In the post a quick look at @null, and the profitability of promoted posts by @furion, after analysing the last ten days of data (01-09-2016 to 10-09-2016), he makes the following conclusion about promoted content:
If we look at all the posts that have been promoted, we can see that in general, they have been very profitable so far.
Sadly this is a mistake. Analysing all promoted posts show that most are not profitable. Most promoted posts failed to earn more than the cost of promoting them.
Results
Only 36.78% of the posts that were promoted ended up being profitable. Out of 764 posts that were promoted only 281 were profitable; 483 were not.

This is what I would expect. I've seen a lot of posts on the promoted pages with a low reward and relatively high promotion cost.
It's a new feature and the cost of promotion hasn't found a balance. Over time a good value for post promotion will be found and more posts will become profitable.
Method
I gathered all transfers to the @null account and parsed out the permlink from the memo
field. I then gathered the post content and compared its promoted
value (the amount spent on promotion) with its total_payout_value
(the amount earned).
I did this only for the following posts:
- That have a
promoted
value greater than zero - That have had their first payout
It's possible that a post may become profitable after the first payout; however, since there's not enough promoted post data, I have to stick to the first payout. Since posts are removed from the promoted feed after the first payout, I think it's more indicative of the promoted feature. Moreover, my feeling is that posts seldom do better on their second payout.