How to judge a location selection consultant's numbers before you pay for them

Choosing where to be is a research problem before it is a property problem, and the people who help with research are not the same people who help with property. This page separates them, says who pays each, and gives you a way to test whether a number somebody sold you means anything.

What to take away

The kinds of help

Commercial property broker. Knows what is available and what has moved. Paid by the landlord or the seller in most cases. Good on buildings, weaker on whether the area suits your business, because that is not what they are paid to answer.

Location or site selection consultant. Works on the question rather than the building, usually paid a fee by you. Worth it where a wrong answer is expensive: a network of sites, a heavy operation, or a business whose location is its product.

Data provider. Sells footfall, demographic or catchment data. The whole value depends on how it was collected, which is the question to ask first and hardest.

The paid names you will meet most often are Esri, whose GIS software and demographic layers sit under many site reports; Placer.ai, which sells footfall estimates built from mobile-device panels; Kalibrate, which models retail demand around a site; Buxton, which builds customer analytics for location decisions; and STDB, the Site To Do Business platform whose demographic and traffic reports are sold through commercial brokers. Each of them sells a model rather than a count, so each of them should be able to answer the five questions further down this page.

Economic development office. Municipal, county or state. Free, knowledgeable about their own area, and openly interested in you locating there. Useful and not neutral, which is easy to allow for.

Small business advisory network. Free general guidance with nothing to sell, listed among the SBA's resource partners. A reasonable second opinion when everyone else is paid on completion.

Traffic and transport consultant. For access, turning, and anything the highway authority will ask about.

Planning consultant or land use lawyer. For whether the activity is permitted, which is the question that voids all the others if the answer is no. How to approach it yourself is in zoning and permits.

The confirmation you want is written. The municipal planning or zoning office issues a zoning verification letter for a specific address, and anything beyond the by-right use of the parcel runs through a conditional use permit (also called a special use permit), a variance, or site plan review, with the certificate of occupancy as the final gate. Ask which of those applies at that address before you commission any study, because no count of passers-by fixes a use that is not permitted.

Ask how the number was made

This is the whole test for anyone selling you data.

A supplier who answers all five plainly is selling something usable. One who treats the method as proprietary is selling confidence. For a sense of how a public body describes its own counting, the Federal Highway Administration publishes Traffic Volume Trends, which puts the numbers and the way they were produced side by side. That is the standard to hold a vendor to: the method travels with the figure, or the figure is not usable.

What to do yourself, free, first

Do that first and you will know which of the paid services you actually need, which is usually fewer than you thought.

Public data will not count the people walking past a door, but it can catch a paid number that is out of range. The Census Bureau's American Community Survey gives population, income and commuting patterns down to the census tract; LEHD OnTheMap, also from the Census Bureau, shows where workers live and where jobs sit; and the Bureau of Labor Statistics' Quarterly Census of Employment and Wages gives employment by industry for a county and its neighbors. If a vendor's picture of a place disagrees sharply with all of those, the burden is on the vendor to explain why.

The scorecard

Work one figure through. Say a vendor sends a footfall estimate for a shopping street and labels the geography "the district". The first question asks how it was counted, and the answer is a mobile-device panel rather than a person with a clicker; the second asks where the district line runs, and it turns out to be a mile-wide shape the vendor drew; the third asks for the collection date. Then spend an afternoon counting at your own door at the hours your business is busy. If the two cannot be reconciled, the estimate has measured a different question than the one you asked, and it does not belong in the decision.

A one-site test

Pick one candidate address and set a deadline. Ask for three things: a footfall or catchment figure with its method, a list of applications pending within a short radius, and a written answer on whether your activity is permitted.

Then compare what comes back with what you found yourself in an afternoon. If the paid answer adds nothing to your own count and a phone call, you have learned what that service is worth on this decision.

What to get in writing before you pay

Nothing on this page works if the answers arrive only by voice. Ask for a written scope that names what will be counted or modeled, the geography, the data source and its vintage, the deliverable and the delivery date, and what the resulting figure may not be used for.

A consultant who sells on method writes that down without being pushed; one who sells on presentation will call it proprietary. Ask what happens if the method statement is not delivered as promised, because the answer tells you what the engagement is really worth. Site selection work is usually a project fee paid by you rather than a commission, which is exactly why the advice does not bend toward a particular landlord.

Where the help stops

Nobody can tell you what a street will be like in five years, and anyone who claims to is guessing with a logo on it. What you can buy is better information about the present and a complete picture of what has been applied for. The judgment is yours, and the framework for it is in location selection.

Once the address is settled, the questions change. The occupancy cost model is in commercial leasing, the case for owning instead is in commercial buying, and for a business whose trade depends on the pitch itself the counting method that matters most is in retail space.

Common questions

Is purchased footfall data worth buying?

Sometimes, for comparing many sites quickly. For a final decision between two, your own counts at the times that matter are usually better and always cheaper.

Should I talk to the economic development office?

Yes. They know their area, they will tell you what is planned, and their interest in you locating there is obvious enough to correct for.

Do I need a consultant for one site?

Rarely. Consultants earn their fee on portfolios, unusual operations, or where the cost of the wrong answer is very large.

What is the one question people forget?

Whether the activity is permitted at that specific address. Everything else on this page is wasted effort if that answer is no.