A Rising Tide, But From Many Boats: Calls Where Demand Broadened
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 show lower evasion (2.36 vs 2.67), lower stress (1.93 vs 2.35), higher specificity (7.88 vs 7.65), higher promotion (5.59 vs 5.08), and higher confidence (7.81 vs 7.36). They raised guidance far more often (43.4% vs 23.5%) and maintained it less often (39.8% vs 51.7%). The strongest overrepresented theme was 'Early Products Growing Fast' at a 1.97x lift. The share of such calls peaked at 0.12 in 2021 and fell to 0.0 by 2025.
- Only 83 of 998 calls (8.3%, 95% CI 6.8%-10.2%) affirmed that demand broadened across independent fronts.
- These calls raised guidance 43.4% of the time versus 23.5% in the base population, and never withdrew it.
- Promotion rose the most among tone deltas (+0.52), followed by confidence (+0.45), while stress fell (-0.42).
- 'Early Products Growing Fast' was the most overrepresented theme at a 1.97x lift, appearing in 78.3% of these calls.
1Introduction
Most earnings calls describe demand as a single engine: one flagship product, one customer segment, one macro tailwind. Rarer is the management team that can point to several independent fronts growing at once — new products, new geographies, new pricing. That kind of breadth is a distinctive claim, and how often it appears, and in what kind of call, tells us something about how companies talk when they believe the story is bigger than any one line item. This study isolates the 83 calls out of 998 (2015-2025) that affirmed the hypothesis 'Demand has broadened across independent fronts' and profiles their tone, guidance behavior, themes, and frequency over time.
2Data & methodology
The corpus comprises 998 earnings-call transcripts published between 2015 and 2025, 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 has broadened across independent fronts" (n = 83; 8.3% of the reference set, 95% Wilson interval 6.8%–10.2%). 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 of these 83 calls is distinctive: evasion (2.36 vs 2.67) and stress (1.93 vs 2.35) both run below the base population, while promotion (+0.52) and confidence (+0.45) run above it — with specificity also higher (7.88 vs 7.65), suggesting the promotional tone is anchored in detail. Guidance behavior is striking: 43.4% raised guidance versus 23.5% of the base, 39.8% maintained it, 10.8% lowered it, and none withdrew it. Thematically, 'Early Products Growing Fast' (1.97x lift), 'Deferred Revenue Growing' (1.96x), and 'Founder-Led Companies' (1.96x) dominate. Over time the share is volatile: 0.03 in 2015, peaking at 0.12 in 2021, then 0.07 in 2022, 0.09 in 2023, and 0.0 in both 2020 and 2025.
| Meter | Study group | Baseline | Δ |
|---|---|---|---|
| Candor | 6.87 | 6.93 | -0.07 |
| Evasion | 2.36 | 2.67 | -0.30 |
| Specificity | 7.88 | 7.65 | +0.23 |
| Stress | 1.93 | 2.35 | -0.42 |
| Promotion | 5.59 | 5.08 | +0.52 |
| Confidence | 7.81 | 7.36 | +0.45 |
| Action | Study group | Baseline |
|---|---|---|
| Raised | 43.4% | 23.5% |
| Maintained | 39.8% | 51.7% |
| Lowered | 10.8% | 12.7% |
| Withdrawn | 0.0% | 1.1% |
| Signal | Lift | In group | Baseline |
|---|---|---|---|
| Early Products Growing Fast | 1.97× | 78.3% | 39.8% |
| Deferred Revenue Growing | 1.96× | 16.9% | 8.6% |
| Founder-Led Companies | 1.96× | 41.0% | 20.9% |
| A Tiny Fraction of the Market | 1.75× | 54.2% | 31.0% |
| Pricing Recovering | 1.46× | 27.7% | 19.0% |
| Ticker | Quarter | Call date | Call grade |
|---|---|---|---|
| WRBY | Q1 2024 | 2024-05-09 | A |
| ASB | Q1 2024 | 2024-04-25 | A |
| E | Q1 2024 | 2024-04-24 | B+ |
| GIII | Q4 2024 | 2024-03-14 | C |
| PSFE | Q4 2023 | 2024-03-07 | C+ |
| FRPH | Q4 2023 | 2024-03-07 | C+ |
| GTLB | Q4 2024 | 2024-03-04 | C+ |
| INTR | Q4 2023 | 2024-02-08 | B+ |
4Discussion
A careful reader should conclude that calls claiming broad-based demand look measurably different: more confident, more specific, less stressed, and far more likely to raise guidance. They should not conclude that this language causes better outcomes or predicts returns — the association is descriptive, drawn from a small sample of 83 calls. The 2021 peak and 2025 low track the macro cycle but cannot be read as a signal. Breadth of demand, as coded here, is a claim made by management; it is a profile of how such calls sound, not a verdict on whether the claim is true.
5Limitations
All fields are AI-read from call transcripts and inherit transcription, parsing, and labeling noise. The returns sample covers 22,449 calls and skews toward liquid names, limiting generalizability. Our own forward tests falsified directional prediction from these signals, so nothing here should be treated as an edge. Finally, LLMs partially remember famous stocks' histories, which contaminates any backtest; apparent patterns may partly reflect memorized narratives rather than genuine inference from the calls themselves. 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.