What the data says, including when it says no.

We replaced a reasoning LLM with a decision-only model that returns a probability per question and no text. Same 500 questions, one transcript: 0.4–1.7 s instead of about four minutes per 250, 150–500× per call, calibrat…

Of 500 things a model checks on every earnings call, three words-only tells stack into a crash signal: “we believe” over “we expect”, “excited” over “confident”, and a question left unanswered. All three on one call, and…

The more often management says “excited” on the call, the worse the stock does over the next year: since 2021 the median stock trailed the S&P 500 by 11 points for silent calls and 25 points for the most excited. The mar…

We counted the word on 174,989 earnings calls since 2016, then followed 22,240 of the stocks for a year. Saying it more raises the odds of beating the S&P 500, from 29% to 38%. The typical stock still loses. And the comp…

We had an AI read 165,000 earnings calls and tested 1,600 trading hypotheses against 12-month returns. One survived every control ever devised, then lost 12.4% against SPY in a real forward test. The autopsy found someth…

This study examines earnings calls where the model flagged the battery item "Critic Ammunition" with a YES, covering 145,051 of 165,182 calls (87.8%, 95% CI 87.7%–88.0%) across 1990–2026. Flagged calls skew slightly towa…

This study examines earnings calls where a language model answered YES to the battery item 'Confidence Proportionate to Evidence'. Of 165,182 calls scored from 1990 to 2026, 144,714 (87.6%, 95% CI 87.4% to 87.8%) met tha…

This study examines earnings calls where the model answered YES to the battery item "Calls That Resolve Doubts" across a corpus spanning 1990 to 2026. Of 165,182 calls, 131,335 (79.51%) met the definition. These calls sh…

We examine 165,182 earnings-call transcripts from 1990 through 2026 and isolate the 118,164 calls (71.5%, 95% CI 71.3% to 71.8%) where the model answered YES to the battery item 'Guidance Worth Underwriting'. These calls…

Artul.ai's model scored 165,182 earnings calls from 1990 to 2026 on a battery of items, including whether the analyst-skeptic voice ended up reassured. In 109,699 calls (66.41%, 95% CI 66.18%–66.64%) the model answered Y…

This study examines 84,487 earnings calls (out of 165,182, or 51.15%) where the model answered YES to the battery item 'Results Worse Than Direction', across a 1990–2026 corpus. Flagged calls show lower candor (6.86 vs 6…

We study 79,206 earnings calls (48.0% of a 165,182-call corpus spanning 1990-2026) where the model answered YES to the battery item 'The Question Left Hanging' — calls that end with a live, unresolved issue. Compared wit…

We ask what management teams reveal when they admit their fixed costs are underused. Using a model-scored battery across 165,182 earnings-call transcripts from 1990 to 2026, we isolate the 68,779 calls (41.6%) where the …

This study examines earnings calls that the model flagged as reading rehearsed, using a corpus of 165,182 calls from 1990 to 2026, of which 38.64% met the flag (95% CI 38.41%-38.88%). Rehearsed-flagged calls score lower …

This study examines 63,629 earnings calls from 1990 to 2026 in which an AI model answered YES to the battery item "Early Products Growing Fast" — 38.5% of the 165,182-call corpus (95% CI 38.3%–38.8%). These calls skew pr…

We studied earnings calls where the model answered YES to the battery item "A Tiny Fraction of the Market" — 49,623 of 165,182 calls (30.04%) across 1990–2026. These calls skew promotional: promotion scores 5.83 versus a…

This study examines 47,036 earnings calls from a 165,182-call corpus (1990-2026) in which a language model answered YES to the battery item "Volume About to Step Up" - 28.5% of all calls (95% CI 28.3%-28.7%). Compared wi…

What does it mean when management starts talking about industry consolidation? This study flags every call in the corpus (1990-2026) where the model answered yes to "Consolidation Among Peers": 42,662 of 165,182 calls, a…

We examine 35,541 earnings calls from a corpus of 165,182 spanning 1990 to 2026 in which the model answered YES to the battery item 'Pricing Recovering'. These calls skew candid and specific: candor 7.01 vs 6.86, specifi…

This study profiles 34,917 earnings calls out of 165,182 in the Artul.ai corpus (21.14%) where the model answered YES to the battery item 'The Hidden Segment', covering calls from 1990 to 2026. Compared with the base pop…

This study examines 33,092 earnings calls (out of 165,182 from 1990-2026) where the model answered YES to the battery item "Founder-Led Companies." Founder-led calls score higher on promotion (+0.61) and confidence (+0.1…

