Analysts open their earnings call questions with "congrats on the quarter" so often that it sounds like a reflex. I wanted to know whether it carries any information, so I read the analyst questions on 59,306 earnings calls held by 3,562 US-listed companies from 2020 to 2025, sorted every congratulation by what it was for, and lined the calls up against the stock's move against SPY.
The short answer
| Analysts who praised the results | Calls | Median return vs SPY, to the next day's close | Beat SPY |
|---|---|---|---|
| None (no congratulations at all) | 35,833 | -0.7% | 44.3% |
| One | 11,199 | +1.0% | 56.7% |
| Two | 4,206 | +1.7% | 60.4% |
| Three or more | 2,961 | +2.6% | 64.6% |
The pleasantry is informative, up to a point. When several analysts praise the results, the stock has usually moved the same way. It is not a forecast.
Five findings
1. On most calls, nobody says it. On 60.4% of calls not one analyst congratulated anybody. Counting each analyst once per call, 12.7% of the 351,985 analysts congratulated someone at the start of a question. "Congrats" (24,797 analysts) beat "congratulations" (20,083).
2. Not every congratulation is about the quarter. Many are about a promotion, a retirement or a deal, and they cluster. On Etsy's call on October 29, 2025, six of seven analysts opened with congratulations, only one of them about the business (the rest were about the leadership transition or for "everyone"), and the stock fell 16.1% against SPY by the next day's close. When the only congratulations were for a new role or retirement, the stock beat SPY 47.0% of the time, about the same as no congratulations at all.
3. It holds across sizes and years. Split into small, mid and large companies by trading volume, calls where two or more analysts praised the results beat SPY 61.2%, 61.4% and 63.4% of the time, against 43.7%, 44.8% and 44.3% with none. It held in every year from 2020 to 2025.
4. It's a verdict, not foresight. By the time the Q&A starts, analysts have read the same release the market is trading on. Over the next 60 sessions the gap mostly closed:
| Analysts who praised the results | Median return vs SPY, next 60 sessions | Beat SPY |
|---|---|---|
| None | -2.6% | 43.2% |
| One | -2.3% | 44.5% |
| Two | -1.2% | 47.1% |
| Three or more | -1.4% | 47.5% |
That small edge is in line with post-earnings drift, and with Milian and Smith's earlier study of analyst praise on calls from 2003 to 2013. A 47.5% hit rate is not something to trade on.
5. "Great question" carries nothing. An executive opened an answer with "great question" or "good question" on 38.9% of calls. Those stocks beat SPY by the next day's close 49.5% of the time; on calls without it, 49.1%.
The traps I hit
- Classify the congratulation, not the call. I took the sentence containing "congrat" in the first 400 characters of an analyst turn and sorted it into people, results, deals or not specified. A sentence that mentions both a person and the results counts as people. A random sample of 25 sorted as results were all about the quarter, results or progress; 20 of 25 sorted as people were about a person.
- Use a window that covers both release times. Returns run from the close before the call to the close of the day after it, minus SPY, so the window covers results released before the open or after the close.
- Praise and price share a cause. Analysts and the market react to the same numbers, and by the time analysts speak they have usually seen the first part of the move. Neither drives the other.
- Who gets called on matters. Companies choose which analysts ask questions, and Cohen, Lou and Malloy found some favor friendly ones.
- Survivors only. Companies no longer listed are mostly missing from the transcript set.
The full study, with the charts by year and by industry, the method and sources, is on Equibles, where the transcripts come from.
Data: earnings call transcripts and daily prices, processed by Equibles. This is not investment advice.
Top comments (0)