Research › Hypotheses Tested
Artul.ai Research LibraryStudy No. 48Hypotheses TestedUpdated 2026-08-28

More Where That Came From: When Earnings Calls Sound Promised-Backed

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

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 calls spanning 2015–2024, 88 calls (22.5%, 95% CI 18.6%–26.9%) qualified. These calls show higher candor (7.0 vs 6.9), specificity (7.8 vs 7.61), and confidence (7.52 vs 7.3), with lower evasion (2.59 vs 2.77) and stress (2.27 vs 2.41). They raise guidance more often (34.1% vs 21.0%) and never withdraw it. Overrepresented themes include "Volume About to Step Up" (lift 1.71) and "Consolidation Among Peers" (1.39); underrepresented themes include "When the CFO Dominates" (0.42) and "The Question Left Hanging" (0.67). Post-call returns were not better than baseline: median -6.8% vs -10.7%, beat rate 42.5% vs 37.0%.

Key findings
  • 88 of 391 calls (22.5%) answered YES to "More where that came from," with a 95% CI of 18.6% to 26.9%.
  • These calls score higher on confidence (7.52 vs 7.3) and specificity (7.8 vs 7.61) and lower on evasion (2.59 vs 2.77) than the corpus.
  • Guidance was raised on 34.1% of these calls versus 21.0% corpus-wide, and withdrawn 0% versus 1.3%.
  • Post-call returns show a median of -6.8% versus -10.7% baseline and a beat rate of 42.5% (CI 28.5%–57.8%) versus 37.0%.

1Introduction

Earnings calls that promise more of the same — continued volume, continued consolidation, continued momentum — are a recognizable genre. Whether that tone signals genuine operational strength or merely well-rehearsed optimism is exactly the kind of question a systematic library of call language can address. This study profiles the 88 calls in Artul.ai's 2015–2024 corpus (391 calls total) that answered YES to the hypothesis "More where that came from." We compare their linguistic profile, guidance behavior, thematic composition, and subsequent price returns against the rest of the corpus, looking for what distinguishes the promise-more cohort.

2Data & methodology

The corpus comprises 391 earnings-call transcripts published between 2015 and 2024, 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 that answered YES to the research hypothesis "More where that came from" (n = 88; 22.5% of the reference set, 95% Wilson interval 18.6%–26.9%). Baseline figures use the set of calls on which this question was tested. 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 promise-more cohort talks differently: candor runs 7.0 versus 6.9, specificity 7.8 versus 7.61, and confidence 7.52 versus 7.3, while evasion (2.59 vs 2.77) and stress (2.27 vs 2.41) run lower. Guidance behavior leans positive — raised on 34.1% of these calls versus 21.0% corpus-wide, lowered on 11.4% versus 13.3%, withdrawn never. Thematically, "Volume About to Step Up" appears 1.71 times more often and "Consolidation Among Peers" 1.39 times; "When the CFO Dominates" (0.42) and "The Question Left Hanging" (0.67) recede. Returns, however, are unremarkable: a median of -6.8% versus -10.7% baseline, and a 42.5% beat rate (CI 28.5%–57.8%) versus 37.0% — intervals too wide to distinguish the groups. The theme's prevalence also collapsed after 2018, with yearly shares from 0.0 to 0.1 in earlier years and 0.0–0.06 afterward.

Table 1. Mean behavioral scores (0–9 scale), study group versus baseline
MeterStudy groupBaselineΔ
Candor7.006.90+0.10
Evasion2.592.77-0.18
Specificity7.807.61+0.18
Stress2.272.41-0.14
Promotion5.235.08+0.14
Confidence7.527.30+0.22
Table 2. Guidance actions, study group versus baseline
ActionStudy groupBaseline
Raised34.1%21.0%
Maintained51.1%53.5%
Lowered11.4%13.3%
Withdrawn0.0%1.3%
Table 3. Co-occurring battery signals ranked by lift (group prevalence ÷ baseline prevalence)
SignalLiftIn groupBaseline
Volume About to Step Up1.71×42.0%24.6%
Consolidation Among Peers1.39×28.4%20.5%
When the CFO Dominates0.42×6.8%16.2%
The Question Left Hanging0.67×29.5%44.2%
Calls That Read Rehearsed0.68×28.7%42.5%
20150.09%
20160.10%
20170.06%
20180.10%
20190.01%
20200.00%
20210.05%
20220.06%
20230.05%
20240.06%
20250.00%
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-6.8%-10.7%
Interquartile range-26.7% to +6.3%
Share beating SPY42.5% (95% CI 29%–58%)37.0%
Observations40154
Table 5. Most recent calls matching the study definition
TickerQuarterCall dateCall grade
MCDQ2 20242024-07-29D
ASOQ1 20242024-06-11C+
EROQ1 20242024-05-10A
FWONKQ1 20242024-05-08C+
LINCQ1 20242024-05-06B+
CTRAQ1 20242024-05-03A
CWQ1 20242024-05-02B+
TTIQ1 20242024-05-01A

4Discussion

A careful reader should conclude that calls answering YES to "More where that came from" are measurably more confident, more specific, and more guidance-forward than the average call, and that this tone was far more common in the corpus's early years. A careful reader should not conclude that this tone causes better outcomes or predicts them: the returns comparison overlaps substantially, the beat-rate confidence interval spans 28.5% to 57.8%, and association in a descriptive profile is not evidence of a trading signal. Treat these figures as a portrait of a call style, not an edge.

5Limitations

The profile scores and theme labels are produced by AI-read fields and are noisy, so small deltas should not be over-interpreted. The returns sample covers 22,449 calls and skews toward liquid names, so the 40-call returns subset may not represent the broader universe. Our own forward tests falsified directional prediction, and this study makes no predictive claim. Additionally, LLMs partially remember famous stocks' histories, contaminating any backtest; label generation and return measurement may not be fully independent. See the full methodology, including the C1 pattern’s forward-test failure and the LLM-memorization finding.

Companion page: every company matching this hypothesis is listed at the question’s own page.

Cite this study Artul.ai Research Group (2026). “More Where That Came From: When Earnings Calls Sound Promised-Backed.” Artul.ai Earnings-Call Research Library, Study No. 48. https://artul.ai/research/hypothesis-more-where-that-came-from

Related studies

The Numbers Are Fine; Everything Else Is Pending: FlRoom to Run and Nowhere to Hide: Calls With UncontesLight at the End of the Tunnel Is Often a Train: RecDon't Stick to the Script: Earnings Calls Where AnswHave Your Cake and Expand the Base Too: Repeat GrowtPressed Harder, Answered Straight: A Profile of 183
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.