The problem with carousels is that you are not going to display an attractive or wanted item to every visitor. The percentage conversion rates probably match:
- A person who saw the offer and was not interested
- Someone who likes offers but didn't wait for the carousel to complete its cycle
- Someone who saw the offer, was interested, but wasn't in a position to purchase or wasn't what they were looking to buy at the time
It is possible through analysis to tailor home page offers to the visitor, but you always run the risk of all three points still. Unless the offer is something that would be attractive to the majority of people, say 15% discount on everything or a section of goods, then a multi-channel e-commerce website would do just as well to have several blocks of products instead of a carousel.
Now, that said, carousels can also have their place when you navigate into a particular section. If, for example, the visitor navigates to the "printers" section of your website it's not outside the realms of possibility to presume they are in the market for a printer. If you have a particularly attractive printer offer on at that time then you could hero that in a large banner because you're not doing it at the expense of other sections.
I wouldn't say the time of the carousel has passed, its use needs to be considered carefully along with the context of the website. Situational awareness is also important.