We examined 23,336 earnings calls (14.1% of a 165,182-call corpus spanning 1990-2026) where the model answered YES to the battery item "When the CFO Dominates." These calls skew toward specifics: specificity runs 7.71 vs…

This study examines 18,297 earnings calls (11.1% of a 165,182-call corpus spanning 1990 to 2026) where a model flagged claims that a company's advantages depend on operating at scale. These calls differ sharply in tone: …

This study examines 14,649 earnings calls (8.87% of a 165,182-call corpus spanning 1990-2026) where the model answered YES to the battery item "Deferred Revenue Growing." Calls in this group show a more promotional and c…

We examined 7,247 earnings calls (4.39% of a 165,182-call corpus spanning 1990-2026) where Artul.ai's model answered YES to the battery item 'The Finished-Story Tell' — calls whose narrative wraps up with suspicious poli…

This study examines 80,591 earnings calls out of a 165,182-call corpus spanning 1990-2026 where guidance was read as maintained - 48.79% of all calls (95% CI 48.55%-49.03%). Maintained-guidance calls show modestly better…

We examine 73,217 earnings calls from a 165,182-call corpus spanning 1990-2026 where the margins signal was read as expanding, a 44.3% share (95% CI 44.1%-44.6%). Speakers on these calls score higher on confidence (7.57 …

This study profiles earnings calls where capex was read as increasing, drawing on 64,183 calls (38.86% of a 165,182-call corpus spanning 1990-2026). Calls with rising capex read as more confident (7.45 vs. 7.21), more pr…

This study examines 58,672 earnings calls—35.5% of a 165,182-call corpus spanning 1990 to 2026—where demand was read as accelerating. These calls show a distinct behavioral profile: confidence runs 7.76 versus a 7.21 bas…

Across 165,182 earnings calls from 1990 to 2026, we identify the 57,249 calls (34.66%, 95% CI 34.43%-34.89%) where backlog was read as growing. Compared with the base population, these calls score higher on confidence (7…

We analyzed 55,774 earnings calls—33.8% of the 165,182-call corpus (1990-2026)—where margins were read as contracting. These calls carry a distinct vocal and verbal signature: confidence runs 6.71 versus 7.21 on the base…

We examine earnings calls where pricing language was read as raising, drawing on 165,182 calls from 1990 to 2026, of which 49,574 (30.0%, 95% CI 29.8% to 30.2%) qualify. These calls show higher confidence (7.46 vs 7.21),…

We analyzed 165,182 earnings-call transcripts from 1990 to 2026 and isolated the 44,091 calls (26.7%) where headcount was read as growing. These calls carry a distinct vocal and verbal signature: promotion language runs …

We examined 41,241 earnings calls — 25.0% of a 165,182-call corpus spanning 1990 to 2026 — in which AI-assisted reads of the transcript flagged pricing as under pressure. On the language profile, these calls show lower c…

This study examines earnings calls where guidance was read as raised, drawing on 165,182 calls from 1990 to 2026. Raised-guidance calls account for 21.05% of the corpus (95% CI 20.86%-21.25%). These calls show a distinct…

This study examines 28,971 earnings calls where headcount was read as cutting, drawn from a 165,182-call corpus spanning 1990 to 2026. These calls make up 17.5% of the corpus (95% CI 17.4% to 17.7%). Management tone shif…

This study examines 23,150 earnings calls (14.0% of a 165,182-call corpus spanning 1990-2026) where capex was read as decreasing. Compared with the rest of the corpus, these calls show lower confidence (6.71 vs 7.21), le…

We studied 19,087 earnings calls where guidance was read as lowered, drawn from 165,182 calls spanning 1990 to 2026. Such cuts make up 11.56% of the corpus (95% CI 11.40% to 11.71%). Compared with other calls, these show…

This study examines 15,797 earnings calls—9.56% of a 165,182-call corpus spanning 1990 to 2026—where demand was read as slowing. Management behavior shifts noticeably on these calls: stress language runs 3.53 versus a 2.…

This study examines 7,699 earnings calls (4.7% of a 165,182-call corpus spanning 1990 to 2026) where backlog was read as shrinking. These calls sound measurably different: stress runs 3.07 versus a 2.43 baseline, while c…

This study examines 4,393 earnings calls where guidance was read as withdrawn — about 2.7% of a 165,182-call corpus spanning 1990 to 2026. On these calls, management language shifts in a consistent direction: stress runs…

