Evgeny Fetisov: Michael, thank you for the question. I guess I would split it in two parts. There are onetime costs, which we clearly like separate and talk about $11.3 million — $11-plus million in 2022. And we expect about $1.3 million in 2023, majority of that in Q1. So — and then we expect that will be over. Now — and then there will be the higher cost plan in the business since we are relocated to more expensive locations. We believe that Q4 is a very solid base of which you can model our business going forward. And as I mentioned, we plan on improving profitability every quarter as we go through 2023, improving margins and again, I probably won’t be repeating myself, but I guess the efficiency. We expect efficiency to be following through every quarter.
Michael Turits: And then — thanks very much for that. And then congratulations having made this transition of the operations of the business. On generative AI, I guess the simple question is, as a company in search engine optimization to the extent that use — that search broadly put shifts over to a chat functionality as opposed to a listed search functionality. How does your ability to add value and optimize people’s moves to certain websites change when that is a chat function versus a listed search function?
Eugene Levin: Thank you. And this is definitely something we’re spending a lot of time discussing internally. I think from — at this point, we don’t know what the final implementation is going to be. We have only seen a couple of examples being implemented their approach to this, which is, I think, very user-friendly. We don’t know what Google is going to do. But I think from what we’ve seen so far, the outputs of this model are actually very similar to what happens if you try to produce feature snippet based on a couple of top articles. So it’s really more of answer that takes more real estate and provides combination from several sources. And in most examples, we’ve seen so far provide expectation to those sources. So from a technical point of view, optimizing for this is very similar to optimizing for future snippet, which we help people to do for a very long time.
And actually, we’ve seen a lot of demand for those features and support of our tracking feature snippet, as well as proactive recommendations about how people can rent and be mentioned in those features snippets. So from a technical point of view, if things go the way that we are seeing them now, this is very similar to what we already have. With minor fine-tuning, we should be able to help our customers to get mentioned in those answers. But like I said, it’s a little bit hard to tell until we see the actual implementation. But we are very positive about our ability to serve our customers when they have those needs.
Michael Turits: Eugene, thanks very much. Evgeny of course good luck. .
Operator: And that does conclude the question-and-answer session and today’s conference call. We thank you for your participation. You may now disconnect.
