What it measures
The Boardroom Index is the historical percentile of three lenses on what public-company management is communicating, extracted directly from earnings call transcripts: average sentiment across the cohort, average confidence (distinct from sentiment — captures hedging), and the net revenue guidance direction (% companies raising minus % lowering). High readings mean management is collectively confident and forward-leaning. Low readings mean management is collectively cautious or hedging.
Why these three
Sentiment captures tone. Confidence captures conviction — companies often say optimistic things while hedging every forward statement. Revenue guidance direction captures the financial commitment. A high boardroom reading requires all three: positive tone, firm commitment, and net upward revenue revisions. Any single dimension can mislead — the composite makes management coherent or makes the inconsistency legible.
How to read it
0-25 (CAUTIOUS): cohort-wide hedging, lowered guidance, defensive tone. The corporate sector is bracing. 25-50 (WATCHFUL): mixed signals. 50-75 (OPTIMISTIC): meaningfully positive across the cohort. 75-100 (EUPHORIC): rare. Past Boardroom highs in late 2020 (post-COVID stimulus) and 2021 (peak AI/SPAC enthusiasm) preceded re-ratings. A high reading is not a buy signal — it is a read on how much complacency is priced into corporate communication.
What it does not say
Management is biased. Optimism on earnings calls is the baseline; lowered guidance is a meaningful negative signal but the population of recent transcripts is small (60-100 per quarter for tracked cohort). The index becomes more stable as the cohort expands beyond megacap. Treat as a regime read on corporate posture, not a tactical signal.
History note
History recomputed 2026-09-21 on the rebuilt five-year Discourse corpus under the canonical discourse.* definition: readings before 2026-09-19 are weekly as-of recomputes that replace the history originally computed on the pre-rebuild corpus. On 2026-09-21 the scale moved from linear (clipped at a raw composite of ±2) to the curved sigmoid scale the other Discourse indices use, because under the recomputed history the raw composite exceeded the linear cap and the index read 100 flat.