How to Audit a Footfall Claim Before You List a Space

Before you put a footfall figure in a listing — your own, an owner's, or one inherited from the last agent — you should be able to test it, because your name goes next to it. Knowing how to audit a footfall claim is a short, repeatable discipline: a handful of questions any claim must answer and a red-flag list that lets you reject the ones that dodge them. This guide gives landlords, brokers and property managers that checklist, and sets out what a defensible Footfall Certificate must contain so that the number you publish survives the first tenant who checks it.

Key takeaways

• Auditing a footfall claim is mostly plain questions: where, when, how measured, what is the margin, and can I check it independently.
• A defensible certificate ties the number to a specific address and observation period, discloses method and sampling, and shows a confidence interval.
• Red flags: a single round number with no range, a headline accuracy percentage with no evidence, no method, no date, no way to verify.
• You do not need statistics to run the first pass — you need the discipline to reject claims that answer none of the questions.
• The fastest way to pass your own audit is to certify the number before you publish it.

How to audit a footfall claim: the five questions

Run any figure through these before it goes near a listing. A claim that cannot answer them is not a measurement.

1. Where, exactly? Is the number tied to this unit and this stretch of pavement, or is it a neighbourhood or postcode figure standing in for a doorway? Traffic on the high street two blocks away is not traffic past this window.
2. When? What was the observation period, and how recent is it? "10,000 a day" from an unstated week in an unstated season is not comparable to anything. A wet February and a festival Saturday are different streets.
3. How was it measured? Guess, clicker, panel model, or a direct count? Each has a different failure mode, and a claim that will not say how it was produced is hiding its failure mode.
4. What is the margin? Is there a range, a confidence interval, a stated error? An honest estimate has one. A single perfectly round number with no margin is either luck or concealment.
5. Can I check it? Is there a method I can read and a record I can verify independently — or am I being asked to take it on faith?

Most weak claims fail at question one or two. You do not need a statistics degree to notice a number that will not name its own address and date.

What a defensible Footfall Certificate must contain

The mirror image of the audit is the specification. A footfall document is defensible — meaning it withstands a counterparty's scrutiny — only if it carries all of the following:

• A specific address and observation period. The what, where and when, pinned down so nobody can move the claim later.
• Per-direction counts. Streets are lopsided; a blended figure hides which frontage actually works.
• An hourly and weekday/weekend profile. When the street fills and empties, so a tenant can match it to their trade.
• A street percentile, clearly framed as a relative rank within the city rather than an absolute grade.
• A disclosed method and sampling window. How the count was produced and how much of the period was actually observed.
• A confidence interval and a confidence label. The margin around any projection, plus the low/medium/high flag that stops a short clip being sold as a full week.
• A limitations note. The observed week's weather, single-week seasonality, any quality flags — stated, not buried.
• A public verification page. A QR-sealed record anyone can scan to confirm the certificate is genuine and unedited.

Miss the last four and the number is unauditable no matter how confident it looks. For a section-by-section reading of a compliant document, see how to read a Footfall Certificate; for why the missing-margin problem is so common, see why you should never trust an unverified footfall number.

The red-flag list

Reject — or at minimum heavily discount — any claim showing these:

• A single round number with no range. Real counts have margins; suspiciously clean numbers usually mean the margin was dropped, not that it did not exist.
• A headline accuracy percentage with nothing to inspect. "99.5% accurate" is itself an unverified claim unless you can examine the evidence behind it. A trustworthy provider shows you the method and the error bars, not a marketing figure.
• No method, no sampling, no date, no address. Any one missing is a gap; several missing is a guess in a suit.
• No independent way to verify. If the only proof is the claimant's word, it is not proof.

None of these red flags requires you to be an analyst. They require you to keep asking "and how do you know that?" until you either get a checkable answer or a certificate.

Where the statistics do matter

The plain-questions pass filters out most bad claims, but two technical points are worth understanding so you can tell a careful provider from a careless one. First, a confidence interval should widen when less was observed — a provider whose margin never changes regardless of coverage is not really computing one. Second, accuracy is better expressed as a disclosed error measure than as a single triumphant percentage; the honest metric is something like a mean error against ground truth, with its limits stated. Both are unpacked in footfall accuracy, MAPE and confidence intervals, which is maintained as shared engine-trust research rather than duplicated here.

Pass your own audit — before you publish

The neatest outcome of learning to audit claims is that you stop publishing ones that fail. If you certify the number first, then by the time a tenant runs the very checklist above, your listing answers every question: address and date on the header, per-direction counts, disclosed method and sampling, a visible confidence interval, a limitations note, and a QR seal they can scan on the spot. That is the entire design of the Footfall Certificate.

Ready to make sure your next listing passes? See the pricing, look at a sample certificate, or, if you list for a living, certify your first listing for $149.

Frequently asked questions

What should I check first? What, where and when: is the number tied to this specific unit over a stated period, or a vague neighbourhood figure with no date? A claim that will not name its address and window fails immediately.

What must a defensible certificate contain? Address and period, per-direction counts, hourly and weekday/weekend profile, street percentile, disclosed method and sampling, a confidence interval and label, a limitations note, and a public QR verification page.

What are the red flags? A single round number with no range; a headline accuracy percentage with no inspectable evidence; no method, sampling, date or address; and no independent way to verify.

Can I audit without technical skills? Yes — most of it is asking where, when, how measured, what is the margin and can I check it, then rejecting claims that dodge those questions.