Which Industries Can You Actually Win?
How we helped a four-county Midwestern EDO move from wishlist thinking to a defensible set of recruitment targets.
01. Executive Summary & Client Challenge
The Client
The economic development partnership for a four-county Midwestern metro — a manufacturing region with a location quotient above 2.2, an OEM assembly anchor, a trimodal inland river port, and a chemicals-and-plastics cluster that most peer regions would envy.
The region was not losing deals because prospects had never heard of it. It was losing deals because, when a corporate site-selection team ran the numbers, something in the stack did not clear.
The Problem the Board Could Name Precisely
- More than $1.6 billion in documented lost projects over the prior cycle — every one citing site availability and readiness as the primary reason.
- The region had real assets: manufacturing depth, port infrastructure, rail access, and an established chemicals cluster.
- What it lacked was an honest, defensible answer to the question: which industries can we actually win a competitive RFP against our peer regions?
- The board had been handed target industry studies before. It wanted something that would survive contact with a prospect's CFO.
At a Glance
The Strategic Ask
The board wanted a target industry study. What it actually needed was to know which targets it could win — and which ones it should stop spending political capital on. Those are different questions, and they require a different methodology.
02. Why Most Target Industry Studies Fail
The conventional target industry study is a growth-and-concentration exercise: run location quotients, find the sectors trending up, produce a list of eight to twelve industries the region “should” pursue, and hand it over. It is analytically defensible and operationally useless — for three reasons.
It never tests cost.
A region can be perfectly concentrated in an industry it cannot win a competitive RFP for, because its power, water, or construction cost stack loses to a peer 200 miles south. Concentration is not competitiveness.
Thresholds drift.
When a screen produces an unpopular answer, the threshold quietly moves. The study then confirms what the room already wanted. This is not analysis — it's validation theater.
It is never validated.
The framework is asserted, not tested. Nobody asks whether it would have predicted the deals the region already lost. A methodology that can't explain your history can't guide your future.
We built this engagement to close all three gaps.
Every threshold locked before analysis ran. Cost tested the way a corporate consultant tests it inside a live RFP. And the methodology run backward against the region's documented losses before a single verdict was reported forward.
03. Approach: Six Phases, One Rule
The Rule
Every threshold is locked before any analysis is run, and every figure traces to a source. Sourced, derived, and pending data are distinguished throughout and never conflated. Where a number was not yet confirmed, it was carried as a flagged placeholder — never invented, never rounded into confidence it had not earned.
Regional Baseline Assessment
Labor force and commuting-shed analysis, cluster mapping, full infrastructure inventory (power, water, wastewater, rail, river, fiber), peer benchmarking against the specific out-of-state regions the client actually loses deals to, and a documented-loss ledger. This established the constraint map every later phase inherited.
Screened Industry Shortlist
198 industries, five independent screens, all thresholds locked in a registry before the first calculation ran: Location Quotient (with parallel wage-bill LQ for quality control), two-window shift-share, automation risk, wage quality tested against real inflation, and supply-chain gap analysis. Output: 12 sub-sectors across three tiers.
Operating Cost & Fatal-Flaw Analysis
The phase most studies skip. Each surviving target was modeled the way a corporate site-selection consultant models it inside a live RFP — five steps, each with a locked decision gate: prototype prospect build, fatal-flaw screen, Total Cost of Project model (Year-0 CapEx + 10-year NPV at 8.5% WACC), incentive alignment overlay, and a competitive position verdict.
Target Profiles & Corporate Personas
Prospect-grade personas for surviving targets: operational footprint, decision triggers, value propositions written to lead with the prospect's pain rather than the region's pride, dollar-quantified ecosystem asks, and target-firm shortlists. Added: a concurrency stress test — can the region actually staff all of these wins at once?
Go-to-Market Playbook
The deployable program: a 50-firm outbound register tiered by siting signal, battlecards and call scripts, collateral and web copy, a trade show and events plan, and a sequenced 13-initiative implementation roadmap with owners and lead times.
Executive Board Presentation
The decision narrative, board-ready. Not a data dump — a defensible answer to the question the board had actually been asking for years: why do we keep losing, and what specifically do we do about it?
04. The Industry Screen: 198 Down to 12
Five independent screens, all thresholds locked in a registry before the first calculation ran. The discipline of the process is what makes the output defensible — not the sophistication of any single screen.
