Like Father, Like Firm: Founder-Led Earnings Calls, Measured at Scale
We examine earnings calls where Artul.ai's model answered YES to 'Founder-Led Companies', a set of 33,092 calls out of 165,182 (20.03%, 95% CI 19.84%-20.23%) spanning 1990-2026. Founder-led calls skew promotional: promotion scores 5.66 vs 5.05, a +0.61 gap, alongside higher confidence (7.36 vs 7.21) and slightly lower candor and specificity. In returns data (3,582 calls), the median next-day move is -0.085 vs -0.072 for the base, and 41.09% beat expectations vs 39.47% baseline. Language lifts cluster around scale claims and early-stage growth narratives.
- Founder-led calls make up 20.03% of the 165,182-call corpus (33,092 calls), with a promotion score of 5.66 vs 5.05 (+0.61).
- Guidance on founder-led calls was raised 22.73% of the time vs 21.05% for the base, and maintained 42.61% vs 48.79%.
- The strongest language lift is 'Scale-Dependent Advantage Claims' at 1.76x (19.45% vs 11.08%); 'The Finished-Story Tell' appears least often at 0.73x.
- In 3,582 founder-led calls with returns, the median move is -0.085 vs -0.072 baseline, with 41.09% beating expectations vs 39.47%.
1Introduction
Founder-led companies occupy a distinct corner of market folklore: the long-horizon owner-operator, presumed to speak with unusual candor and skin in the game. If that reputation shapes how listeners weigh management's words, it should leave fingerprints on the calls themselves. Using Artul.ai's 165,182-call library (1990-2026), we isolate the 33,092 calls flagged as founder-led and compare their language profiles, guidance behavior, thematic lifts, and post-call return distributions against everything else. The study asks a descriptive question only: do founder-led calls actually look, and trade, differently?
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 the model answered YES to "Founder-Led Companies", tracked by year (n = 33,092; 20.0% of the reference set, 95% Wilson interval 19.8%–20.2%). 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 language deltas cut against the folk wisdom: founder-led calls score higher on promotion (5.66 vs 5.05) and confidence (7.36 vs 7.21), and slightly lower on candor (6.64 vs 6.86) and specificity (7.41 vs 7.56). They raise guidance more often (22.73% vs 21.05%) and maintain it less (42.61% vs 48.79%). Theme lifts center on scale narratives ('Scale-Dependent Advantage Claims' 1.76x; 'Deferred Revenue Growing' 1.54x) while the finished-story tell is rare (0.73x). The share of founder-led calls peaked at 22.82% in 2022 and slid to 18.08% by 2025. Returns show a modestly wider distribution (Q1 -0.326, Q3 0.174) and a median of -0.085 vs -0.072.
| Meter | Study group | Baseline | Δ |
|---|---|---|---|
| Candor | 6.64 | 6.86 | -0.22 |
| Evasion | 2.79 | 2.70 | +0.09 |
| Specificity | 7.41 | 7.56 | -0.15 |
| Stress | 2.49 | 2.43 | +0.06 |
| Promotion | 5.66 | 5.05 | +0.61 |
| Confidence | 7.36 | 7.21 | +0.14 |
| Action | Study group | Baseline |
|---|---|---|
| Raised | 22.7% | 21.1% |
| Maintained | 42.6% | 48.8% |
| Lowered | 10.6% | 11.6% |
| Withdrawn | 2.4% | 2.7% |
| Signal | Lift | In group | Baseline |
|---|---|---|---|
| Scale-Dependent Advantage Claims | 1.76× | 19.4% | 11.1% |
| A Tiny Fraction of the Market | 1.60× | 48.0% | 30.0% |
| Deferred Revenue Growing | 1.54× | 13.7% | 8.9% |
| Early Products Growing Fast | 1.38× | 53.0% | 38.5% |
| Volume About to Step Up | 1.26× | 35.9% | 28.5% |
| The Finished-Story Tell | 0.73× | 3.2% | 4.4% |
| Statistic | Study group | Returns sample |
|---|---|---|
| Median excess return | -8.5% | -7.2% |
| Interquartile range | -32.6% to +17.4% | — |
| Share beating SPY | 41.1% (95% CI 39%–43%) | 39.5% |
| Observations | 3,582 | 22,449 |
| Ticker | Quarter | Call date | Call grade |
|---|---|---|---|
| KNSL | Q2 2025 | 2025-07-25 | C+ |
| LBTSF | Q2 2025 | 2025-07-25 | C+ |
| ASPS | Q2 2025 | 2025-07-25 | D |
| WZZAF | Q1 2026 | 2025-07-25 | F |
| FINW | Q2 2025 | 2025-07-25 | B |
| MLLGF | Q2 2025 | 2025-07-25 | C+ |
| LABFF | Q2 2025 | 2025-07-24 | B+ |
| SAM | Q2 2025 | 2025-07-24 | D |
4Discussion
A careful reader should treat these as descriptive associations, not verdicts. Founder-led calls in this corpus do skew more promotional and more confident, and their thematic fingerprint leans toward scale and early-growth claims - but nothing here establishes that founder leadership causes any language pattern or market outcome. The beat-rate gap (41.09% vs 39.47%) sits within a range consistent with sampling variation, and the median return difference of roughly one basis point is economically small. The right conclusion is that the founder-led register is measurably distinct, not that it is better or worse.
5Limitations
Behavioral fields are produced by an AI model and are noisy proxies for human-graded qualities. The returns sample covers 22,449 calls, skewed toward liquid, widely covered names, so results may not generalize. Our own forward tests falsified directional prediction from these signals. Additionally, LLMs partially remember famous stocks' histories, contaminating any backtest interpretation of model-labeled features. See the full methodology, including the C1 pattern’s forward-test failure and the LLM-memorization finding.