Stats available for ordinary mortals are decreasing, but not so badly. I think the death of the site might happen after a decade or more later.
SE gives us a way to create a lot of useful statistics from a large part of the public data available here.
Here you can see the count of the open, closed and deleted questions by month. Numbers are decreasing, but not very heavily.
Here you can see the "attractive power" of the questions generated by that month. By "attractive power" I understand the sum of the average daily clicks of the questions posted in the given month.
What I see as a larger problem, is that the size of the Internet (i.e., count of the people and their activity to find content) has more than doubled in the last ten years. In my opinion, in this relative sense, the fact of the failure is beyond doubt.
And PSE had the capability to revolutionize the Internet as we know it. At least for us, "intermediate people": we with a STEM mindset, education and profession, with a strong physics interest, but without (or with very little) formal physics education. I think we were the largest group of the visitors and even contributors of the site, yet our interests and views were considered the least. That was a mistake, and that mistake was fatal.
Between physics and any other, there is a very characteristic difference. High level physics is very hard and it can not be learnt on the Internet effectively. To be a master of a programming language, you only need focus, time and experience. To know cutting edge physics well, most of us don't have any chance.
I think PSE could have become a bridge to it.
I have learnt basic QM on my own. As I bought my first QM textbook, I did not understand anything in it. Note, I knew math well enough (operators, wave functions, eigen-things and so). The problem was that the text had a lot of things obvious for the writer and for a beginner physics student, but not for me. I needed to buy multiple textbooks, compare them, and hunt Google a lot. It was much longer and it required much more effort, as an ordinary physics student does in the frame of his/her formal education.
For example, that "x" as a function operator means the operator which converts f(x) to x*f(x), and that was for me months to understand. All my textbooks, without a single exception, simply forgot to mention this little nuance and I could not ask anyone: What is that? PSE did not exist at the time, but today, with my current skills, I would still find hard to formulate such a question in a way which would not be closed on the site now. I think, asking this is practically impossible on the current site for a beginner (both in physics and in PSE).
It could have become a pretty learning platform where we could learn a lot of physics, what we normally could not learn on the Internet (or not effectively). It did not. Opinions differ; why not? My opinion is the large amount of expelled users and unfairly killed content. Contradicting opinions typically cite the need to preserve the site quality, and the indirect deeds of the Stack Exchange company seem to side with them.
You can search for Stack Exchange critics on the Internet. I think their common point that they are mostly false in the "logical framework" of the site rules and working mechanisms, but they are painfully right in the viewpoint of people just looking for answers to the questions he/she could not ever ask from anyone.