Checking Whether Earnings Data Appeared Before the Official Release

# Could Earnings Figures be available already before the oficial relase?


# Could Earnings Figures be available already before the oficial relase?












I'm using Bloomberg as a market data provider, as well as a few uncommon and extremely cheap market data providers to scrape US fundamental data. For one company, I realized that one of the cheap providers already had a value for the actual EPS one day before the release date.

At first, I was very sure that this was just another data bug, and I even reported it to them. However, today, the official press release was published in the investor relations section of the company and was also filed in Bloomberg. Interestingly, the EPS value in the press release exactly matched the value that the cheap provider had already published yesterday. But I have checked that website already yesterday and there the press release certainly didn't exist, so the time stamp dated as February 6th (amc) is right.

Now, I am extremely confused as to how this is possible. I think the following possibilities exist:

- It was just an extremely good guess by the provider. However, since there was a huge estimation surprise, I don't think anyone could have predicted this so accurately. Also, the EPS value was 2.86, making it very unlikely that a data bug accidentally arrived at the correct value. (Though this would be more plausible if the EPS were, for example, close to 0.)

- The cheap market data provider has insider information. If this were the case, I wouldn't expect them to sell this data at such a ridiculously low price. Additionally, someone would probably have reported this to the authorities, as the data is more or less publicly available with a cheap subscription.

- The "Report Date" in Bloomberg and the timestamp of the press release in the investor relations section are not actually the "Earnings Release Date." Instead, the company may have decided to release key KPIs (such as EPS) one day earlier through a different distribution channel that Bloomberg does not cover. This is, of course, the most likely explanation. However, I still find it hard to believe, given that the share price only reacted significantly to the earnings release today.

Can anyone help me clear up this confusion?

EDIT: The company ticker was "APD" and the small data provider is called financialmodelingprep.com. I fortunately took a screenshot of their Json Response for that ticker as of February 5th, so one day before the release. You also can see on the "updatedFromDate" parameter in the Json, that the eps of 2.86 was filled on Wednesday already. I attached the screenshot (I hope this is fine, if not I could delete the infos again).

## Answer by lehalle (score 1)

https://quant.stackexchange.com/a/81885

This is an interesting question.

First I went to the web site of your provider, apparently it assigns the date of the 6 feb to the release:

All the sources I had access to are also focus on this date.

Since the web site of the company mention `Feb 6, 2025 at 8:00 AM EST`, could it be an error in the assignment of the date because of a timezone issue? (you `updatedFrom` field could be on the East of the release timestamp?).

Nevertheless, in this academic paper (A growing disparity in earnings disclosure mechanisms: The rise of concurrently released earnings announcements and 10-Ks, 2019), it is mentioned that:

> Firms have traditionally disclosed their annual financial performance via a two-step mechanism, in which they first disclose a ‘stand-alone’ earnings announcement (EA) with highlights of firm performance and subsequently provide a second, more comprehensive disclosure of performance via the 10-K filing. In this paper, we document a growing disparity over time in how firms disclose earnings to the market. Specifically, firms are increasingly switching from the conventional “two-step” disclosure strategy to a “single-step” disclosure strategy in which the EA is disclosed with the 10-K filing (hereafter ‘concurrent EA/10-K’). Overall, we find that the percentage of firms that release concurrent EA/10-Ks has risen from approximately 9% in 2002 to 43% by 2016

This is your case number 3, but it is very difficult to know if it was the case.

For your information (but it is not what happened in your case), there may be some providers that are late in recording the earnings.