Research › Hypotheses Tested
Artul.ai Research LibraryStudy No. 39Hypotheses TestedUpdated 2026-08-28

Off the Charts, On the Record: The Language of Demand-Blown Calls

By Artul.ai Research Group · n = 74 earnings calls · First published 2026-08-28
Abstract

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 baseline, promotion 5.77 versus 5.13, and stress 2.16 versus 2.38. Guidance behavior diverges sharply: 47.3% raised guidance and none withdrew it. Topic lifts show "Deferred Revenue Growing" at 2.71x and "Volume About to Step Up" at 1.93x, while "The Hidden Segment" appears at only 0.39x. The pattern spiked to 14% of calls in 2021 before fading to 2% by 2024.

Key findings
  • The 74 demand-surge calls show confidence of 8.03 versus a 7.3 baseline, a +0.72 delta.
  • Guidance was raised on 47.3% of these calls versus 24.0% of the baseline corpus, and never withdrawn.
  • The topic "Deferred Revenue Growing" appears 2.71x more often than in the baseline; "Volume About to Step Up" lifts 1.93x.
  • Underweighted topics include "The Hidden Segment" at 0.39x and "When the CFO Dominates" at 0.58x.

1Introduction

Earnings calls are where executives choose their words under pressure, and few choices are as loud as the ones surrounding a genuine demand surge. When a management team believes demand is truly off the charts, the language tends to shift: more confident tone, more promotion, less stress, and guidance that gets raised rather than hedged. For anyone who dissects calls for a living, knowing what those surges actually look like on the page is useful context. This study examines the 74 calls out of 3,399 (2015-2026) that answered YES to the hypothesis "Demand off the charts", profiling their tone, guidance behavior, topic lifts, and frequency over time.

2Data & methodology

The corpus comprises 3,399 earnings-call transcripts published between 2015 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 that answered YES to the research hypothesis "Demand off the charts" (n = 74; 2.2% of the reference set, 95% Wilson interval 1.7%–2.7%). 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 tone profile is distinct across every measured dimension: confidence 8.03 vs 7.3, promotion 5.77 vs 5.13, specificity 7.84 vs 7.62, and candor 6.99 vs 6.88, while evasion (2.47 vs 2.68) and stress (2.16 vs 2.38) run below baseline. Guidance skews positive: 47.3% raised versus 24.0% baseline, 40.5% maintained, 5.4% lowered, and 0% withdrawn. Topic lifts concentrate on concrete demand signals: "Deferred Revenue Growing" at 2.71x, "Volume About to Step Up" at 1.93x, and "Early Products Growing Fast" at 1.69x. The annual share peaked at 14% of calls in 2021 and 13% in 2022, then fell to 2% in 2024 and 2025.

Table 1. Mean behavioral scores (0–9 scale), study group versus baseline
MeterStudy groupBaselineΔ
Candor6.996.88+0.10
Evasion2.472.68-0.21
Specificity7.847.62+0.22
Stress2.162.38-0.22
Promotion5.775.13+0.64
Confidence8.037.30+0.72
Table 2. Guidance actions, study group versus baseline
ActionStudy groupBaseline
Raised47.3%24.0%
Maintained40.5%50.0%
Lowered5.4%12.0%
Withdrawn0.0%1.2%
Table 3. Co-occurring battery signals ranked by lift (group prevalence ÷ baseline prevalence)
SignalLiftIn groupBaseline
Deferred Revenue Growing2.71×25.7%9.5%
Volume About to Step Up1.93×54.1%28.1%
Early Products Growing Fast1.69×67.6%40.0%
Founder-Led Companies1.49×32.4%21.8%
The Hidden Segment0.39×8.1%21.0%
When the CFO Dominates0.58×8.1%13.9%
Results Worse Than Direction0.59×28.4%48.1%
Calls That Read Rehearsed0.72×27.0%37.6%
The Question Left Hanging0.72×32.4%44.8%
20150.03%
20160.01%
20170.01%
20180.02%
20190.01%
20200.01%
20210.14%
20220.13%
20230.05%
20240.02%
20250.02%
Figure 1. Share of all analyzed calls matching the study definition, by year.
Table 4. Most recent calls matching the study definition
TickerQuarterCall dateCall grade
VIKQ1 20252025-05-20B+
EXCQ3 20242024-10-30B
FTAIQ2 20242024-07-24A
ACMQ2 20242024-05-07B
FWONAQ3 20232023-11-03C+
PLTRQ3 20232023-11-02B
MACQ3 20232023-10-31D
LLYQ2 20232023-08-08B+

4Discussion

A careful reader should conclude that calls flagged as showing off-the-charts demand co-occur with a more confident, more promotional tone, more raised guidance, and topics about deferred revenue and volume. That is a description of correlation in the data, not evidence that the language causes anything or that these stocks outperform. The 2021-2022 spike likely reflects broader market conditions rather than anything unique to these companies. Tone deltas of 0.6 to 0.7 points are meaningful in aggregate but say little about any single call.

5Limitations

The profile scores and topic labels are AI-read and inherently noisy, so individual classifications can misfire. Any returns discussion would rest on a 22,449-call sample skewed toward liquid names, and our own forward tests falsified directional prediction outright. Additionally, LLMs partially remember famous stocks' histories, contaminating any backtest that mixes model knowledge with call text. Treat this study as descriptive context, not a signal. 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). “Off the Charts, On the Record: The Language of Demand-Blown Calls.” Artul.ai Earnings-Call Research Library, Study No. 39. https://artul.ai/research/hypothesis-demand-off-the-charts

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.