The C1 Pattern: Clean Calls With Growing Prepayments: When Two Signals Appear Together

When an AI read 165,182 earnings-call transcripts and scored each on a battery of questions, it found 2,343 calls — about 1.4% of the corpus — where two signals appeared together: clean, candid language and growing customer prepayments. That combination matters because prepayments are one of the harder things to fake, and a call that is both honest and backed by customers putting cash down early is a pattern worth knowing how to spot.

The profile differences are stark: confidence averaged 8.03 versus 7.21 baseline, stress 1.29 versus 2.43, and evasion 2.13 versus 2.70. Guidance behavior diverged sharply too — 55.2% raised guidance versus 21.1% baseline, and only 0.8% lowered it versus 11.6%. The pattern peaked in 2021 at 2.17% of calls, then fell to roughly 1% by 2024. Among 468 calls with tracked outcomes, the median next-quarter return was 10.1% versus −7.2% for the 22,449-call baseline, with 62.8% beating versus 39.5%.

These are descriptive readings from an AI scorer, and AI-read fields are noisy — the same call can score differently across runs. The return figures are historical descriptions from a 22,449-call sample, not a trading signal; our own forward tests falsified directional prediction. Language models also partially remember famous stocks' histories, which contaminates any backtest of language-based patterns like this one.

Behavior profile

Guidance actions in this group

What travels with it

Share of calls by year

Recent examples