QWI R2026Q1 · t0=2021, t1=2024 Industry-structure peers from public QWI · outcomes layered on after · Methods →
Window applies on Update · quadrant, peers, map & divergence re-window; leaderboard & Industry Detail stay on the pipeline window
Concentration × growth quadrant
Opportunity leaderboard
Structural peers are the metros whose industry portfolios most resemble this one — measured by cosine similarity on standardized 3-digit-NAICS concentration (LQ) vectors, with rarer industries weighted up (TF-IDF). Peers are chosen on structure only; economic outcomes play no role in selection, so outcome comparisons against peers stay honest. Rank 1 is the closest match. Persistence tiers ask how durable each peer relationship is: recompute the peer set year by year and count how often a metro stays in the top 20. Core peers (≥ half of years) reflect stable structural kinship; episodic and rare peers drift in and out as portfolios evolve.
Structural peers
Persistence tiers
Industries: metro vs peer-median Δshare
Peer map
Metro divergence: vs national vs peers
Each point is a metro. X-axis: how much that metro's share of this industry changed relative to the national trajectory (right of 0 = beating the national trend). Y-axis: the same change measured against the metro's own structural peers (above 0 = beating the metros most like it). Upper-right metros are outperforming on both yardsticks; lower-left are lagging both. Points sized by data suppression — larger points carry noisier underlying QWI data, so read them with more caution. This shows divergence, not causation: it says where performance differs, not why.
Top metros by Opportunity
Why this tab exists. The Census Bureau suppresses QWI cells when employment counts are small enough that publishing them could risk identifying individual employers. Suppression isn't random — it hits smaller metros and thinner industries hardest — so two metros with identical published numbers can differ a lot in how much of their data is actually observed versus inferred. Figures here reflect the pipeline's full current-vintage window and do not re-window with the analysis-period selector — treat them as a metro-level data-quality constant. This table reports each metro's suppression burden so you can calibrate confidence: a high-suppression metro's opportunity scores, peer gaps, and growth figures rest on noisier foundations. We surface this rather than hide it — transparency about data quality is part of the method.
QWI suppression burden by metro
Click a row to make that metro the anchor on the other tabs.
About this tool

MEMO

Re: The Regional Opportunity Scan — what it does, how it works, and how to use it

Data vintage: QWI R2026Q1 · t0=2021, t1=2024

Purpose

Every region asks some version of the same question: how are we actually doing? The honest answer depends on the comparison. Measured against the national economy, a manufacturing-heavy metro will look like it is lagging whenever manufacturing lags nationally — that is composition, not performance. This tool supplies a fairer yardstick: it compares each metro against the metros that are structurally most like it, alongside the conventional national and all-metro benchmarks, so users can see which parts of a region’s trajectory reflect its industrial inheritance and which parts are genuinely distinctive.

Method, briefly

All figures derive from the Census Bureau’s Quarterly Workforce Indicators (QWI), a public dataset built from linked employer–employee records. Each metro’s economy is summarized as a portfolio of about one hundred 3-digit NAICS industries, expressed as location quotients (concentration relative to the nation). Its structural peers are the twenty metros with the most similar portfolios, found by cosine similarity on standardized, rarity-weighted industry vectors. Two design choices matter most. First, peers are selected on structure only — no outcome data enters the peer definition, so outcome comparisons against peers are not circular. Second, every industry is scored against three benchmarks at once: the national trend, the median metro, and the peer median. Where the three disagree is usually where the interesting story is. The composite Opportunity score bundles undersupply, momentum, and relative gaps into a single 0–100 heuristic for scanning; its components are always shown so no one has to take the bundle on faith. Peer relationships are also recomputed year by year to distinguish durable structural kinship from transient similarity.

How to use it

Start on the Leaderboard: pick your metro, orient with the quadrant chart (your biggest industries, their local concentration, and how the nation is treating them), then scan the table for industries where the metro diverges from its benchmarks. Move to Peer Comparison and the Peer Map to see who the metro’s peers are and how durable those relationships have been — the peer list itself is often the most useful output, since it tells you which regions’ playbooks are worth studying. Use Industry Detail to flip the lens and see one industry across all metros. Check Data Quality before leaning hard on any figure for a smaller metro: QWI suppression is uneven, and this tab tells you how much of a metro’s data is actually observed.

What this tool will not tell you

The scan shows divergence, not causation. It identifies where a metro’s trajectory departs from comparable places; it cannot say whether policy, geography, luck, or a single large employer explains the gap. It also does not prescribe: an “opportunity” here means a statistical pattern worth investigating, not a recommendation to recruit an industry. Treat every flag as the beginning of a question, not the end of one.

Data notes

Metro boundaries use the Census Bureau’s 2023-vintage CBSA delineations, applied uniformly across all years shown (QWI publishes history on current delineations), so comparisons across long windows describe today’s metro footprints, not the boundaries in force at the time — early-2000s comparisons deserve that caveat. Metros that span multiple states are grouped into a single metro observation: QWI publishes each state’s portion separately, and the pipeline combines the portions so a metro like Cincinnati (OH-KY-IN) appears once. Data suppression figures reflect the full current pipeline window regardless of any analysis window selected in the header. When a custom analysis window is active, peer sets are recomputed from industry portfolios at the selected t0 and are labeled as such; the Opportunity leaderboard remains on the pipeline’s canonical window.

Where the detail lives

The methodology — peer construction and validation, benchmark definitions, winsorization, sensitivity checks, and the full limitations discussion — is documented in the accompanying papers, which are the citable references for any figure produced here:

Questions, corrections, or requests for the underlying peer sets: use the CSV download on the Peer Comparison tab, which embeds the data vintage for reproducibility.