No pressure. The big promise of a framework is usually gaining some advantage of speed on some metric. We are usually optimizing for speed on some facet of the software project!
We can evaluate our dependencies very clinically. Frameworks usually promise on speed to deliver and this usually comes with an opaque tradeoff cost of one of the other aforementioned facets of speed.
Often speed in refactoring/maintenance can slow down with a framework. Either yours or theirs. Newer frameworks change more in the early days creating thrash. But they usually stabilize with time. Likewise userland apps bolt on new functionality accruing dept with time further locking the codebase into an older and older framework. Neither of these things is inherently bad as that many frameworks nowadays value API stability. But regardless you should independently verify that by looking at the changelog, releases, issue tracker and stackoverflow. Maybe your team can take some of the framework maintenance on and contribute back?
Of course a stable API might not matter much if the tradeoff is speed in terms of raw network performance and compute execution. Obviously we can look that very clinically. How big are the payloads? Etc. I like using DaisyDisk to check out the disk cost and either Firefox or Chrome devtools to ascertain the network cost of webby stuff.
Frame explorations in terms of testable experiments attempting to falsify a hypothesis. You can evaluate a smaller dependency in a day. For something larger like a full blown web framework or language I'd budget least a week. More ideally.
It is valuable for everyone involved to do the best diligence research for your long term organization and project needs. And given that no two projects ever have the same dep tree it is clear there is no singular right answer for all scenarios.
Also know that at some point somebody has to make a call and sometimes that call will be wrong but a good team will catch that and correct it with a blameless retrospective.
“If we have data, let’s look at data. If all we have are opinions, let’s go with mine.” Jim Barksdale