Mall Holds |
| Footwear & Accessories | 42 | 26 | 14 | Mall Holds |
| Groceries & Daily Needs | 38 | 24 | 20 | Mall Holds |
| Electronics & Gadgets | 25 | 33 | 24 | Online Leads |
| Beauty & Personal Care | 28 | 39 | 15 | Online Leads |
15-29% of respondents show no stated preference - consumers who could be pulled either way. Electronics (24 undecided) and groceries (20 undecided) are the most contested. These are not loyal to either channel; they will follow whoever reduces friction first. For the mall that means faster checkout, better parking, cleaner navigation - not price cuts.
The mall's defensible moat is tactile experience - categories where fitting, touching, and trying still matter. Electronics and beauty have been commodified: spec sheets are enough, and doorstep delivery wins. Doubling down on clothing and footwear experiencability - better fitting rooms, styling assistance, exclusive in-mall collections - is the priority. Ceding electronics is rational. Ceding clothing would be fatal.
Three observations.
Each with a clear operational implication.
These are directional, not prescriptive. The evidence base is a single-location survey. Validation at scale is a logical next step.
The operational implication is not subtle: design the floor for discovery, not navigation. Pop-up formats at atrium nodes, rotating activations, corridors as encounter space rather than throughput. Separately monetized pop-up real estate also creates a revenue layer less exposed to anchor tenant stress.
The implication is to embed in the online journey rather than compete with it. In-mall return counters for e-commerce orders drive footfall with purchase-adjacent intent. Try-before-you-order QR pilots convert the browsing session into a physical visit. Exclusive in-store SKUs unavailable online create a reason to show up that digital channels cannot replicate.
Reporting commercial footfall separately from total headcount would be a harder number to present to tenants and investors. It would also be more honest. For habitual non-buyers, two options: design targeted micro-conversion pathways - student bundles, affordable food anchors, gamified spend triggers - or accept their presence as the cost of the social infrastructure role the mall now plays. Either is a coherent position. Pretending they don't exist is not.
The questions this research opens are more interesting
than the ones it closes.
Self-reported spend is not verified spend
Respondents stated what they typically spend - we didn't observe transactions. Social desirability bias likely inflates spend figures and deflates "I came for nothing" responses. The real conversion gap is probably wider than the data shows.
60% students is a sampling artifact
Students are overrepresented relative to their share of the actual spending population. Mall intercept during weekday afternoons structurally produces this skew. The findings about student behavior are robust; the aggregate spend findings should be read with this in mind.
Single location, single city
Magneto Mall is a specific asset in a specific market. The patterns identified here - the archetypes, the paradoxes - are plausible as general phenomena, but they cannot be validated from one site. Replication across mall tiers and geographies is the obvious next step.
No transaction data
The most powerful version of this study would cross-reference survey responses with actual POS transaction data from mall retailers. That would allow direct validation of the spend claims and segmentation. The current analysis is survey-based only.
Does the serendipity premium hold across mall tiers?
Window shoppers outspending intentional buyers was the study's most surprising finding. Does this pattern appear at premium malls? At failing ones? Understanding where it breaks down would tell you something important about what drives it.
At what point does online shopping frequency flip from complement to substitute?
Moderate online shopping correlated with higher mall spend. "Multiple times per week" online shoppers show lower mall spend (1.89). Is there a threshold effect - and if so, where is the inflection?
What would commercial footfall actually look like as a reported metric?
If malls reported "visitors who made at least one retail purchase" alongside total headcount, how would the numbers compare? What would that do to tenant lease negotiations? No one seems to be publishing this.
"Retail research in India tends to ask whether consumers like malls. The more useful question - and the one this study begins to answer - is whether the people who use them are the people the business model is built around."
Every finding in this study traces back
to these cross-tabulations.
Spend coded 0-4: Nothing=0, up to Rs500=1, Rs501-1500=2, Rs1501-3000=3, Rs3001-5000=4. Raw counts from n=82 mall intercept survey, Magneto Mall, Raipur, March 2026.
| Purpose | n | Nothing | Low | Mid | High | V.High | Avg Score |
|---|---|---|---|---|---|---|---|
| Window shopping | 8 | 0 | 1 | 4 | 2 | 1 | 2.38 |
| Food & dining | 10 | 1 | 2 | 4 | 3 | 0 | 2.30 |
| Movies / entertainment | 22 | 2 | 2 | 8 | 8 | 2 | 2.27 |
| Socializing / timepass | 14 | 2 | 1 | 6 | 4 | 1 | 2.07 |
| Shopping - specific items | 25 | 4 | 2 | 10 | 7 | 2 | 2.12 |
| Work / Job | 3 | 2 | 1 | 0 | 0 | 0 | 0.33 |
| Frequency | n | Avg Spend Score | Note |
|---|---|---|---|
| Once a week | 11 | 2.27 | Highest of regular visitors |
| Once a month or less | 36 | 2.14 | Largest group |
| 2-3 times/month | 27 | 2.11 | |
| First time today | 2 | 1.50 | Small n |
| More than once/week | 6 | 1.50 | Paradox 3 - lowest despite highest loyalty |
| Online Frequency | n | Avg Spend Score | Note |
|---|---|---|---|
| Once a week online | 4 | 3.00 | Small n - treat with caution |
| 2-3 times/month online | 24 | 2.42 | Paradox 2 - highest reliable group |
| Once a month online | 29 | 2.03 | Largest group |
| Multiple times/week | 9 | 1.89 | Heavy online, lower mall spend |
| Rarely / never online | 16 | 1.56 | Lowest - contradicts naive intuition |
| Score | Spend Bracket | Count (n=82) | Share |
|---|---|---|---|
| 0 | Nothing | 12 | 15% |
| 1 | Up to Rs500 | 9 | 11% |
| 2 | Rs501 - Rs1,500 | 29 | 35% |
| 3 | Rs1,501 - Rs3,000 | 24 | 29% |
| 4 | Rs3,001 - Rs5,000 | 8 | 10% |