Location Quotient
Run with a parallel wage-bill LQ as quality control — catches headcount concentrations propped up by low wages.
Two-Window Shift-Share
Pre-2020 and post-2020 windows run separately, isolating genuine local competitive effect from national trend.
Automation Risk
Conditional ceiling and hard-eliminate threshold taken from published research — not analyst judgment.
Wage Quality
Tested against real inflation. Industries with positive nominal but negative real wage growth scored Conditional, not Pass.
Supply-Chain Gap
Quantified import leakage the region is already paying for — identifying where local capacity is structurally undersupplied.
Output: Three Tiers
Cleared all five screens with robust scores. Advance to cost model.
Identified deliberately outside the LQ screen — structural assets with no existing concentration.
Conditional scores. Advance to cost model with elevated scrutiny.
A Methodological Note That Mattered
One industry's apparently extreme concentration turned out to be an artifact of data suppression in the public wage series — the figure attributable to a broader aggregate rather than the specific sub-sector. We caught it, documented it, and reclassified the target. That reconciliation is in the appendix, not buried. The integrity of the output depends on the integrity of the inputs, and that means naming the anomalies rather than smoothing them.
05. Operating Cost & Fatal-Flaw Analysis
This is the phase most studies skip. We modeled each target the way a corporate site-selection consultant models it inside a live RFP — five steps, each with a locked decision gate.
24 region-industry pairs entered this phase. 17 advanced to the full cost model. The eliminations at the fatal-flaw gate were not judgment calls — they were arithmetic.
Why Fatal Flaws Before Cost
Screening physical capacity before modeling cost means you never spend a week building a cost model for a project the region physically cannot host. The eliminations become unarguable: not “this looks expensive,” but “the water draw exceeds available capacity at the 150% headroom standard, and here is the specific utility investment that would reopen it.”
The Funnel
Prototype Prospect Build
A realistic mid-range project specified across ten dimensions — CapEx, jobs, square footage, acreage, peak MW, water and wastewater draw, logistics, permitting. Not a generic placeholder. An actual project.
Fatal-Flaw Screen
Binary pass/fail across eight infrastructure dimensions at 150% of prototype peak demand — the site-selector standard for growth contingency. Any single failure eliminates the region-industry pair. If the client's own sites fail, the industry leaves the study entirely.
Total Cost of Project Model
Year-0 construction CapEx plus the 10-year NPV of six operating components at a locked 8.5% WACC — including an explicit Year-1 workforce ramp and training cost. A real differential in a sub-3% unemployment market that most cost models silently omit.
Incentive Alignment Overlay
A four-layer stack — state, local, utility, federal — netted against gross cost. For the two industries where incentive math most drives outcomes, the stack was calibrated against 10–16 comparable real peer-region wins rather than statutory maximums. Statutory maximums are a fantasy no prospect ever receives.
Competitive Position Decision
Pass within 7.5% of the net-cost leader. Conditional at 7.5–15% with a costed path to close. Fail above 15% or on any fatal flaw. Every verdict re-run at 80% and 120% on labor, power, and construction cost. A verdict only counts as Robust if it survives the band.
06. The Three Moves That Made the Difference
Methodology is not differentiation. The execution of it is. Three specific decisions separated this study from the stack of reports already on the client's shelf.
The Threshold Registry
Every threshold — LQ floors, automation ceilings, wage floors, verdict gates, the discount rate, the headroom multiplier — was set and recorded in a control tab before analysis began, and carried unchanged through every calculation.
This is the single cheapest defense against the primary methodological risk in any target industry study: post-hoc threshold adjustment to confirm a preferred conclusion. When a Tier 1 industry failed, nobody could argue the goalposts had moved, because the goalposts were in writing, dated, and locked.
Fatal Flaws Before Cost
Screening physical capacity before modeling cost meant we never spent a week building a cost model for a project the region physically could not host. The eliminations were unarguable: not “this looks expensive,” but “the water draw exceeds available capacity at the 150% headroom standard, and here is the specific utility investment that would reopen it.”
A target list that never eliminates anything is a marketing document, not a strategy. The eliminations are what make the approvals credible.
We Tested the Framework Against History Before We Trusted It
Before reporting a single verdict forward, we ran the methodology backward against the region's four documented losses. Every one of those deals turned on site readiness — and the fatal-flaw screen is the formal, repeatable expression of exactly that variable. The model reproduced the loss mechanism rather than discovering it after the fact.
