Research › Business Verdicts
Artul.ai Research LibraryStudy No. 2Business VerdictsUpdated 2026-08-28

Read the Room, Lower the Bar: What Happens After Guidance Gets Cut

By Artul.ai Research Group · n = 19,087 earnings calls · First published 2026-08-28
Abstract

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 elevated stress (3.41 vs 2.43) and evasion (2.82 vs 2.70), and lower confidence (6.37 vs 7.21) and promotion (4.56 vs 5.05). 'Results Worse Than Direction' narratives appear 1.53x more often, while 'Skeptic Reassured' appears 0.54x as often. Among 3,134 calls with measured post-call outcomes, the median return was -9.06% versus -7.16% for the base sample. These are descriptive patterns, not predictions.

Key findings
  • Lowered-guidance calls make up 11.56% of the corpus (95% CI 11.40%-11.71%).
  • Stress scores run 3.41 vs 2.43 on other calls, while confidence runs 6.37 vs 7.21.
  • The 'Results Worse Than Direction' narrative appears 1.53x more often; 'Skeptic Reassured' appears 0.54x as often.
  • Among 3,134 calls with measured outcomes, the median return was -9.06% versus -7.16% for the base sample.

1Introduction

Few sentences reset a call's mood faster than a guidance cut. Investors, analysts, and management all reprice expectations in real time, and the language around the cut can matter as much as the cut itself. Yet the tone of these calls is rarely measured systematically at scale. Using Artul.ai's library of 165,182 earnings calls from 1990 through 2026, we isolate the 19,087 calls where guidance was read as lowered and profile how their language, narrative patterns, and post-call outcomes differ from the rest of the corpus.

2Data & methodology

The corpus comprises 165,182 earnings-call transcripts published between 1990 and 2026, each scored independently by a large language model on an identical 37-field battery: seven categorical business verdicts, eight 0–9 behavioral meters, and twenty yes/no judgments. The study group is defined as calls where guidance was read as lowered (n = 19,087; 11.6% of the reference set, 95% Wilson interval 11.4%–11.7%). Baseline figures use all scored calls. Market outcomes join a fixed sample of 22,449 calls with twelve-month total returns in excess of SPY, measured from the first close after each call; this sample skews toward liquid U.S. names and is reported as descriptive history only.

3Results

The behavioral profile is distinctive: stress is elevated by 0.98 points (3.41 vs 2.43) and confidence falls by 0.84 (6.37 vs 7.21), while specificity barely moves (7.54 vs 7.56), suggesting management stays detailed even as mood sours. Narrative patterns shift too: 'Results Worse Than Direction' appears 1.53x more often and 'The Hidden Segment' 1.47x more often, while 'Skeptic Reassured' appears only 0.54x as often. The trend series is volatile, peaking at 14.86 in 2022 and dipping to 6.36 in 2021.

Table 1. Mean behavioral scores (0–9 scale), study group versus baseline
MeterStudy groupBaselineΔ
Candor7.166.86+0.30
Evasion2.822.70+0.13
Specificity7.547.56-0.02
Stress3.412.43+0.98
Promotion4.565.05-0.49
Confidence6.377.21-0.84
Table 2. Guidance actions, study group versus baseline
ActionStudy groupBaseline
Raised0.0%21.1%
Maintained0.0%48.8%
Lowered100.0%11.6%
Withdrawn0.0%2.7%
Table 3. Co-occurring battery signals ranked by lift (group prevalence ÷ baseline prevalence)
SignalLiftIn groupBaseline
Results Worse Than Direction1.53×78.1%51.1%
The Hidden Segment1.47×31.1%21.1%
The Question Left Hanging1.33×63.9%48.0%
Underused Fixed Costs1.32×55.0%41.6%
Skeptic Reassured0.54×36.1%66.4%
Pricing Recovering0.72×15.6%21.5%
Volume About to Step Up0.75×21.3%28.5%
201514.70%
201612.65%
20179.98%
201810.55%
201913.97%
202010.24%
20216.36%
202214.86%
202313.14%
202411.92%
202510.88%
Figure 1. Share of all analyzed calls matching the study definition, by year.
Table 4. Twelve-month excess total returns versus SPY (descriptive history, not a signal)
StatisticStudy groupReturns sample
Median excess return-9.1%-7.2%
Interquartile range-30.2% to +12.9%
Share beating SPY38.1% (95% CI 36%–40%)39.5%
Observations3,13422,449
Table 5. Most recent calls matching the study definition
TickerQuarterCall dateCall grade
CNCQ2 20252025-07-25F
FLGQ2 20252025-07-25B
MTHQ2 20252025-07-25C
TRATFQ2 20252025-07-25F
VWAGYQ2 20252025-07-25C
GRBMFQ2 20252025-07-25D
INTCQ2 20252025-07-24D
SAMQ2 20252025-07-24D

4Discussion

A careful reader should conclude that lowered-guidance calls sound measurably different: more stress, less confidence, and a different mix of recurring narratives. They should not conclude that any tone or narrative causes the cut, the outcomes, or subsequent returns. The return gap (-9.06% vs -7.16% median) is a descriptive comparison across samples, not evidence of an exploitable pattern. Year-to-year shares also reflect market conditions rather than any structural law. Treat every number here as a characterization of this corpus.

5Limitations

The fields in this study are AI-read and therefore noisy; misclassification of guidance direction and narrative labels is possible. The returns sample covers 22,449 calls with measured outcomes but skews toward liquid, well-covered names. Our own forward tests falsified directional prediction from these signals. Additionally, LLMs partially remember the history of famous stocks, which can contaminate any backtest of language-based signals against realized outcomes. See the full methodology, including the C1 pattern’s forward-test failure and the LLM-memorization finding.

Cite this study Artul.ai Research Group (2026). “Read the Room, Lower the Bar: What Happens After Guidance Gets Cut.” Artul.ai Earnings-Call Research Library, Study No. 2. https://artul.ai/research/after-guidance-is-cut-earnings-calls

Related studies

Steady As She Goes: What Happens When Guidance HoldsMargins Are Expanding, and So Is the Confidence: ReaCapex Up, Stress Down: Reading 64,183 Calls Where SpGuidance Is a Vibe: Calls Where Demand Reads as AcceBacklog Is Growing, and So Is the Confidence: EarninMargins Are Fine, Thanks for Asking: What Contractio
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.