Try and concentrate on the reason for change
Sometimes, yes, it is because a New Shiny has appeared and the person genuinely believes it's an improvement. Ask yourself if it's any good, are they trying to accommodate people who point out flaws in their vision *cough* Lennart *cough*, or are they upfront on their position and that absolutely defines the limits of how they can change (de Raadt, Stallman..)
Also ask yourself if it actually matters. I have preferences for editors (vi or X2), window managers (cwm), operating systems (BSD), and coding (C++, awk, python). What I actually need is a web browser, an editor, and a mail client - it doesn't really matter which. I've used CP/M, DOS, Windows, OS/2, and Unix as my main OS and they all got the job done. It was more important what they enabled in my life. It all involves compromise : currently I'm trying to have FreeBSD in a VM as my main desktop (home, work is mostly Windows), and whilst it works I have to admit Windows manages extremely well here.
The above is why 'emotional intelligence' (blech) is an important part of providing a solution.
Many, many other times it's because there is money involved. Those with long memories may remember when Windows remote desktop start becoming A Thing. Like any New Shiny (conveniently based on Old Shiny), it was pushed hard as the solution to everything - never mind that in the early days printing and requirements for local storage were somewhat of an issue to name only a couple of issues. Now it's devops, and VDI, and agile.
There is never going to be consensus on any new trend, in any area (not just IT), because to immediately take a mature and rational view would get in the way of making money. Far better to stick the new idea in somewhere where it just about works, patch around the issues, and after the initial gold rush actually use it for genuinely useful instances.