Original research
Original research, downloadable aggregate datasets, and drawing-review benchmarks with disclosed methodology and limitations.
497
published projects
120
aggregate benchmarks
15
issue topics
8
discipline groups
Published collections
Each collection documents its source population, filtering rules, measurement definitions, and limitations alongside the results.
Project-level findings, assigned severity, normalized disciplines, project types, and recurring issue-title patterns from the published case-study corpus.
Topic-by-discipline cohort comparisons with aggregate project counts, finding counts, assigned severity, methodology, and limitations.
Topic summaries
Each topic summary compares the same disclosed title-term group across eight discipline cohorts, with key findings, severity, FAQs, methodology, and links to the underlying benchmark pages.
35,077 aggregate finding records across eight discipline cohorts.
View topic16,367 aggregate finding records across eight discipline cohorts.
View topic27,655 aggregate finding records across eight discipline cohorts.
View topic14,002 aggregate finding records across eight discipline cohorts.
View topic5,539 aggregate finding records across eight discipline cohorts.
View topic13,838 aggregate finding records across eight discipline cohorts.
View topic19,327 aggregate finding records across eight discipline cohorts.
View topic18,216 aggregate finding records across eight discipline cohorts.
View topic8,291 aggregate finding records across eight discipline cohorts.
View topic11,765 aggregate finding records across eight discipline cohorts.
View topic7,345 aggregate finding records across eight discipline cohorts.
View topic2,711 aggregate finding records across eight discipline cohorts.
View topic4,878 aggregate finding records across eight discipline cohorts.
View topic2,757 aggregate finding records across eight discipline cohorts.
View topic4,006 aggregate finding records across eight discipline cohorts.
View topicThe research distinguishes platform totals, project-level counts, and structured public records. Each report discloses which sample supports each metric so larger context numbers are not mistaken for downloadable row-level data.
Public downloads contain aggregate fields only. They exclude customer names, project names, addresses, drawings, account identities, project identifiers, and finding text.
The library includes the Construction Drawing Error Index, 120 aggregate issue benchmarks across 15 topics and 8 discipline groups, and machine-readable CSV and JSON downloads.
Yes. Both research collections provide free aggregate CSV and JSON files with their source window, methodology, and limitations documented on the corresponding report page.
No. The aggregate downloads exclude customer names, project names, addresses, drawings, account identities, project identifiers, and finding text.
No. The counts describe potential findings surfaced during InspectMind reviews. Categories can overlap, and project scope, document size, design stage, and reviewer validation vary, so the data must not be treated as prevalence, defect rates, or discipline quality rankings.
The published benchmark source window runs from 2026-03-21 through 2026-09-04. Each report explains its own source population and filtering rules.
Run an AI-assisted plan review, then have the responsible project team validate every potential finding.
5+ issues or full refund · No demo required
Sample report: 282 issues found|Pricing after first check
227,181+ customer-visible issues across 2,000+ customer accounts