We then ran the same discipline forward in Phase 4, mapping the client's live pipeline against the persona set. Coverage came in at 41.7% — validated, not asserted.
A framework that predicts the deals you already lost is a framework you can take to a board.
07. What We Found
Three of four Tier 1 industries passed the cost screen — all three robustly. Two of them, once the incentive stack was applied, made the region the lowest net-effective-cost location in its entire peer set.
That headline is not the important part. These three findings are.
We Killed a Tier 1 Target
One of the four — a capital-intensive chemicals play the region badly wanted — failed at the fatal-flaw screen on a physical water constraint. It was never costed, because costing an industry the region cannot host is theater.
We eliminated it, documented the specific utility investment that would reopen it, and redirected the chemicals thesis into an adjacent sub-sector that fit inside existing capacity.
A target list that never eliminates anything is a marketing document, not a strategy. The elimination is not a failure — it is the most useful sentence in the report. It tells the region exactly where to invest capital if it wants to change the answer.
We Refused to Oversell the Glamour Target
The data center track cleared the cost gate — but only after engineering an abatement so aggressive it pushed the host tax base to its limit, and the worst-case sensitivity band widened the gap past the threshold.
So the verdict was Conditional, not Pass. The recommendation: pivot from hyperscale toward right-sized sub-40 MW enterprise and edge nodes, where the power penalty collapses and existing headroom (200+ MW) can absorb the load with no transmission build.
That target is carried in gold, not teal — a real opportunity, stated with its fiscal caveat in the client's own collateral rather than quietly omitted from it. Prospects find out on their own what the real constraints are. Better that they find out from you.
We Did Not Paper Over the Cost Disadvantage
The region's power rates run meaningfully above its southern peer set — a genuine structural finding, and one the client's stakeholders would have discovered on their own the first time a prospect asked.
So we said it. The defensible claim is not that power is cheap. It is that for the specific targets on this roster, power cost is not the decisive line — because the incentive framework closes the gap, and the analysis shows exactly where and by how much.
That is a claim that survives contact with a prospect's CFO. The other kind does not.
08. What the Client Received
Locked Five-Target Recruitment Roster
Ratified in working session. Not a list that requires interpretation — a roster the business development team could open on Monday and work.
Six Deliverables with Full Analytical Appendix
Every workbook, every threshold registry, every version-history entry. The client can recreate any number in the report from first principles.
50-Firm Outbound Register
Tiered by siting signal, with battlecards and call scripts — not a contact list, but a sequenced prospecting program.
13-Initiative Implementation Roadmap
Sequenced by dependency and lead time, with assigned owners. Not a recommendations slide — an execution plan.
A Defensible Board Narrative
An honest answer to the question the board had been asking for years: why do we keep losing, and what specific investment changes the answer?
Corporate Persona Set with Concurrency Stress Test
Prospect-grade personas for each surviving target, including an honest assessment of whether the region can staff all wins simultaneously.
09. Transferable Insights
The next quarter's work is execution, not study. But every engagement produces findings that extend beyond the specific client. These apply to any EDO, port authority, or certified tech park working on a targeting strategy.
A target list that never eliminates anything is a marketing document.
The eliminations are what make the approvals credible. The value of saying no to the right targets is measured in the political capital you stop spending on deals you were never going to win.
Concentration is not competitiveness.
A region can be deeply concentrated in an industry and still lose every RFP because its cost stack loses to a peer 200 miles south. Location quotient measures what you have, not whether you can win.
Thresholds set after the analysis are a different product.
Lock them before the first calculation runs. The threshold registry is the cheapest methodological safeguard available, and most studies skip it entirely.
The framework should predict your losses before you trust it with your wins.
If the methodology doesn't reproduce the mechanism behind the deals you already lost, you don't know what the methodology is actually measuring.
Honest findings are a competitive advantage in the room.
A consultant who tells you what you can't win, and why, and what it would cost to change that answer, has given you something you can actually use. The other kind gives you a list you already half-believed.
You Don't Need Another List of Industries to Pursue.
You need to know which ones you can win, which ones you should stop spending political capital on, and what specific investment would change the answer. That is the work.
Start a Conversation →Client identity withheld by agreement. Utility capacity figures generalized; precise negotiated values, absolute cost-model outputs, and named prospect firms excluded. Methodology, thresholds, and verdict logic described as executed.