This study asks what distinguishes earnings calls that are highly specific—scoring 6 or higher on Artul.ai's 0–9 specificity meter. Of 165,182 calls from 1990 to 2026, 161,829 (97.97%) meet the bar. High-specificity call…

Does calling a company 'candid' tell you anything? Artul.ai scored 165,182 earnings-call transcripts from 1990 through 2026 on a 0–9 candor meter and studied the 158,979 calls scoring 6 or higher. The first surprise is h…

This study examines earnings calls scoring 6 or higher on a 0-9 AI-assigned confidence meter. Of 165,182 calls in a 1990-2026 corpus, 157,716 (95.5%) clear the bar, with a 95% interval of 95.4%-95.6%. High-confidence cal…

We examined 86,188 earnings calls—52.2% of a 165,182-call corpus spanning 1990 to 2026—that scored 2 or lower on a 0–9 evasion meter. These candid calls ran higher on candor (7.12 vs 6.86), specificity (7.85 vs 7.56), an…

This study examines earnings calls scoring 6 or higher on Artul.ai's 0-9 complexity meter, a group spanning 64,523 of 165,182 calls (39.1%) from 1990 to 2026. High-complexity calls sound more evasive (evasion 3.04 vs. 2.…

This study asks what distinguishes earnings calls that score 6 or higher on Artul.ai's 0-9 promotion meter, a measure of upbeat language. Of 165,182 calls spanning 1990-2026, 61,671 (37.3%) meet the threshold. High-scori…

We examine 18,704 earnings calls scoring 6 or higher on a 0-9 uncertainty meter, drawn from a corpus of 165,182 calls spanning 1990 to 2026. These high-uncertainty calls make up 11.32% of the corpus (95% CI 11.17% to 11.…

This study examines earnings calls scoring 6 or higher on a 0-9 evasion meter, identifying 3,661 such calls among 165,182 transcripts spanning 1990-2026 (2.2%, 95% CI 2.1%-2.3%). High-evasion calls average 6.23 on the ev…

This study examines earnings calls scoring 6 or higher on Artul.ai's 0-9 stress meter, a group covering just 1.75% of the 165,182-call corpus from 1990 to 2026 (n = 2,893). Stressed calls show sharply different language …

This study examines earnings calls scoring 2 or lower on Artul.ai's 0-9 complexity meter: 861 calls, roughly 0.52% of the 165,182-call corpus spanning 1990-2026. Low-complexity calls read as unusually candid (7.09 vs 6.8…

This study examines 431 earnings calls, out of a corpus of 165,182 calls spanning 1990 to 2026, that score 2 or lower on Artul.ai's 0-9 specificity meter — roughly 0.26% of all calls. These calls are strikingly quieter a…

This study examines the rarest tail of Artul.ai's candor meter: earnings calls scoring 2 or lower on the 0-9 scale. Across a corpus of 165,182 calls spanning 1990 to 2026, only 92 calls qualify, a share of 0.057% (95% CI…

We examine 145,051 earnings calls from 1990 to 2026 where Artul.ai's model answered YES to "Critic Ammunition" — material a critic could seize on. These calls make up 87.81% of the 165,182-call corpus (95% CI: 87.65%–87.…

Artul.ai's model rated 165,182 earnings calls from 1990 to 2026 on whether the skeptical analyst voice was ultimately reassured; it answered YES on 109,699 calls, a 66.4% share (95% CI 66.2% to 66.6%). This paper profile…

We asked a language model a single question about 165,182 earnings calls from 1990 to 2026: was a question left hanging? It answered YES for 79,206 calls, or 47.95% of the corpus (95% CI 47.71% to 48.19%). Calls with a h…

We tracked 63,830 earnings calls (out of 165,182) where the model answered YES to "Calls That Read Rehearsed" — 38.64% of the corpus, with a 95% CI of 38.41% to 38.88%. The share climbed from 38.04% of calls in 2015 to 4…

This study examines earnings calls where Artul.ai's model answered YES to "Volume About to Step Up" — 47,036 of 165,182 calls, or 28.5% of the corpus. These calls show a promotional tilt: promotion scores run 5.58 vs 5.0…

We examine earnings calls where Artul.ai's model answered YES to 'Founder-Led Companies', a set of 33,092 calls out of 165,182 (20.03%, 95% CI 19.84%-20.23%) spanning 1990-2026. Founder-led calls skew promotional: promot…

We studied 23,336 earnings calls out of 165,182 (14.13%, 95% CI 13.96%–14.30%) where the model answered YES to "When the CFO Dominates." CFO-dominated calls read less promotional (promotion 4.46 vs 5.05) and less confide…

