Research › Hypotheses Tested
Artul.ai Research LibraryStudy No. 63Hypotheses TestedUpdated 2026-08-28

Hypothesis Tested: Two-sided intensity

By Artul.ai Research Group · n = 88 earnings calls · First published 2026-08-28

Some earnings calls read like a seesaw: strong promotion and confidence on one side, with candor and specificity on the other. This study tests whether that two-sided intensity pattern means anything. Out of 165,000 transcripts read by an AI, 88 calls (0.177% of a 496-call base group) showed the pattern. If you follow earnings calls closely, it is worth knowing whether this tonal combination marks a distinctive kind of management behavior or just a stylistic quirk with no substance behind it.

The profile shows these calls score higher on promotion (5.73 vs 5.10) and confidence (7.81 vs 7.33), with slightly higher specificity (7.72 vs 7.65) and candor (6.78 vs 6.92 essentially flat). Guidance skews positive: 30.7% raised versus 22.0% in the base, and only 5.7% lowered versus 13.3%. The pattern is heavily overrepresented among founder-led companies (1.73x lift, 36.4% vs 21.0%) and underrepresented when results are worse than direction (0.55x). The trend is spiky, peaking at 0.14 in 2022. Among 37 calls with returns data, the median was -13.0% versus -10.5% in the base.

These fields come from AI readings of transcripts and are noisy; small groups like this one amplify that noise. The return figures are descriptive history from a 22,449-call sample, not a trading signal. Our own forward tests falsified directional prediction, and language models partially remember famous stocks' history, which contaminates any backtest. Treat everything here as pattern description, not prediction.

Matching Calls
88
17.7% of corpus
12m Excess vs SPY (median)
-13.0%
sample baseline -10.5%
Beat SPY
41%
baseline 38% · n=37

Behavior profile

MeterThis groupAll calls
Candor6.78/96.92/9
Evasion2.69/92.70/9
Specificity7.72/97.65/9
Stress1.94/92.39/9
Promotion5.73/95.10/9
Confidence7.81/97.33/9

Guidance actions in this group

ActionThis groupAll calls
raised30.7%22.0%
maintained56.8%53.0%
lowered5.7%13.3%
withdrawn1.1%1.0%

What travels with it

SignalLiftIn groupOverall
Founder-Led Companies1.73×36.4%21.0%
A Tiny Fraction of the Market1.69×52.3%30.8%
Early Products Growing Fast1.61×62.5%38.7%
Results Worse Than Direction0.55×26.1%47.2%
When the CFO Dominates0.62×9.1%14.5%
The Question Left Hanging0.66×28.4%42.7%

Share of calls by year

YearShare
20150.06%
20160.05%
20170.08%
20180.06%
20190.01%
20200.00%
20210.06%
20220.14%
20230.07%
20240.02%
20250.00%

Recent examples

TickerCallDateGrade
CRGOQ1 20242024-05-20C+
WRBYQ1 20242024-05-09A
LINCQ1 20242024-05-06B+
SNVQ1 20242024-04-18B
PBRQ4 20232024-03-08D
NICEQ4 20232024-02-22B+
SBSQ3 20232023-11-10C+
DASHQ3 20232023-11-01C+

This hypothesis has its own page with every matching company: see the full question.

Cite this study Artul.ai Research Group (2026). “Hypothesis Tested: Two-sided intensity.” Artul.ai Earnings-Call Research Library, Study No. 63. https://artul.ai/research/hypothesis-two-sided-intensity

Related studies

Hypothesis Tested: Identified near-term contributorsHypothesis Tested: Uncontested runwayHypothesis Tested: Coming out of the tunnelHypothesis Tested: Answers go deeper than the scriptHypothesis Tested: Repeat expansion inside the base Hypothesis Tested: Earned edge, pressed harder
Not investment advice. Artul.ai publishes AI-generated earnings-call quality grades and expected-volatility estimates — never buy or sell recommendations. We tested over 1,600 predictive hypotheses against 165,000 transcripts; the honest result, including what failed, is documented in our methodology.