What a footfall benchmark is actually worth
A benchmark is only useful when the things being compared were measured the same way. In footfall, that condition is harder to meet than it looks.
Benchmarks are satisfying to produce and dangerous to misread. Your flagship’s footfall is up twelve per cent year-on-year. Your nearest competitor’s location is down eight per cent against the same period. Someone has made a slide deck. Before that slide goes to the board, one question is worth asking: are those two numbers actually measuring the same thing?
Often, they are not.
The appeal of the comparison
Absolute footfall counts are useful for internal planning. You need to know how many people come through on a Tuesday morning because you are staffing for it. The hourly pattern tells you when the queue at the fitting rooms becomes a problem. Set against the till, the same count becomes the conversion rate behind footfall and sales telling different stories. These are operational questions; they only need your own data, collected consistently.
The moment you compare your number to anything external (a competitor, an industry index, your own portfolio across different counting systems) the requirements change entirely. The comparison only has value if what is being compared was measured the same way, under the same rules, using the same definitions. In practice, retail footfall data rarely meets that standard, and the gaps are not trivial.
What the method actually counts
Consider two sites, both using electronic visitor counting, both reporting daily totals. The first uses a camera-based system that has been calibrated against manual control counts twice a year; it counts inbound visitors only, staff excluded via a separate entrance. The second uses a bi-directional door sensor; staff use the same entrance as customers, and the counter has not been recalibrated since installation. (A Wi-Fi system counts devices, not people, and leans on exactly that calibration step to convert one into the other.)
The two numbers are not comparable. They may be in the same spreadsheet. They may be averaged together in a chain-wide report. But they are measuring something subtly different, with different error rates and different definitional assumptions, and any benchmark built from them carries those errors forward into every decision it informs.
This is not an abstract concern.
A 2021 paper published in PLOS ONE on retail footfall archetypes found significant variation in footfall patterns across location types, which is expected, but also highlighted the degree to which measurement methodology affects the comparability of data across sites.
The conclusion was essentially that footfall data is useful for within-site trend analysis, and considerably more complicated for cross-site comparison.
The index question
One response to the comparability problem is to express performance as an index rather than an absolute count. An index sets each site (or each period, or each location type) to a baseline of 100 and expresses subsequent performance relative to that point. The advantage is that it removes the absolute-volume difference between sites: a large regional shopping centre and a small high-street unit can both be benchmarked against their own baseline without the comparison being distorted by the fact that one receives ten times the visitors of the other.
Ipsos Retail Management’s UK footfall index, which tracks more than 1.2 billion retail visits annually, works on this principle, tracking relative movement across location types rather than claiming a single absolute count for “retail.”
The index is informative precisely because it normalises the data before comparison.
The limitation is that an index still requires the underlying counts to be consistent. If your individual site data is inconsistent from year to year (because a sensor was replaced, a counting rule changed, or a new staff entrance was opened) the index will faithfully reflect an artificial movement rather than a real one.
What the data deliverables need to include
For a benchmark to be usable, the data behind it needs to be documented as thoroughly as the count itself. That means knowing the counting technology and its calibration history, the definitional rules (inbound only or net, staff excluded or not, children counted or not), the known error range, and any changes to any of these variables over the period being compared.
Without that documentation, you are comparing numbers that may look the same but represent different things. A chain that has deployed four different counting technologies across its estate of retail locations over the past decade cannot produce a meaningful ten-year trend unless it has tracked the impact of each transition.
Mall operators and property owners face the same challenge at a larger scale: multiple tenants, multiple counting systems, and a headline footfall figure that is often the product of multiple methodological choices that no two tenants made the same way.
The competitor comparison problem
Cross-competitor benchmarking is the version of this that most readily goes wrong. Published footfall indices aggregate data from multiple operators and technologies, which is useful for understanding macro trends, whether retail footfall overall is rising or falling against the prior year. What they cannot reliably tell you is whether your specific site is outperforming or underperforming relative to a specific competitor in the same retail format, because the underlying data came from different counting systems with different calibration standards.
The honest use of a published index is as a directional signal: is the market moving against us or with us?
That is a useful question. The overconfident use is as proof that your locations specifically are beating the competition, a claim the data methodology typically cannot support.
Where it helps most: lease negotiations
One place where internally consistent footfall data produces clear, defensible value is lease negotiations. A tenant with validated footfall records for their unit (measured consistently, documented properly, and produced by a system that distinguishes customers from staff) has a concrete basis for a rent review conversation that anecdote alone cannot provide.
A landlord in the same position can demonstrate to prospective tenants that the centre’s footfall is real, consistent, and independently verifiable, a claim that matters more to sophisticated tenants than a headline number produced by an unknown system at an unknown calibration standard.
The analytics platform that supports this needs to do more than count visitors; it needs to maintain the audit trail that makes the count credible.
The foundation the benchmark rests on
Comparing performance over time, across a portfolio, and against market indices gives a genuinely useful view of whether a location is improving or declining. The risk is not in the comparison itself but in the assumptions it requires.
The assumption most quietly elided is consistency: that what you counted this year was counted the same way as last year, and the same way as the site two hundred miles away. That assumption is only as good as the measurement behind it. Consistent method, documented definitions, calibrated hardware, a data trail that survives staff turnover and system upgrades, that is the foundation any footfall benchmark rests on. The analytics platform and the data deliverables it produces need to maintain that audit trail, not just the count itself. Whether you are comparing yesterday to last Tuesday or your best site to the market index, everything follows from that foundation, or it doesn’t follow at all.
- 1.2 billion
- Retail visits tracked annually by the index
Frequently asked questions
What is footfall benchmarking?
Footfall benchmarking compares visitor counts across locations, time periods, or competitors to establish a reference point for performance. A site that attracts more visitors than comparable sites, or more than it did the previous year, is said to outperform the benchmark. The difficulty is that meaningful comparison requires the same counting method, the same definitions of what counts as a visit, and the same approach to handling staff, service entrances, and multiple-entry visitors.
Should footfall benchmarks use absolute counts or indices?
Both have a place, but for cross-site and cross-operator comparisons, an index is often more useful than an absolute count. An index expresses each site's performance relative to a baseline, typically 100, so that sites with very different absolute volumes can be compared fairly. Absolute counts are more useful for internal planning, staffing, and revenue correlation where the actual number of visitors matters.
What makes a footfall benchmark misleading?
The most common problem is comparing counts produced by different measurement methods, a Wi-Fi-based system counting devices rather than people, a camera system that has not been calibrated against a manual control count, a sensor that does not distinguish staff from customers. Even within the same method, definitional differences compound: does the count include visitors who only enter the lobby? Are children counted? What happens to bi-directional door sensors when two people pass at once?
How do I use footfall data for lease negotiations?
Consistently measured footfall data provides an objective basis for negotiating rent and service charges. A tenant who can demonstrate that their location draws significantly lower traffic than comparable units in the centre, from validated data rather than an anecdotal impression, has a concrete argument. A landlord who can demonstrate that footfall at their property outperforms the broader market has equally strong grounds.