We asked a language model a single yes/no question about every earnings call in a 165,182-call corpus spanning 1990 to 2026: is deferred revenue growing? It answered yes on 14,649 calls, an 8.87% share (95% CI 8.73%-9.01…

This study examines 35,111 earnings calls — 21.3% of a 165,182-call corpus spanning 1990 to 2026 — where a language model answered YES to both "Early Products Growing Fast" and "A Tiny Fraction of the Market." These call…

This study isolates earnings calls where Artul.ai's model answered YES to "Skeptic Reassured" and guidance was read as raised: 31,112 calls from a 165,182-call corpus spanning 1990-2026, or 18.8% of calls (95% CI 18.6%-1…

We study earnings calls where Artul.ai's model answered YES to Founder-Led Companies and candor scored 6/9 or higher. Of 165,182 calls from 1990-2026, 18.5% meet this definition. These calls skew toward promotion languag…

This study examines 23,140 earnings calls—14.01% of the 165,182-call Artul.ai corpus spanning 1990 to 2026—where the model answered YES to "Volume About to Step Up" and the backlog was read as growing. Compared with the …

We identify 18,659 earnings calls, or 11.3% of a 165,182-call corpus spanning 1990 to 2026, where the model answered YES to Founder-Led Companies and the promotion score reached 6/9 or higher. These calls look distinctiv…

This study examines 9,192 earnings calls (5.6% of a 165,182-call corpus spanning 1990-2026) where a language model answered YES to 'When the CFO Dominates' and margins were read as contracting. These calls are markedly s…

We examine 3,629 earnings calls (2.20% of a 165,182-call corpus spanning 1990-2026) where the model answered YES to "The Question Left Hanging" and evasion scored 6/9 or higher. These calls are behaviorally distinctive: …

This study profiles earnings calls where the model found no critic ammunition (answered NO) and saw deferred revenue growing (answered YES) — 2,343 calls out of 165,182, or 1.42%. Compared with the base population, these…

We examine a rare slice of the Artul.ai corpus of 165,182 earnings calls (1990-2026): the 657 calls (0.40%) where both evasion and stress scored 6/9 or higher. These calls read very differently from the corpus: candor av…

This study profiles the 378 earnings calls (0.23% of a 165,182-call corpus spanning 1990-2026) where a language model flagged the call as reading rehearsed AND scored specificity at 2/9 or lower. Compared with corpus bas…

We studied 992 earnings-call transcripts from 2015 to 2025 and identified the 426 calls (42.9%, 95% CI 39.9% to 46.0%) where management answered YES to the research hypothesis "Identified near-term contributors not yet i…

This study examines 474 earnings calls from 2015-2024, of which 268 (56.5%) answered YES to the research hypothesis "Uncontested runway." Compared with the base population, these calls showed higher confidence (7.62 vs 7…

This study examines 215 earnings calls from a 500-call corpus (2015-2024) that answered YES to the research hypothesis "Coming out of the tunnel" - management claims that the company is emerging from a difficult period. …

Of 1,445 earnings calls scored between 2015 and 2025, 200 (13.8%) answered yes to a simple hypothesis: the answers went deeper than the script. What do those calls look like? Their language profile leans candid and speci…

This study examines 186 earnings calls from 2015-2024 that answered YES to the hypothesis 'Repeat expansion inside the base while new logos still arrive' - the pattern of growing existing customers while continuing to si…

We study earnings calls from 2015 through 2024 in which the research hypothesis "Earned edge, pressed harder" was answered YES. Of 485 calls in the corpus, 183 (37.73%, 95% CI 33.53% to 42.13%) qualify. Relative to the b…

We examine 153 earnings calls (2015-2024) that answered YES to the research hypothesis 'Second demand front open and funded' out of a 494-call base, a 0.3097 share. These calls show a modest promotion-language lift (5.41…

This study examines earnings calls that answered YES to the hypothesis "This quarter could not have been described last quarter" — calls whose content genuinely exceeded what was knowable one quarter earlier. Of 318 call…

This study examines 130 earnings calls (out of a 371-call corpus spanning 2015 to 2024) where the research hypothesis held: Q&A flipped from problems to possibilities and management answered accordingly. These calls show…

We ask how often earnings calls feature the claim that existing customers are pulling more from the company than originally said, and what such calls look like. Of 438 calls in a 2015-2024 corpus, 126 (28.8%) answered ye…

This study examines earnings calls where management signaled simultaneous improvement across the front, middle, and back of the business. Of 495 calls from 2015 to 2024, 125 calls (25.3%, 95% CI 21.6% to 29.3%) met the c…

