Coverage, freshness, and limits

This page states the boundaries. It exists because the expensive failure in a diligence tool is not a missing number, it is a related number presented as if it answered the question.

Metrics the metro index does not measure

These are common requests. This table describes only what the currently connected metro-index reads report; it does not claim that every public or paid provider lacks the data. When the index marks a metric not measured, it gives the reason rather than substituting a proxy. A report linked to a deal may supply a separate figure for its own geography and period.

Market reports fill part of this gap. When you link a broker market report to a deal, EQUIRE extracts the figures it publishes, including vacancy, availability, absorption, inventory, deliveries, space under construction, sublease space, rents, and concessions. Each one stays labelled by its publisher, geography, and period, and appears beside the index rather than replacing it. Concessions are carried as the report's own wording, not as a number.

The research view reads the producer's unmeasured-metric list from its current contract at runtime. This page describes the index limits, not the availability of every external source.

Measurements that were retired

Three families of columns were withdrawn after they were found to mean something other than their names suggested. None of them is quoted as a market figure anywhere in the product, and the metro-index read strips them before the assistant sees the row.

A vacancy column that was not vacancy. It held 100 minus physical occupancy at securitization. Read as market vacancy it was wrong twice over: wrong measurement, and wrong date. It read as exactly zero on a large share of rows. Its metro-ranking companion ranked markets on the same column and went with it.

Blended cap rate and price per square foot. These pooled valuations struck at securitization with broker asking quotes into one distribution without publishing the mix, so a gap between two metros could be a composition artifact rather than a market difference. Each is replaced by a stated-basis pair: an appraised-at-securitization figure and an asking figure, each carrying its own sample count.

An occupancy column under an ambiguous name. It was identical to the occupancy-at-securitization figure but did not say so in its name. The honestly named twin is used instead, and it ships a sample count.

Sample sizes travel with the figure

A median over three observations and a median over three hundred are not the same claim, so EQUIRE reports the count alongside the figure rather than presenting bare medians.

Two thresholds are enforced in code. A median is qualified only at three or more observations. Quartiles are withheld below five, so the Market research distributions that would otherwise show a range show the median alone and disclose the count behind it. A median published below the qualifying count is labelled preliminary rather than presented as a settled figure.

Geography

Coverage is source-specific, not one shared geography set. Economic-indicator and CRE metro-index reads are keyed to a CBSA; an asking-listing summary may return submarket, CBSA, state, or national evidence. Availability also varies by metric, property type, and source refresh, so this page does not publish a single metro count. Read each figure's returned geography, source, and observation period.

A linked market report can contribute submarket evidence, labelled by its own publisher, grain, and period. Some listing summaries may also be submarket-level; check the geography attached to each result.

Freshness

Three different dates attach to any figure, and conflating them is the common error:

• The observation date. When the underlying thing was measured or filed. This is the one that matters for underwriting.
• The index refresh time. When EQUIRE last rebuilt the joined metro index. A freshly refreshed index can still carry old observations.
• The research run time. When you last ran Market research on the deal.

Market research shows these separately in Sources & basis. Do not read a recent refresh as recent data.

Vintages differ by series and are not forced onto a common calendar. Read each figure's observation date and source period; a dataset refresh does not make an older observation current.

When a read fails

A failed read is an unresolved comparison, not adverse evidence. If an index read fails, Market research says the comparison could not be resolved and keeps the comparisons that did resolve. It does not fall back to a different basis to fill the gap, and it does not report a gap as a finding against the deal.

Related

• CRE market evidence for what each basis measures.
• Economic and demographic indicators for the metro layer and its computed scores.
• Review and release for how preliminary versus released deal data is disclosed.