Sometimes no action is the best action. It's certainly the easiest action, like everybody I get lots of spam email most of which are phishing attempts and simply ignoring them costs me no effort at all.
I don't open them, attempt to unsubscribe, or even mark them as spam because with my email application I know that tomorrow they will be "gone".
The Smart filter in Spark shows me my email in classified sections as People, Notifications, Newsletters, Pins and Seen. I deal with People first and Notifications second. Newsletters are of little concern and I address them when I have time. Pins are the things I want to action or not loose for later. In all cases I'm only shown the three most recent and as I read them they are moved to Seen and replaced with the next most recent.
Once you get used to this system it's easy to take care of recent priorities first and leave the rest for later and I can with a single click change to Classic view just to make sure I'm not missing something important. I can also expand any section to show all email under that classification.
The point is that the emails I suspect are spam I simply don't open and the next day they are effectively "gone" replaced with more recent emails.
Once a week I go though "all" in each section and trash the unread that I know are spam and any others I have read but do not need to keep. The rest I put into folders.
Any time I check Classic view I get an insight into what most people see when they launch their email client; a very long list of emails, usually displayed by date, personal emails mixed with spam, newsletters and notifications, some flagged, some read and some unread and I realise why I like this Smart filter so much.
Yes, it takes a little practice and double checking at first but I find it makes prioritising so much easier now and the view is so much less confronting than it used to be.
Of course you can still swap back and forward between Mail and Spark at any time but I seldom do.
I suppose I should add a disclaimer: I am not affiliated in any way with Spark or it's developers, the above is just a personal opinion based on my own use of the product.