These things aren't mutually exclusive. If it takes 10 devs 3 months to ship a Java EE app that requires 2 full time SREs to keep running, vs. 2 devs, 3 months and a part time SRE to ship something built with ShinyNewTech, then you better believe that's delivering business value.
ShinyNewTech bites people when they aren't intentional about introducing it and haven't done an accurate cost/benefit analysis. If you've done that analysis, and it really is a big difference in cost to market & cost of operation, then you're leaving money on the table by not taking that route. This happens all the time.
The cost of rejecting new stuff because the status quo seems to be working well enough is massive, especially because the status quo at most software shops is room-on-fire-this-is-fine.jpg
A good engineer certainly should know when to bite on new technology if it will deliver legitimate value but a FAR more common story in this industry is that whole stacks get rewritten for the new hotness with very little actual tangible gain to show for it. Opportunity cost of jumping at new tech when you shouldn't is much higher than the opportunity cost of the opposite.
ShinyNewTech bites people when they aren't intentional about introducing it and haven't done an accurate cost/benefit analysis. If you've done that analysis, and it really is a big difference in cost to market & cost of operation, then you're leaving money on the table by not taking that route. This happens all the time.
The cost of rejecting new stuff because the status quo seems to be working well enough is massive, especially because the status quo at most software shops is room-on-fire-this-is-fine.jpg