Bimecc Insights All articles
Market Analysis

The Map Is the Model: Why Site Selection Belongs to Intelligence, Not Real Estate

Bimecc Insights
The Map Is the Model: Why Site Selection Belongs to Intelligence, Not Real Estate

Photo: Md. Ahsanul Haque Nayem, CC BY-SA 4.0, via Wikimedia Commons

A manufacturer in the upper Midwest narrows an expansion decision to three metropolitan areas. The broker supplies a polished deck. The chamber of commerce supplies an incentive offer. A demographic report supplies population growth projections. The executive committee selects the site with the most enthusiastic economic development authority and the shortest drive from the chief executive's hometown.

Meanwhile, three datasets sit unread on a server: federal freight flow records showing a corridor realignment that will cut two hours from westbound transit, occupational wage data showing that the labor pool for the firm's two hardest-to-fill roles is thinning in one of the finalist metros and thickening in another, and county permit records showing that a direct competitor has already optioned land eighteen miles away.

Geographic advantage is measurable. It is simply rarely measured.

Why Location Decisions Still Default to Anecdote

The reasons are structural rather than intellectual. Site selection is episodic, so firms never build the muscle. It is cross-functional, so no single executive owns it. And it is heavily intermediated by parties, brokers and development authorities among them, whose compensation depends on a transaction closing.

That last point deserves emphasis without cynicism. Brokers are not adversaries. They are participants in a transaction with a specific structure, and their information advantage is real. The problem is that the buyer's information advantage is also real and goes unused.

The Six Layers of a Location Model

A defensible site-selection framework combines six data layers, weighted to the specific decision. No layer is novel. The discipline lies in combining them into a single scoring model that can be inspected, challenged, and refreshed.

One: Labor Is a Market With a Price

Start with the roles that will actually be staffed, not with the county's total workforce. Occupational employment statistics, wage percentiles by metropolitan statistical area, commuting patterns, and the presence of adjacent employers who compete for the same skills will tell you more than any headline employment figure. A metro with 1.2 million people and no bench for your maintenance technicians is smaller than it looks.

Two: Freight Corridors and the Cost of Distance

Transportation is not a single number. Model inbound and outbound separately, by mode, and include the projected changes. Corridor improvements, intermodal capacity, port throughput trends, and Class I rail service levels all move over a five-year horizon that outlasts most leases. A site that is adequate today can be structurally expensive in 2031.

Three: Customer Density Against Competitor Saturation

A catchment analysis that counts customers without counting competitors describes a market that does not exist. Overlay your target customer distribution with the physical footprint of competing and adjacent providers, drawn from permits, licensing records, and commercial property filings. Penetration-adjusted density is a better predictor of first-year revenue than raw population growth.

Four: Incentives, Read Through the Clawback

An incentive package is a contract, not a gift. Discount the headline figure to net present value, then read the performance conditions: job creation thresholds, wage floors, capital investment minimums, recapture provisions, and the political durability of the authorizing jurisdiction. A ten-year abatement with an aggressive clawback and a two-year budget cycle is worth less than its stated value.

Five: Climate and Physical Exposure as a Balance Sheet Item

Flood plains, wildfire risk, heat stress on equipment and workforce, water availability, and grid reliability belong in the model as cost and continuity variables, not as sustainability language. Insurers have already repriced this risk. A site decision that ignores it is deferring a cost to a future executive committee that will not thank you.

Six: Competitor Footprint From Public Records

Competitor location data is abundant and underused. Business registrations, building permits, environmental filings, and licensing databases frequently reveal expansion before any announcement does. A quarterly refresh of this layer is one of the highest-yield intelligence activities available to a mid-market firm, and it costs almost nothing.

Building a Repeatable Framework

Begin with the decision itself: what the facility must accomplish, what volume it must serve, and what failure looks like. Then weight the six layers. A distribution node weights freight heavily. A technical service center weights labor. A customer-facing site weights density. Weights should be set before the sites are chosen, not adjusted after, which is the single most common analytical failure in this work.

Build a composite index with the weights visible. Run scenarios: what happens to the ranking if wage growth runs two points hotter, if the corridor slips a year, if a competitor opens within thirty miles. Record the assumptions in a decision log so the model can be audited when the outcome is known.

Finally, monitor after the decision. Refresh the layers annually and compare predicted with realized performance. Most firms never close this loop, which is why their second expansion is no better informed than their first.

From Transaction to Discipline

Location strategy fails as a real estate function because real estate functions close transactions. It succeeds as an intelligence function because intelligence functions maintain models. The firms that out-expand their rivals over the next several years will not be the ones with better brokers. They will be the ones whose geography is governed, refreshed, and argued about in the same room where capital allocation happens.

All Articles

Related Articles

Ninety Days to a Working Intelligence Function: A Week-by-Week Build for Mid-Market Leaders

Ninety Days to a Working Intelligence Function: A Week-by-Week Build for Mid-Market Leaders

Acquired Intelligence: How PE-Backed Consolidators Are Turning Customer Data Into a Competitive Weapon

Acquired Intelligence: How PE-Backed Consolidators Are Turning Customer Data Into a Competitive Weapon

The Revenue Concentration Blind Spot: Measuring the Customer Dependency Risk That Threatens Enterprise Valuation

The Revenue Concentration Blind Spot: Measuring the Customer Dependency Risk That Threatens Enterprise Valuation