This study profiles earnings calls whose language answered YES to the hypothesis "Paid to expand," drawing on 498 calls from 2015 to 2024, of which 123 (about a quarter) qualified. These calls ran slightly more candid (7…

This study examines 109 earnings calls (out of 497 scored, a 21.9% share with a 95% CI of 18.5% to 25.8%) that answered YES to the hypothesis "External validators converging," drawn from calls between 2015 and 2024. Thes…

This study examines 107 earnings calls (21% of a 499-call corpus spanning 2015 to 2024) whose language was flagged for a distinctive quality: a tone of discovery, where management seems to be working out problems in publ…

This study examines 106 earnings calls (21.8% of a 487-call corpus, 2015-2024) whose transcripts were judged YES on the hypothesis 'Still getting better as they speak' — management teams that seemed to hit their stride w…

This study examines 103 earnings calls (out of a 391-call corpus, 26.3%) that answered YES to the research hypothesis 'Engine explained, runway named' — calls where management explained the growth engine and named the ru…

This study examines 101 earnings calls (out of 460, a 0.22 share) that fit the pattern "Strong facts, held-back story" — calls where management presents solid specifics while underplaying narrative promotion. These calls…

We asked which earnings calls read as 'proven engine, fuel already in the tank,' and what accompanied that verdict. Of 496 calls from 2015-2024, 100 answered YES (20.2%, CI 16.9%-23.9%). Their language leaned promotional…

This study profiles earnings calls that answer YES to the hypothesis "Substance without an audience" — substantive calls that appear to lack an audience. Of 979 calls scored from 2015 to 2025, 97 (9.9%, 95% CI 8.2%–11.9%…

This study examines 95 earnings calls (21.2% of a 429-call corpus spanning 2015-2024) that answered YES to the hypothesis 'Fresh change already executed, benefit mostly still ahead' - companies that claim transformation …

Winning business away from a rival is one of the most repeatable claims in the earnings-call repertoire — and one of the hardest to verify. Artul.ai's research library flags calls that claim such a win and state a reason…

We study earnings calls that answered YES to the hypothesis 'The feared thing keeps not happening' — calls where management describes a widely anticipated risk that persistently fails to materialize. Of 994 calls spannin…

This study profiles 92 earnings calls from 2015-2024 that answered YES to the hypothesis 'Secure base, several live doors' out of a 342-call corpus (26.9%, 95% CI 22.5%-31.8%). These calls show modestly more confident la…

Artul.ai's research library classifies earnings-call language against research hypotheses; this study profiles calls answering YES to "More where that came from" — calls that read as promising more of the same. Of 391 ca…

This study examines 88 earnings calls (17.7% of a 496-call corpus, 2015-2024) flagged by Artul.ai's research hypothesis 'Two-sided intensity' -- calls where both upbeat and downbeat signals register strongly. These calls…

This study examines earnings calls that answered YES to the hypothesis 'Running the company for a size the numbers don't show yet' — companies acting like they are bigger than current results imply. Of 497 calls analyzed…

This study examines earnings calls that answered YES to the research hypothesis 'Leaning into the storm' — management teams that confront rather than sidestep turbulent conditions. Of 382 calls in the 2015–2024 corpus, 8…

We asked a simple question of 470 earnings calls from 2015 to 2024: which ones answered YES to the hypothesis that management acted like the company is already bigger than it is? 86 calls, or 18.3% of the corpus (95% CI …

We examined 1,998 earnings-call transcripts from 2015 through 2026 for a specific signal: calls that answered YES to the research hypothesis 'senior hire recruited.' Only 84 calls, or 4.2% of the corpus, met that criteri…

This study examines earnings calls that answered YES to the research hypothesis "Compounding evidence" — management teams presenting steadily accumulating supporting facts. Of 470 calls in the 2015–2024 corpus, 84 calls …

This study examines earnings calls that answered YES to the hypothesis 'Demand has broadened across independent fronts' — 83 of 998 calls (8.3%) in the 2015-2025 corpus. Compared with the base population, these calls sho…

This study profiles 81 earnings calls (out of a 332-call base, 24.4%) that answered YES to the hypothesis 'Candid stewardship of one big open question,' drawn from calls spanning 2015 to 2024. Compared with the broader c…

We studied 74 earnings calls that answered YES to the research hypothesis "Demand off the charts" out of a 3,399-call corpus spanning 2015 to 2026. These calls read differently: confidence scores average 8.03 versus 7.3 …